`wand.image` — Image objects¶

Opens and manipulates images. Image objects can be used in with statement, and these resources will be automatically managed (even if any error happened):

with Image(filename='pikachu.png') as i:
    print('width =', i.width)
    print('height =', i.height)

wand.image.ALPHA_CHANNEL_TYPES = ('undefined', 'activate', 'associate', 'background', 'copy', 'deactivate', 'discrete', 'disassociate', 'extract', 'off', 'on', 'opaque', 'remove', 'set', 'shape', 'transparent')¶

(tuple) The list of alpha_channel types.

'undefined'
'activate'
'background'
'copy'
'deactivate'
'discrete' - Only available in ImageMagick-7
'extract'
'off' - Only available in ImageMagick-7
'on' - Only available in ImageMagick-7
'opaque'
'reset' - Only available in ImageMagick-6
'set'
'shape'
'transparent'
'flatten' - Only available in ImageMagick-6
'remove'

See also

ImageMagick Image Channel: Describes the SetImageAlphaChannel method which can be used to modify alpha channel. Also describes AlphaChannelType

wand.image.AUTO_THRESHOLD_METHODS = ('undefined', 'kapur', 'otsu', 'triangle')¶

(tuple) The list of methods used by Image.auto_threshold()

'undefined'
'kapur'
'otsu'
'triangle'

New in version 0.5.5.

class wand.image.ArtifactTree(image)¶

Splay tree to map image artifacts. Values defined here are intended to be used elseware, and will not be written to the encoded image.

For example:

# Omit timestamp from PNG file headers.
with Image(filename='input.png') as img:
    img.artifacts['png:exclude-chunks'] = 'tIME'
    img.save(filename='output.png')

Parameters:: image (Image) – an image instance

Note

You don’t have to use this by yourself. Use Image.artifacts property instead.

New in version 0.5.0.

class wand.image.BaseImage(wand)¶

The abstract base of Image (container) and SingleImage. That means the most of operations, defined in this abstract class, are possible for both Image and SingleImage.

New in version 0.3.0.

adaptive_blur(radius=0.0, sigma=0.0, channel=None)¶

Adaptively blurs the image by decreasing Gaussian as the operator approaches detected edges.

See:

Example of Adaptive Blur.

Parameters:

radius (numbers.Real) – size of gaussian aperture.
sigma (numbers.Real) – Standard deviation of the gaussian filter.
channel (str) – Apply the blur effect on a specific channel. See CHANNELS.

New in version 0.5.3.

Changed in version 0.5.5: Added optional channel argument

adaptive_resize(columns=None, rows=None)¶

Adaptively resize image by applying Mesh interpolation.

Parameters:

columns (numbers.Integral) – width of resized image.
rows (numbers.Integral) – height of resized image.

New in version 0.5.3.

adaptive_sharpen(radius=0.0, sigma=0.0, channel=None)¶

Adaptively sharpens the image by sharpening more intensely near image edges and less intensely far from edges.

See:

Example of Adaptive Sharpen.

Parameters:

radius (numbers.Real) – size of gaussian aperture.
sigma (numbers.Real) – Standard deviation of the gaussian filter.
channel (str) – Apply the sharpen effect on a specific channel. See CHANNELS.

New in version 0.5.3.

Changed in version 0.5.5: Added optional channel argument

adaptive_threshold(width, height, offset=0.0)¶

Applies threshold for each pixel based on neighboring pixel values.

Parameters:

width (numbers.Integral) – size of neighboring pixels on the X-axis.
height (numbers.Integral) – size of neighboring pixels on the Y-axis.
offset (numbers.Real) – normalized number between 0.0 and quantum_range. Forces the pixels to black if values are below offset.

New in version 0.5.3.

affine(sx=1, rx=0, ry=0, sy=1, tx=0, ty=0)¶

Transforms an image by applying an affine matrix.

                                  | sx  rx  0 |
| x', y', 1 |  =  | x, y, 1 |  *  | ry  sy  0 |
                                  | tx  ty  1 |

Where sx & sy control the scale / shearing, rx & ry control the rotational values, and the tx & ty control the translations.

Set the virtual_pixel attribute before invoking affine() method.

from wand.image import Image

with Image(filename='wizard:') as img:
    img.virtual_pixel = 'mirror'
    img.affine(sx=0.9, ry=1.1)
    img.save(filename='output.png')

Parameters:

sx (numbers.Real) – Horizontal scale/shear. Default value 1.0
rx (numbers.Real) – Horizontal rotation. Default value 0.0
ry (numbers.Real) – Vertical rotation. Default value 0.0
sy (numbers.Real) – Vertical scale/shear. Default value 1.0
tx (numbers.Real) – Horizontal translation. Default value 0.0
ty (numbers.Real) – Vertical translation. Default value 0.0

New in version 0.7.0.

property alpha_channel¶

(bool) Get state of image alpha channel. It can also be used to enable/disable alpha channel, but with different behavior such as new, copied, or existing.

Behavior of setting alpha_channel is defined with the following values:

'activate', 'on', or True will enable an images
alpha channel. Existing alpha data is preserved.
'deactivate', 'off', or False will disable an images
alpha channel. Any data on the alpha will be preserved.
'associate' & 'disassociate' toggle alpha channel flag in
certain image-file specifications.
'set' enables and resets any data in an images alpha channel.
'opaque' enables alpha/matte channel, and forces full opaque
image.
'transparent' enables alpha/matte channel, and forces full
transparent image.
'extract' copies data in alpha channel across all other channels,
and disables alpha channel.
'copy' calculates the gray-scale of RGB channels,
and applies it to alpha channel.
'shape' is identical to 'copy', but will color the resulting
image with the value defined with background_color.
'remove' will composite background_color value.
'background' replaces full-transparent color with background
color.

Note

The alpha_channel attribute will always return True if alpha channel is enabled, and False otherwise. Setting this property with a string value from ALPHA_CHANNEL_TYPES will resolve to a bool after applying channel operations listed above.

With ImageMagick-6, values 'on' & 'off' are aliased to 'activate' & 'deactivate'. However in ImageMagick-7, both 'on' & 'off' have their own behavior.

New in version 0.2.1.

Changed in version 0.4.1: Support for additional setting values. However Image.alpha_channel will continue to return bool if the current alpha/matte state is enabled.

Changed in version 0.6.0: Setting the alpha channel will apply the change to all frames in the image stack.

property animation¶

(bool) Whether the image is animation or not. It doesn’t only mean that the image has two or more images (frames), but all frames are even the same size. It’s about image format, not content. It’s False even if image/ico consists of two or more images of the same size.

For example, it’s False for image/jpeg, image/gif, image/ico.

If image/gif has two or more frames, it’s True. If image/gif has only one frame, it’s False.

New in version 0.3.0.

Changed in version 0.3.8: Became to accept image/x-gif as well.

annotate(text, drawing_wand, left=0, baseline=0, angle=0)¶

Draws text on an image. This method differs from caption() as it uses Drawing class to manage the font configuration & style context.

from wand.drawing import Drawing
from wand.image import Image

with Image(filename='input.jpg') as img:
    with Drawing() as ctx:
        ctx.font_family = 'Times New Roman, Nimbus Roman No9'
        ctx.font_size = 18
        ctx.text_decoration = 'underline'
        ctx.text_kerning = -1
        img.annotate('Hello World', ctx, left=20, baseline=50)
    img.save(filename='output.jpg')

Parameters:

text (wand.drawing.Drawing) – String to annotate on image.
drawing_wand – Font configuration & style context.
left (numbers.Real) – X-axis offset of the text baseline.
baseline (numbers.Real) – Y-axis offset of the bottom of the text.
angle (numbers.Real) – Degree rotation to draw text at.

New in version 0.5.6.

property antialias¶: (bool) If vectors & fonts will use anti-aliasing.

Changed in version 0.5.0: Previously named font_antialias.

auto_gamma()¶: Adjust the gamma level of an image.

New in version 0.5.4.

auto_level()¶: Scale the minimum and maximum values to a full quantum range.

New in version 0.5.4.

auto_orient()¶: Adjusts an image so that its orientation is suitable for viewing (i.e. top-left orientation). If available it uses MagickAutoOrientImage() (was added in ImageMagick 6.8.9+) if you have an older magick library, it will use _auto_orient() method for fallback.

New in version 0.4.1.

auto_threshold(method='kapur')¶

Automatically performs threshold method to reduce grayscale data down to a binary black & white image. Included algorithms are Kapur, Otsu, and Triangle methods.

Warning

This class method is only available with ImageMagick 7.0.8-41, or greater.

Parameters:: method (str) – Which threshold method to apply. See AUTO_THRESHOLD_METHODS. Defaults to 'kapur'.
Raises:: WandLibraryVersionError – if function is not available on system’s library.

New in version 0.5.5.

property background_color¶: (wand.color.Color) The image background color. It can also be set to change the background color.

New in version 0.1.9.

Changed in version 0.6.7: Allow property to be set before image read.

bilateral_blur(width=1.0, height=None, intensity=None, spatial=None)¶

Noise-reducing Gaussian distrbution filter. Preserves edges by replacing pixel intensity with weighted averages of neighboring pixels. Parameters width & height define the area-size around each pixel to include in the neighboring sample.

Warning

This method is only available with ImageMagick version 7.1.1 or greater.

Parameters:

width (numbers.Real) – The neighboring pixel area width.
height (numbers.Real) – The neighboring pixel area height. Default the given width value.
intensity (numbers.Real) – Influence the distance between colors values. Default 75% of the neighboring pixel area.
spatial (numbers.Real) – Influence the coordinate distance weights. Default 25% of the neighboring pixel area.

New in version 0.7.0.

black_threshold(threshold)¶

Forces all pixels above a given color as black. Leaves pixels above threshold unaltered.

Parameters:: threshold (Color) – Color to be referenced as a threshold.

New in version 0.5.3.

property blue_primary¶: (tuple) The chromatic blue primary point for the image. With ImageMagick-6 the primary value is (x, y) coordinates; however, ImageMagick-7 has (x, y, z).

New in version 0.5.2.

blue_shift(factor=1.5)¶

Mutes colors of the image by shifting blue values.

See:: Example of Blue Shift
Parameters:: factor (numbers.Real) – Amount to adjust values.

New in version 0.5.3.

blur(radius=0.0, sigma=0.0, channel=None)¶

Blurs the image. Convolve the image with a gaussian operator of the given radius and standard deviation (sigma). For reasonable results, the radius should be larger than sigma. Use a radius of 0 and blur() selects a suitable radius for you.

See:

Example of Blur.

Parameters:

radius (numbers.Real) – the radius of the, in pixels, not counting the center pixel. Default is 0.0.
sigma (numbers.Real) – the standard deviation of the, in pixels. Default value is 0.0.
channel (str) – Optional color channel to apply blur. See CHANNELS.

New in version 0.4.5.

Changed in version 0.5.5: Added optional channel argument.

Changed in version 0.5.7: Positional arguments radius & sigman have been converted to key-word arguments.

border(color, width, height, compose='copy')¶

Surrounds the image with a border.

Parameters:

bordercolor – the border color pixel wand
width (numbers.Integral) – the border width
height (numbers.Integral) – the border height
compose (str) – Use composite operator when applying frame. Only used if called with ImageMagick 7+.

New in version 0.3.0.

Changed in version 0.5.0: Added compose parameter, and ImageMagick 7 support.

property border_color¶: (wand.color.Color) The image border color. Used for special effects like polaroid().

New in version 0.5.4.

brightness_contrast(brightness=0.0, contrast=0.0, channel=None)¶

Converts brightness & contrast parameters into a slope & intercept, and applies a polynomial function.

Parameters:

brightness (numbers.Real) – between -100.0 and 100.0. Default is 0.0 for unchanged.
contrast (numbers.Real) – between -100.0 and 100.0. Default is 0.0 for unchanged.
channel – Isolate a single color channel to apply contrast. See CHANNELS.

New in version 0.5.4.

Changed in version 0.5.5: Optional channel argument added.

c_clear_exception = wand.api.library.MagickClearException¶: (ctypes.CFUNCTYPE) The ctypes function that clears an exception of the resource.

Note

It is an abstract attribute that has to be implemented in the subclass.

c_destroy_resource = wand.api.library.DestroyMagickWand¶: (ctypes.CFUNCTYPE) The ctypes function that destroys the resource.

Note

It is an abstract attribute that has to be implemented in the subclass.

c_get_exception = wand.api.library.MagickGetException¶: (ctypes.CFUNCTYPE) The ctypes function that gets an exception from the resource.

Note

It is an abstract attribute that has to be implemented in the subclass.

c_is_resource = wand.api.library.IsMagickWand¶: (ctypes.CFUNCTYPE) The ctypes predicate function that returns whether the given pointer (that contains a resource data usually) is a valid resource.

Note

It is an abstract attribute that has to be implemented in the subclass.

canny(radius=0.0, sigma=1.0, lower_percent=0.1, upper_percent=0.3)¶

Detect edges by leveraging a multi-stage Canny algorithm.

Warning

This class method is only available with ImageMagick 7.0.8-41, or greater.

Parameters:

radius (numbers.Real) – Size of gaussian filter.
sigma (numbers.Real) – Standard deviation of gaussian filter.
lower_percent (numbers.Real) – Normalized lower threshold. Values between 0.0 (0%) and 1.0 (100%). The default value is 0.1 or 10%.
upper_percent (numbers.Real) – Normalized upper threshold. Values between 0.0 (0%) and 1.0 (100%). The default value is 0.3 or 30%.

Raises:

WandLibraryVersionError – if function is not available on system’s library.

New in version 0.5.5.

caption(text, left=0, top=0, width=None, height=None, font=None, gravity=None)¶

Writes a caption text into the position.

Parameters:

text (str) – text to write
left (numbers.Integral) – x offset in pixels
top (numbers.Integral) – y offset in pixels
width (numbers.Integral) – width of caption in pixels. default is width of the image
height (numbers.Integral) – height of caption in pixels. default is height of the image
font (wand.font.Font) – font to use. default is font of the image
gravity (str) – text placement gravity. uses the current gravity setting of the image by default

New in version 0.3.0.

cdl(ccc)¶: Alias for color_decision_list().

New in version 0.5.7.

charcoal(radius, sigma)¶

Transform an image into a simulated charcoal drawing.

See:

Example of Charcoal.

Parameters:

radius (numbers.Real) – The size of the Gaussian operator.
sigma (numbers.Real) – The standard deviation of the Gaussian.

New in version 0.5.3.

chop(width=None, height=None, x=None, y=None, gravity=None)¶

Removes a region of an image, and reduces the image size accordingly.

Parameters:

width (numbers.Integral) – Size of region.
height (numbers.Integral) – Size of region.
x (numbers.Integral) – Offset on the X-axis.
y (numbers.Integral) – Offset on the Y-axis.
gravity (str) – Helper to auto-calculate offset. See GRAVITY_TYPES.

New in version 0.5.5.

Changed in version 0.6.8: Added gravity argument.

Changed in version 0.6.12: Allow zero values for width & height arguments.

Changed in version 0.7.0: Allow x & y offset to apply relative to gravity.

clahe(width, height, number_bins, clip_limit)¶

Contrast limited adaptive histogram equalization.

Warning

The CLAHE method is only available with ImageMagick-7.

Parameters:

width (numbers.Integral) – Tile division width.
height (numbers.Integral) – Tile division height.
number_bins (numbers.Real) – Histogram bins.
clip_limit (numbers.Real) – contrast limit.

Raises:

WandLibraryVersionError – If system’s version of ImageMagick does not support this method.

New in version 0.5.5.

clamp(channel=None)¶

Restrict color values between 0 and quantum range. This is useful when applying arithmetic operations that could result in color values over/under-flowing.

Parameters:: channel (str) – Optional color channel.

New in version 0.5.0.

Changed in version 0.5.5: Added channel argument.

clone()¶

Clones the image. It is equivalent to call Image with image parameter.

with img.clone() as cloned:
    # manipulate the cloned image
    pass

Returns:: the cloned new image
Return type:: Image

New in version 0.1.1.

clut(image, method='undefined', channel=None)¶

Replace color values by referencing another image as a Color Look Up Table.

Parameters:

image (wand.image.BaseImage) – Color Look Up Table image.
method (str) – Pixel Interpolate method. Only available with ImageMagick-7. See PIXEL_INTERPOLATE_METHODS
channel (str) – Optional color channel to target. See CHANNELS

New in version 0.5.0.

Changed in version 0.5.5: Added optional channel argument.

coalesce()¶: Rebuilds image sequence with each frame size the same as first frame, and composites each frame atop of previous.

Note

Only affects GIF, and other formats with multiple pages/layers.

New in version 0.5.0.

color_decision_list(ccc)¶

Applies color correction from a Color Correction Collection (CCC) xml string. An example of xml:

<ColorCorrectionCollection xmlns="urn:ASC:CDL:v1.2">
    <ColorCorrection id="cc03345">
        <SOPNode>
            <Slope> 0.9 1.2 0.5 </Slope>
            <Offset> 0.4 -0.5 0.6 </Offset>
            <Power> 1.0 0.8 1.5 </Power>
        </SOPNode>
        <SATNode>
            <Saturation> 0.85 </Saturation>
        </SATNode>
    </ColorCorrection>
</ColorCorrectionCollection>

Parameters:: ccc (str) – A XML string of the CCC contents.

New in version 0.5.7.

color_map(index, color=None)¶

Get & Set a color at a palette index. If color is given, the color at the index location will be set & returned. Omitting the color argument will only return the color value at index.

Valid indexes are between 0 and total colors of the image.

Note

Ensure the image type is set to 'palette' before calling the color_map() method. For example:

with Image(filename='graph.png') as img:
    img.type = 'palette'
    palette = [img.color_map(idx) for idx in range(img.colors)]
    # ...

Parameters:

index (numbers.Integral) – The color position of the image palette.
color (wand.color.Color) – Optional color to _set_ at the given index.

Returns:

Color at index.

Return type:

wand.color.Color

New in version 0.5.3.

color_matrix(matrix)¶

Adjust color values by applying a matrix transform per pixel.

Matrix should be given as 2D list, with a max size of 6x6.

An example of 3x3 matrix:

matrix = [
    [1.0, 0.0, 0.0],
    [0.0, 1.0, 0.0],
    [0.0, 0.0, 1.0],
]

Which would translate RGB color channels by calculating the following:

\[\begin{split}\begin{aligned} red' &= 1.0 * red + 0.0 * green + 0.0 * blue\\ green' &= 0.0 * red + 1.0 * green + 0.0 * blue\\ blue' &= 0.0 * red + 0.0 * green + 1.0 * blue\\ \end{aligned}\end{split}\]

For RGB colorspace images, the rows & columns are laid out as:

	Red	Green	Blue
Red’	1	0	0
Green’	0	1	0
Blue’	0	0	1
n/a	0	0	0
Alpha’	0	0	0
Offset’	0	0	0

Or for a CMYK colorspace image:

	Cyan	Yellow	Magenta	Black
Cyan’	1	0	0	0
Yellow’	0	1	0	0
Magenta’	0	0	1	0
Black’	0	0	0	1
Alpha’	0	0	0	0
Offset’	0	0	0	0

See color-matrix for examples.

See:: Example of Color Matrix.
Parameters:: matrix (collections.abc.Sequence) – 2D List of doubles.

New in version 0.5.3.

color_threshold(start=None, stop=None)¶

Forces all pixels in color range to white, and all other pixels to black.

Note

This method is only works with ImageMagick-7.0.10, or later.

Parameters:

start (wand.color.Color) – Color to begin color range.
stop (wand.color.Color) – Color to end color range.

New in version 0.6.4.

colorize(color=None, alpha=None)¶

Blends a given fill color over the image. The amount of blend is determined by the color channels given by the alpha argument.

See:

Example of Colorize.

Parameters:

color (wand.color.Color) – Color to paint image with.
alpha (wand.color.Color) – Defines how to blend color.

New in version 0.5.3.

property colors¶: (numbers.Integral) Count of unique colors used within the image. This is READ ONLY property.

New in version 0.5.3.

property colorspace¶

(str) The image colorspace.

Defines image colorspace as in COLORSPACE_TYPES enumeration.

It may raise ValueError when the colorspace is unknown.

New in version 0.3.4.

combine(channel='rgb_channels', colorspace='rgb')¶

Creates an image where each color channel is assigned by a grayscale image in a sequence.

Warning

If your using ImageMagick-6, use channel argument to control the color-channel order. With ImageMagick-7, the channel argument has been replaced with colorspace.

For example:

for wand.image import Image

with Image() as img:
    img.read(filename='red_channel.png')
    img.read(filename='green_channel.png')
    img.read(filename='blue_channel.png')
    img.combine(colorspace='rgb')
    img.save(filename='output.png')

Parameters:

channel (str) – Determines the colorchannel ordering of the sequence. Only used for ImageMagick-6. See CHANNELS.
colorspace (str) – Determines the colorchannel ordering of the sequence. Only used for ImageMagick-7. See COLORSPACE_TYPES.

New in version 0.5.9.

compare(image, metric='undefined', highlight=None, lowlight=None)¶

Compares an image with another, and returns a reconstructed image & computed distortion. The reconstructed image will show the differences colored with highlight, and similarities with lowlight.

If you need the computed distortion between to images without a image being reconstructed, use get_image_distortion() method.

Set fuzz property to adjust pixel-compare thresholds.

For example:

from wand.image import Image

with Image(filename='input.jpg') as base:
    with Image(filename='subject.jpg') as img:
        base.fuzz = base.quantum_range * 0.20  # Threshold of 20%
        result_image, result_metric = base.compare(img)
        with result_image:
            result_image.save(filename='diff.jpg')

Parameters:

image (wand.image.Image) – The reference image
metric (str) – The metric type to use for comparing. See COMPARE_METRICS
highlight (Color or str) – Set the color of the delta pixels in the resulting difference image.
lowlight (Color or str) – Set the color of the similar pixels in the resulting difference image.

Returns:

The difference image(wand.image.Image), the computed distortion between the images (numbers.Real)

Return type:

tuple ( Image, numbers.Real )

New in version 0.4.3.

Changed in version 0.5.3: Added support for highlight & lowlight.

complex(operator='undefined', snr=None)¶

Performs complex mathematics against two images in a sequence, and generates a new image with two results.

from wand.image import Image

with Image(filename='real_part.png') as imgA:
    with Image(filename='imaginary_part.png') as imgB:
        imgA.sequence.append(imgB)
    with imgA.complex('conjugate') as results:
        results.save(filename='output-%02d.png')

Warning

This class method is only available with ImageMagick 7.0.8-41, or greater.

Parameters:

operator (str) – Define which mathematic operator to perform. See COMPLEX_OPERATORS.
snr (str) – Optional SNR parameter for 'divide' operator.

Raises:

WandLibraryVersionError – If ImageMagick library does not support this function.

New in version 0.5.5.

property compose¶: (str) The type of image compose. It’s a string from COMPOSITE_OPERATORS list. It also can be set.

New in version 0.5.1.

composite(image, left=None, top=None, operator='over', arguments=None, gravity=None)¶

Places the supplied image over the current image, with the top left corner of image at coordinates left, top of the current image. The dimensions of the current image are not changed.

Parameters:

image (wand.image.Image) – the image placed over the current image
left (numbers.Integral) – the x-coordinate where image will be placed
top (numbers.Integral) – the y-coordinate where image will be placed
operator (str) – the operator that affects how the composite is applied to the image. available values can be found in the COMPOSITE_OPERATORS list. Default is 'over'.
arguments (str) – Additional numbers given as a geometry string, or comma delimited values. This is needed for 'blend', 'displace', 'dissolve', and 'modulate' operators.
gravity – Calculate the top & left values based on gravity value from GRAVITY_TYPES.

Type:

gravity: str

New in version 0.2.0.

Changed in version 0.5.3: The operator can be set, as well as additional composite arguments.

Changed in version 0.5.3: Optional gravity argument was added.

Changed in version 0.7.0: Allow x & y offset to apply relative to gravity.

composite_channel(channel, image, operator, left=None, top=None, arguments=None, gravity=None)¶

Composite two images using the particular channel.

Parameters:

channel – the channel type. available values can be found in the CHANNELS mapping
image (Image) – the composited source image. (the receiver image becomes the destination)
operator (str) – the operator that affects how the composite is applied to the image. available values can be found in the COMPOSITE_OPERATORS list
left (numbers.Integral) – the column offset of the composited source image
top (numbers.Integral) – the row offset of the composited source image
arguments (str) – Additional numbers given as a geometry string, or comma delimited values. This is needed for 'blend', 'displace', 'dissolve', and 'modulate' operators.
gravity – Calculate the top & left values based on gravity value from GRAVITY_TYPES.

Type:

gravity: str

Raises:

ValueError – when the given channel or operator is invalid

New in version 0.3.0.

Changed in version 0.5.3: Support for optional composite arguments has been added.

Changed in version 0.5.3: Optional gravity argument was added.

Changed in version 0.7.0: Allow x & y offset to apply relative to gravity.

property compression¶: (str) The type of image compression. It’s a string from COMPRESSION_TYPES list. It also can be set.

New in version 0.3.6.

Changed in version 0.5.2: Setting compression now sets both image_info and images in the internal image stack.

property compression_quality¶: (numbers.Integral) Compression quality of this image.

New in version 0.2.0.

Changed in version 0.5.2: Setting compression_quality now sets both image_info and images in the internal image stack.

concat(stacked=False)¶

Concatenates images in stack into a single image. Left-to-right by default, top-to-bottom if stacked is True.

Parameters:: stacked (bool) – stack images in a column, or in a row (default)

New in version 0.5.0.

connected_components(**kwargs)¶

Evaluates binary image, and groups connected pixels into objects. This method will also return a list of ConnectedComponentObject instances that will describe an object’s features.

from wand.image import Image

with Image(filename='objects.gif') as img:
    objects = img.connected_components()
for cc_obj in objects:
    print("{0._id}: {0.size} {0.offset}".format(cc_obj))

#=> 0: (256, 171) (0, 0)
#=> 2: (120, 135) (104, 18)
#=> 3: (50, 36) (129, 44)
#=> 4: (21, 23) (0, 45)
#=> 1: (4, 10) (252, 0)

Warning

This class method is only available with ImageMagick 7.0.8-41, or greater.

Tip

Set fuzz property to increase pixel matching by reducing tolerance of color-value comparisons:

from wand.image import Image
from wand.version import QUANTUM_RANGE

with Image(filename='objects.gif') as img:
    img.fuzz = 0.1 * QUANTUM_RANGE  # 10%
    objects = img.connected_components()

Parameters:

angle_threshold (str) – Optional argument that merges any region with an equivalent ellipse smaller than a given value. Requires ImageMagick-7.0.9-24, or greater.
area_threshold (str) – Optional argument to merge objects under an area size.
background_id (str) – Optional argument to identify which object should be the background. Requires ImageMagick-7.0.9-24, or greater.
circularity_threshold (str) – Optional argument that merges any region smaller than value defined as: 4*pi*area/perimeter^2. Requires ImageMagick-7.0.9-24, or greater.
connectivity (numbers.Integral) – Either 4, or 8. A value of 4 will evaluate each pixels top-bottom, & left-right neighbors. A value of 8 will use the same pixels as with 4, but will also include the four corners of each pixel. Default value of 4.
diameter_threshold (str) – Optional argument to merge any region under a given value. A region is defined as: sqr(4*area/pi). Requires ImageMagick-7.0.9-24.
eccentricity_threshold – Optional argument to merge any region with ellipse eccentricity under a given value. Requires ImageMagick-7.0.9-24, or greater.
keep (str) – Comma separated list of object IDs to isolate, the reset are converted to transparent.
keep_colors (str) – Semicolon separated list of objects to keep by their color value. Requires ImageMagick-7.0.9-24, or greater.
keep_top (str) – Keeps only the top number of objects by area. Requires ImageMagick-7.0.9-24, or greater.
major_axis_threshold (str) – Optional argument to merge any ellipse with a major axis smaller then given value. Requires ImageMagick-7.0.9-24, or greater.
mean_color (bool) – Optional argument. Replace object color with mean color of the source image.
minor_axis_threshold (str) – Optional argument to merge any ellipse with a minor axis smaller then given value. Requires ImageMagick-7.0.9-24, or greater.
perimeter_threshold – Optional argument to merge any region with a perimeter smaller than the given value. Requires ImageMagick-7.0.9-24, or greater.
remove (str) – Comma separated list of object IDs to ignore, and convert to transparent.
remove_colors (str) – Semicolon separated list of objects to remove by there color. Requires ImageMagick-7.0.9-24, or greater.

Returns:

A list of ConnectedComponentObject.

Return type:

list [ConnectedComponentObject]

Raises:

WandLibraryVersionError – If ImageMagick library does not support this method.

New in version 0.5.5.

Changed in version 0.5.6: Added mean_color, keep, & remove optional arguments.

Changed in version 0.6.4: Added angle_threshold, circularity_threshold, diameter_threshold, eccentricity_threshold, keep_colors, major_axis_threshold, minor_axis_threshold, perimeter_threshold, and remove_colors optional arguments.

contrast(sharpen=True)¶

Enhances the difference between lighter & darker values of the image. Set sharpen to False to reduce contrast.

Parameters:: sharpen (bool) – Increase, or decrease, contrast. Default is True for increased contrast.

New in version 0.5.7.

contrast_stretch(black_point=0.0, white_point=None, channel=None)¶

Enhance contrast of image by adjusting the span of the available colors.

Parameters:

black_point (numbers.Real) – black point between 0.0 and 1.0. default is 0.0
white_point (numbers.Real) – white point between 0.0 and 1.0. Defaults to the same value given to the black_point argument.
channel (CHANNELS) – optional color channel to apply contrast stretch

Raises:

ValueError – if channel is not in CHANNELS

New in version 0.4.1.

Changed in version 0.5.5: The white_point argument will now default to the value given by the black_point argument.

convex_hull(background=None)¶

Find the smallest convex polygon, and returns a list of points.

Note

Requires ImageMagick-7.0.10 or greater.

You can pass the list of points directly to Drawing.polygon() method to draw the convex hull shape on the image.

from wand.image import Image
from wand.drawing import Drawing

with Image(filename='kdf_black.png') as img:
  points = img.convex_hull()
  with Drawing() as ctx:
    ctx.fill_color = 'transparent'
    ctx.stroke_color = 'red'
    ctx.polygon(points=points)
    ctx(img)
img.save(filename='kdf_black_convex_hull.png')

Parameters:: background (str or Color) – Define which color value to evaluate as the background.
Returns:: list of points
Return type:: list [ tuple ( float, float ) ]

New in version 0.6.4.

crop(left=0, top=0, right=None, bottom=None, width=None, height=None, reset_coords=True, gravity=None)¶

Crops the image in-place.

+--------------------------------------------------+
|              ^                         ^         |
|              |                         |         |
|             top                        |         |
|              |                         |         |
|              v                         |         |
| <-- left --> +-------------------+  bottom       |
|              |             ^     |     |         |
|              | <-- width --|---> |     |         |
|              |           height  |     |         |
|              |             |     |     |         |
|              |             v     |     |         |
|              +-------------------+     v         |
| <--------------- right ---------->               |
+--------------------------------------------------+

Parameters:

left (numbers.Integral) – x-offset of the cropped image. default is 0
top (numbers.Integral) – y-offset of the cropped image. default is 0
right (numbers.Integral) – second x-offset of the cropped image. default is the width of the image. this parameter and width parameter are exclusive each other
bottom (numbers.Integral) – second y-offset of the cropped image. default is the height of the image. this parameter and height parameter are exclusive each other
width (numbers.Integral) – the width of the cropped image. default is the width of the image. this parameter and right parameter are exclusive each other
height (numbers.Integral) – the height of the cropped image. default is the height of the image. this parameter and bottom parameter are exclusive each other
reset_coords (bool) – optional flag. If set, after the rotation, the coordinate frame will be relocated to the upper-left corner of the new image. By default is True.
gravity (GRAVITY_TYPES) – optional flag. If set, will calculate the top and left attributes. This requires both width and height parameters to be included.

Raises:

ValueError – when one or more arguments are invalid

Note

If you want to crop the image but not in-place, use slicing operator.

New in version 0.1.7.

Changed in version 0.1.8: Made to raise ValueError instead of IndexError for invalid width/height arguments.

Changed in version 0.4.1: Added gravity option. Using gravity along with width & height to auto-adjust left & top attributes.

Changed in version 0.7.0: Allow x & y offset to apply relative to gravity.

cycle_color_map(offset=1)¶

Shift the image color-map by a given offset.

Parameters:: offset (numbers.Integral) – number of steps to rotate index by.

New in version 0.5.3.

decipher(passphrase)¶

Decrypt ciphered pixels into original values.

Note

ImageError will be thrown if the system’s ImageMagick library was compiled without cipher support.

Parameters:: passphrase (str) – the secret passphrase to decrypt with.

New in version 0.6.3.

deconstruct()¶: Iterates over internal image stack, and adjust each frame size to minimum bounding region of any changes from the previous frame.

New in version 0.5.0.

property delay¶: (numbers.Integral) The number of ticks between frames.

New in version 0.5.9.

property depth¶: (numbers.Integral) The depth of this image.

New in version 0.2.1.

deskew(threshold)¶

Attempts to remove skew artifacts common with most scanning & optical import devices.

Params threshold:: limit between foreground & background. Use a real number between 0.0 & 1.0 to match CLI’s percent argument.

New in version 0.5.0.

despeckle()¶

Applies filter to reduce noise in image.

See:: Example of Despeckle.

New in version 0.5.0.

dirty = None¶: (bool) Whether the image is changed or not.

property dispose¶: (str) Controls how the image data is handled during animations. Values are from DISPOSE_TYPES list, and can also be set.

See also

Dispose Images section in Animation Basics article.

New in version 0.5.0.

distort(method, arguments, best_fit=False, filter=None)¶

Distorts an image using various distorting methods.

from wand.image import Image
from wand.color import Color

with Image(filename='checks.png') as img:
    img.virtual_pixel = 'background'
    img.background_color = Color('green')
    img.matte_color = Color('skyblue')
    arguments = (0, 0, 20, 60,
                 90, 0, 70, 63,
                 0, 90, 5, 83,
                 90, 90, 85, 88)
    img.distort('perspective', arguments)
    img.save(filename='checks_perspective.png')

Use virtual_pixel, background_color, and matte_color properties to control the behavior of pixels rendered outside of the image boundaries.

Use interpolate_method to control how images scale-up.

Distortion viewport, and scale, can be defined by using Image.artifacts dictionary. For example:

img.artifacts['distort:viewport'] = '44x44+15+0'
img.artifacts['distort:scale'] = '10'

See:

Additional examples of Distortion.

Parameters:

method (str) – Distortion method name from DISTORTION_METHODS
arguments (collections.abc.Sequence) – List of distorting float arguments unique to distortion method
best_fit (bool) – Attempt to resize resulting image fit distortion. Defaults False
filter (str) – Optional resampling filter used when calculating pixel-value. Defaults to 'mitchell', or 'lanczos' based on image type & operation.

New in version 0.4.1.

Changed in version 0.6.11: Included filter= parameter.

edge(radius=0.0)¶

Applies convolution filter to detect edges.

See:: Example of Edge.
Parameters:: radius (numbers.Real) – aperture of detection filter.

New in version 0.5.0.

emboss(radius=0.0, sigma=0.0)¶

Applies convolution filter against Gaussians filter.

Note

The radius value should be larger than sigma for best results.

See:

Example of Emboss.

Parameters:

radius (numbers.Real) – filter aperture size.
sigma (numbers.Real) – standard deviation.

New in version 0.5.0.

encipher(passphrase)¶

Encrypt plain pixels into ciphered values.

Note

ImageError will be thrown if the system’s ImageMagick library was compiled without cipher support.

Parameters:: passphrase (str) – the secret passphrase to encrypt with.

New in version 0.6.3.

Changed in version 0.6.8: Fixed C-API call.

enhance()¶

Applies digital filter to reduce noise.

See:: Example of Enhance.

New in version 0.5.0.

equalize(channel=None)¶

Equalizes the image histogram

Parameters:: channel (str) – Optional channel. See CHANNELS.

New in version 0.3.10.

Changed in version 0.5.5: Added optional channel argument.

evaluate(operator=None, value=0.0, channel=None)¶

Apply arithmetic, relational, or logical expression to an image.

Percent values must be calculated against the quantum range of the image:

fifty_percent = img.quantum_range * 0.5
img.evaluate(operator='set', value=fifty_percent)

See:

Example of Evaluate Expression.

Parameters:

operator (EVALUATE_OPS) – Type of operation to calculate
value (numbers.Real) – Number to calculate with operator
channel (CHANNELS) – Optional channel to apply operation on.

Raises:

TypeError – When value is not numeric.
ValueError – When operator, or channel are not defined in constants.

New in version 0.4.1.

evaluate_images(operator=None)¶

Create a new image by applying arithmetic operation between all images loaded in the image-stack.

For example:

with Image() as img:
     img.read(filename='first.png')
     img.read(filename='second.png')
     with img.evaluate_images('add') as sum:
         sum.save(filename='sum.png')

Parameters:: operator (EVALUATE_OPS) – Type of operation to calculate.

New in version 0.7.0.

export_pixels(x=0, y=0, width=None, height=None, channel_map='RGBA', storage='char')¶

Export pixel data from a raster image to a list of values.

The channel_map tells ImageMagick which color channels to export, and what order they should be written as – per pixel. Valid entries for channel_map are:

'R' - Red channel
'G' - Green channel
'B' - Blue channel
'A' - Alpha channel (0 is transparent)
'O' - Alpha channel (0 is opaque)
'C' - Cyan channel
'Y' - Yellow channel
'M' - Magenta channel
'K' - Black channel
'I' - Intensity channel (only for grayscale)
'P' - Padding

See STORAGE_TYPES for a list of valid storage options. This tells ImageMagick what type of data it should calculate & write to. For example; a storage type of 'char' will write a 8-bit value between 0 ~ 255, a storage type of 'short' will write a 16-bit value between 0 ~ 65535, and a 'integer' will write a 32-bit value between 0 ~ 4294967295.

Note

By default, the entire image will be exported as 'char' storage with each pixel mapping Red, Green, Blue, & Alpha channels.

Parameters:

x (numbers.Integral) – horizontal starting coordinate of raster.
y (numbers.Integral) – vertical starting coordinate of raster.
width (numbers.Integral) – horizontal length of raster.
height (numbers.Integral) – vertical length of raster.
channel_map (str) – a string listing the channel data format for each pixel.
storage (str) – what data type each value should be calculated as.

Returns:

list of values.

Return type:

collections.abc.Sequence

New in version 0.5.0.

Changed in version 0.6.11: Update storage type size for “long” & “quantum” values.

extent(width=None, height=None, x=None, y=None, gravity=None)¶

Adjust the canvas size of the image. Use x & y to offset the image’s relative placement in the canvas, or gravity helper for quick position placement.

Parameters:

width (numbers.Integral) – the target width of the extended image. Default is the width of the image.
height (numbers.Integral) – the target height of the extended image. Default is the height of the image.
x (numbers.Integral) – the x-axis offset of the extended image. Default is 0, and can not be used with gravity.
y (numbers.Integral) – the y offset of the extended image. Default is 0, and can not be used with gravity.
gravity (str) – position of the item extent when not using x & y. See GRAVITY_TYPES.

New in version 0.4.5.

Changed in version 0.6.8: Added gravity argument.

Changed in version 0.7.0: Allow x & y offset to apply relative to gravity.

features(distance)¶

Calculate directional image features for each color channel. Feature metrics including:

angular second moment
contrast
correlation
variance sum of squares
inverse difference moment
sum average
sum variance
sum entropy
entropy
difference variance
difference entropy
information measures of correlation 1
information measures of correlation 2
maximum correlation coefficient

With each metric containing horizontal, vertical, left & right diagonal values.

from wand.image import Image

with Image(filename='rose:') as img:
    channel_features = img.features(distance=32)
    for channels, features in channel_features.items():
        print(channels)
        for feature, directions in features.items():
            print('  ', feature)
            for name, value in directions.items():
                print('    ', name, value)

Parameters:: distance (numbers.Integral) – Define the distance if pixels to calculate.
Returns:: a dict mapping each color channel with a dict of each feature.
Return type:: dict

New in version 0.5.5.

fft(magnitude=True)¶: Alias for forward_fourier_transform().

New in version 0.5.7.

flip()¶

Creates a vertical mirror image by reflecting the pixels around the central x-axis. It manipulates the image in place.

See:: Example of Flip and flop.

New in version 0.3.0.

floodfill(fill='none', fuzz=0.1, bordercolor=None, x=0, y=0, invert=False, channel='default_channels')¶

Changes pixel value to fill color at (x, y) coordinate, and repeats for all neighboring pixels until border is reached. Use fuzz and/or bordercolor to defined tolarance & border values.

Parameters:

fill (wand.color.Color or str) – New target color value.
fuzz (numbers.Real) – Fill threashold. Values between 0.0 and 1.0. Default 0.1 (10%).
bordercolor (wand.color.Color or str) – Optional color to use as a border. Default None.
x (number.Integral) – Starting point X coordinate. Default 0.
y (number.Integral) – Starting point Y coordinate. Default 0.
invert (bool) – Only paint none-matching pixels. Default False.
channel (CHANNELS) – Optional channel mask. Default 'default_channels'.

New in version 0.7.0.

flop()¶

Creates a horizontal mirror image by reflecting the pixels around the central y-axis. It manipulates the image in place.

See:: Example of Flip and flop.

New in version 0.3.0.

property font¶: (wand.font.Font) The current font options.

property font_antialias¶: Deprecated since version 0.5.0: Use antialias instead.

property font_path¶: (str) The path of the current font. It also can be set.

property font_size¶: (numbers.Real) The font size. It also can be set.

property format¶

(str) The image format.

If you want to convert the image format, just reset this property:

assert isinstance(img, wand.image.Image)
img.format = 'png'

It may raise ValueError when the format is unsupported.

See also

ImageMagick Image Formats: ImageMagick uses an ASCII string known as magick (e.g. GIF) to identify file formats, algorithms acting as formats, built-in patterns, and embedded profile types.

New in version 0.1.6.

forward_fourier_transform(magnitude=True)¶

Performs a discrete Fourier transform. The image stack is replaced with the results. Either a pair of magnitude & phase images, or real & imaginary (HDRI).

from wand.image import Image
from wand.version import QUANTUM_RANGE

with Image(filename='source.png') as img:
    img.forward_fourier_transform()
    img.depth = QUANTUM_RANGE
    img.save(filename='fft_%02d.png')

Note

ImageMagick must have HDRI support to compute real & imaginary components (i.e. magnitude=False).

Parameters:: magnitude (bool) – If True, generate magnitude & phase, else real & imaginary. Default True

New in version 0.5.5.

frame(matte=None, width=1, height=1, inner_bevel=0, outer_bevel=0, compose='over')¶

Creates a bordered frame around image. Inner & outer bevel can simulate a 3D effect.

Parameters:

matte (wand.color.Color) – color of the frame
width (numbers.Integral) – total size of frame on x-axis
height (numbers.Integral) – total size of frame on y-axis
inner_bevel (numbers.Real) – inset shadow length
outer_bevel (numbers.Real) – outset highlight length
compose (str) – Optional composite operator. Default 'over', and only available with ImageMagick-7.

New in version 0.4.1.

Changed in version 0.5.6: Added optional compose parameter.

function(function, arguments, channel=None)¶

Apply an arithmetic, relational, or logical expression to an image.

Defaults entire image, but can isolate affects to single color channel by passing CHANNELS value to channel parameter.

Note

Support for function methods added in the following versions of ImageMagick.

'polynomial' >= 6.4.8-8
'sinusoid' >= 6.4.8-8
'arcsin' >= 6.5.3-1
'arctan' >= 6.5.3-1

See:

Example of Function Expression.

Parameters:

function (str) – a string listed in FUNCTION_TYPES
arguments (collections.abc.Sequence) – a sequence of doubles to apply against function
channel (str) – optional CHANNELS, defaults all

Raises:

ValueError – when a function, or channel is not defined in there respected constant
TypeError – if arguments is not a sequence

New in version 0.4.1.

property fuzz¶: (numbers.Real) The normalized real number between 0.0 and quantum_range. This property influences the accuracy of compare().

New in version 0.5.3.

fx(expression, channel=None)¶

Manipulate each pixel of an image by given expression.

FX will preserver current wand instance, and return a new instance of Image containing affected pixels.

Defaults entire image, but can isolate affects to single color channel by passing CHANNELS value to channel parameter.

See also

The anatomy of FX expressions can be found at http://www.imagemagick.org/script/fx.php

See:

Example of FX.

Parameters:

expression (str) – The entire FX expression to apply
channel (CHANNELS) – Optional channel to target.

Returns:

A new instance of an image with expression applied

Return type:

Image

New in version 0.4.1.

Changed in version 0.6.9: Will raise WandRuntimeError if method is unable to generate a new image & doesn’t throw an exception.

gamma(adjustment_value=1.0, channel=None)¶

Gamma correct image.

Specific color channels can be correct individual. Typical values range between 0.8 and 2.3.

See:

Example of Gamma.

Parameters:

adjustment_value (numbers.Real) – value to adjust gamma level. Default 1.0
channel (str) – optional channel to apply gamma correction

Raises:

TypeError – if gamma_point is not a numbers.Real
ValueError – if channel is not in CHANNELS

New in version 0.4.1.

gaussian_blur(radius=0.0, sigma=0.0, channel=None)¶

Blurs the image. We convolve the image with a gaussian operator of the given radius and standard deviation (sigma). For reasonable results, the radius should be larger than sigma. Use a radius of 0 and blur() selects a suitable radius for you.

See:

Example of Gaussian Blur.

Parameters:

radius (numbers.Real) – the radius of the, in pixels, not counting the center pixel
sigma (numbers.Real) – the standard deviation of the, in pixels
channel (str) – Optional color channel to target. See CHANNELS

New in version 0.3.3.

Changed in version 0.5.5: Added channel argument.

Changed in version 0.5.7: Positional arguments radius & sigma have been converted to keyword arguments.

get_image_distortion(image, metric='undefined')¶

Compares two images, and return the specified distortion metric.

This method is faster than compare() method as ImageMagick will not need to reconstruct an image.

Parameters:

image (wand.image.BaseImage) – Image to reference.
metric (str) – Compare disortion metric to use. See COMPARE_METRICS.

Returns:

Computed value of the distortion metric used.

Return type:

numbers.Real

New in version 0.6.6.

property gravity¶: (str) The text placement gravity used when annotating with text. It’s a string from GRAVITY_TYPES list. It also can be set.

property green_primary¶: (tuple) The chromatic green primary point for the image. With ImageMagick-6 the primary value is (x, y) coordinates; however, ImageMagick-7 has (x, y, z).

New in version 0.5.2.

hald_clut(image, channel=None)¶

Replace color values by referencing a Higher And Lower Dimension (HALD) Color Look Up Table (CLUT). You can generate a HALD image by using ImageMagick’s hald: protocol.

with Image(filename='rose:') as img:
    with Image(filename='hald:3') as hald:
        hald.gamma(1.367)
        img.hald_clut(hald)

Parameters:

image (wand.image.BaseImage) – The HALD color matrix.
channel (str) – Optional color channel to target. See CHANNELS

New in version 0.5.0.

Changed in version 0.5.5: Added channel argument.

property height¶: (numbers.Integral) The height of this image.

property histogram¶

(HistogramDict) The mapping that represents the histogram. Keys are Color objects, and values are the number of pixels.

Tip

True-color photos can have millions of color values. If performance is more valuable than accuracy, remember to quantize() the image before generating a HistogramDict.

with Image(filename=’hd_photo.jpg’) as img:
img.quantize(255, ‘RGB’, 0, False, False) hist = img.histogram

New in version 0.3.0.

hough_lines(width, height=None, threshold=40)¶

Identify lines within an image. Use canny() to reduce image to a binary edge before calling this method.

Warning

This class method is only available with ImageMagick 7.0.8-41, or greater.

Parameters:

width (numbers.Integral) – Local maxima of neighboring pixels.
height (numbers.Integral) – Local maxima of neighboring pixels.
threshold (numbers.Integral) – Line count to limit. Default to 40.

Raises:

WandLibraryVersionError – If system’s version of ImageMagick does not support this method.

New in version 0.5.5.

ift(phase, magnitude=True)¶: Alias for inverse_fourier_transform().

New in version 0.5.7.

implode(amount=0.0, method='undefined')¶

Creates a “imploding” effect by pulling pixels towards the center of the image.

See:

Example of Implode.

Parameters:

amount (numbers.Real) – Normalized degree of effect between 0.0 & 1.0.
method (str) – Which interpolate method to apply to effected pixels. See PIXEL_INTERPOLATE_METHODS for a list of options. Only available with ImageMagick-7.

New in version 0.5.2.

import_pixels(x=0, y=0, width=None, height=None, channel_map='RGB', storage='char', data=None)¶

Import pixel data from a byte-string to the image. The instance of Image must already be allocated with the correct size.

The channel_map tells ImageMagick which color channels to export, and what order they should be written as – per pixel. Valid entries for channel_map are:

'R' - Red channel
'G' - Green channel
'B' - Blue channel
'A' - Alpha channel (0 is transparent)
'O' - Alpha channel (0 is opaque)
'C' - Cyan channel
'Y' - Yellow channel
'M' - Magenta channel
'K' - Black channel
'I' - Intensity channel (only for grayscale)
'P' - Padding

See STORAGE_TYPES for a list of valid storage options. This tells ImageMagick what type of data it should calculate & write to. For example; a storage type of 'char' will write a 8-bit value between 0 ~ 255, a storage type of 'short' will write a 16-bit value between 0 ~ 65535, and a 'integer' will write a 32-bit value between 0 ~ 4294967295.

Note

By default, the entire image will be exported as 'char' storage with each pixel mapping Red, Green, Blue, & Alpha channels.

Parameters:

x (numbers.Integral) – horizontal starting coordinate of raster.
y (numbers.Integral) – vertical starting coordinate of raster.
width (numbers.Integral) – horizontal length of raster.
height (numbers.Integral) – vertical length of raster.
channel_map (str) – a string listing the channel data format for each pixel.
storage (str) – what data type each value should be calculated as.

New in version 0.5.0.

Changed in version 0.6.11: Update storage type size for “long” & “quantum” values.

property interlace_scheme¶: (str) The interlace used by the image. See INTERLACE_TYPES.

New in version 0.5.2.

Changed in version 0.6.2: The interlace_scheme property now points to the image. Previously was pointing to the MagickWand.

property interpolate_method¶: (str) The interpolation method of the image. See PIXEL_INTERPOLATE_METHODS.

New in version 0.5.2.

inverse_fourier_transform(phase, magnitude=True)¶

Applies the inverse of a discrete Fourier transform. The image stack is replaced with the results. Either a pair of magnitude & phase images, or real & imaginary (HDRI).

from wand.image import Image

with Image(filename='magnitude.png') as img:
    with Image(filename='phase.png') as phase:
        img.inverse_fourier_transform(phase)
    img.save(filename='output.png')

Note

ImageMagick must have HDRI support to compute real & imaginary components (i.e. magnitude=False).

Parameters:

phase (BaseImage) – Second part (image) of the transform. Either the phase, or the imaginary part.
magnitude (bool) – If True, accept magnitude & phase input, else real & imaginary. Default True

New in version 0.5.5.

iterator_first()¶: Sets the internal image-stack iterator to the first image. Useful for prepending an image at the start of the stack.

New in version 0.6.2.

iterator_get()¶

Returns the position of the internal image-stack index.

Return type:: int

New in version 0.6.2.

iterator_last()¶: Sets the internal image-stack iterator to the last image. Useful for appending an image to the end of the stack.

New in version 0.6.2.

iterator_length()¶

Get the count of images in the image-stack.

Return type:: int

New in version 0.6.2.

iterator_next()¶

Steps the image-stack index forward by one

Return type:: bool

New in version 0.6.2.

iterator_previous()¶

Steps the image-stack index back by one.

Return type:: bool

New in version 0.6.2.

iterator_reset()¶: Reset internal image-stack iterator. Useful for iterating over the image-stack without allocating Sequence.

New in version 0.6.2.

iterator_set(index)¶

Sets the index of the internal image-stack.

Return type:: bool

New in version 0.6.2.

kmeans(number_colors=None, max_iterations=100, tolerance=0.01)¶

Reduces the number of colors in an image by applying the K-means clustering algorithm.

Note

Requires ImageMagick-7.0.10-37, or later.

Parameters:

number_colors (numbers.Integral) – the target number of colors to use as seeds.
max_iterations (numbers.Integral) – maximum number of iterations needed until convergence. Default 100.
tolerance (numbers.Real) – maximum tolerance between distrotion iterations. Default 0.01

New in version 0.6.4.

property kurtosis¶: (numbers.Real) The kurtosis of the image.

Tip

If you want both kurtosis & skewness, it would be faster to call kurtosis_channel() directly.

New in version 0.5.3.

kurtosis_channel(channel='default_channels')¶

Calculates the kurtosis and skewness of the image.

from wand.image import Image

with Image(filename='input.jpg') as img:
    kurtosis, skewness = img.kurtosis_channel()

Parameters:: channel (str) – Select which color channel to evaluate. See CHANNELS. Default 'default_channels'.
Returns:: Tuple of kurtosis & skewness values.
Return type:: tuple

New in version 0.5.3.

kuwahara(radius=1.0, sigma=None)¶

Edge preserving noise reduction filter.

https://en.wikipedia.org/wiki/Kuwahara_filter

If sigma is not given, the value will be calculated as:

sigma = radius - 0.5

To match original algorithm’s behavior, increase radius value by one:

myImage.kuwahara(myRadius + 1, mySigma)

Warning

This class method is only available with ImageMagick 7.0.8-41, or greater.

See:

Example of Kuwahara.

Parameters:

radius (numbers.Real) – Size of the filter aperture.
sigma (numbers.Real) – Standard deviation of Gaussian filter.

Raises:

WandLibraryVersionError – If system’s version of ImageMagick does not support this method.

New in version 0.5.5.

label(text, left=None, top=None, font=None, gravity=None, background_color='transparent')¶

Writes a label text into the position on top of the existing canvas. This method doesn’t autofit text like caption(). Use left & top, or gravity, to position the text.

Parameters:

text (str) – text to write.
left (numbers.Integral) – x offset in pixels.
top (numbers.Integral) – y offset in pixels.
font (wand.font.Font) – font to use. default is font of the image.
gravity (str) – text placement gravity.

New in version 0.6.8.

property length_of_bytes¶: (numbers.Integral) The original size, in bytes, of the image read. This will return 0 if the image was modified in a way that would invalidate the original length value.

New in version 0.5.4.

level(black=0.0, white=None, gamma=1.0, channel=None)¶

Adjusts the levels of an image by scaling the colors falling between specified black and white points to the full available quantum range.

If only black is given, white will be adjusted inward.

See:

Example of Level.

Parameters:

black (numbers.Real) – Black point, as a percentage of the system’s quantum range. Defaults to 0.
white (numbers.Real) – White point, as a percentage of the system’s quantum range. Defaults to 1.0.
gamma (numbers.Real) – Optional gamma adjustment. Values > 1.0 lighten the image’s midtones while values < 1.0 darken them.
channel (CHANNELS) – The channel type. Available values can be found in the CHANNELS mapping. If None, normalize all channels.

Note

Images may not be affected if the white value is equal to or less than the black value.

New in version 0.4.1.

level_colors(black_color, white_color, channel=None)¶

Maps given colors to “black” & “white” values.

Warning

This class method is only available with ImageMagick 7.0.8-54, or greater.

Parameters:

black_color (Color) – linearly map given color as “black” point.
white_color (Color) – linearly map given color as “white” point.
channel (str) – target a specific color-channel to levelize.

Raises:

WandLibraryVersionError – If system’s version of ImageMagick does not support this method.

New in version 0.5.6.

levelize(black=0.0, white=None, gamma=1.0, channel=None)¶

Reverse of level(), this method compresses the range of colors between black & white values.

If only black is given, white will be adjusted inward.

Warning

This class method is only available with ImageMagick 7.0.8-41, or greater.

Parameters:

black (numbers.Real) – Black point, as a percentage of the system’s quantum range. Defaults to 0.
white (numbers.Real) – White point, as a percentage of the system’s quantum range. Defaults to 1.0.
gamma (numbers.Real) – Optional gamma adjustment. Values > 1.0 lighten the image’s midtones while values < 1.0 darken them.
channel (CHANNELS) – The channel type. Available values can be found in the CHANNELS mapping. If None, normalize all channels.

Raises:

WandLibraryVersionError – If system’s version of ImageMagick does not support this method.

New in version 0.5.5.

levelize_colors(black_color, white_color, channel=None)¶

Reverse of level_colors(), and creates a de-contrasting gradient of given colors. This works best with grayscale images.

Warning

This class method is only available with ImageMagick 7.0.8-54, or greater.

Parameters:

black_color (Color) – tint map given color as “black” point.
white_color (Color) – tint map given color as “white” point.
channel (str) – target a specific color-channel to levelize.

Raises:

WandLibraryVersionError – If system’s version of ImageMagick does not support this method.

New in version 0.5.6.

linear_stretch(black_point=0.0, white_point=1.0)¶

Enhance saturation intensity of an image.

Parameters:

black_point (numbers.Real) – Black point between 0.0 and 1.0. Default 0.0
white_point (numbers.Real) – White point between 0.0 and 1.0. Default 1.0

New in version 0.4.1.

liquid_rescale(width, height, delta_x=1, rigidity=0)¶

Rescales the image with seam carving, also known as image retargeting, content-aware resizing, or liquid rescaling.

Parameters:

width (numbers.Integral) – the width in the scaled image
height (numbers.Integral) – the height in the scaled image
delta_x (numbers.Real) – maximum seam transversal step. 0 means straight seams. default is 0
rigidity (numbers.Real) – introduce a bias for non-straight seams. default is 1

Raises:

wand.exceptions.MissingDelegateError – when ImageMagick isn’t configured --with-lqr option.

Note

This feature requires ImageMagick to be configured --with-lqr option. Or it will raise MissingDelegateError:

See also

Seam carving — Wikipedia: The article which explains what seam carving is on Wikipedia.

Changed in version 0.7.0: Updated default value of delta_x to 1 to match CLI.

local_contrast(radius=10, strength=12.5)¶

Increase light-dark transitions within image.

Warning

This class method is only available with ImageMagick 6.9.3, or greater.

Parameters:

radius (numbers.Real) – The size of the Gaussian operator. Default value is 10.0.
strength (numbers.Real) – Percentage of blur mask to apply. Values can be between 0.0 and 100 with a default of 12.5.

New in version 0.5.7.

property loop¶: (numbers.Integral) Number of frame iterations. A value of 0 will loop forever.

magnify()¶: Quickly double an image in size. This is a convenience method. Use resize(), resample(), or sample() for more control.

New in version 0.5.5.

property matte_color¶: (wand.color.Color) The color value of the matte channel. This can also be set.

New in version 0.4.1.

property maxima¶: (numbers.Real) The maximum quantum value within the image. Value between 0.0 and quantum_range

Tip

If you want both maxima & minima, it would be faster to call range_channel() directly.

New in version 0.5.3.

property mean¶: (numbers.Real) The mean of the image, and have a value between 0.0 and quantum_range

Tip

If you want both mean & standard_deviation, it would be faster to call mean_channel() directly.

New in version 0.5.3.

mean_channel(channel='default_channels')¶

Calculates the mean and standard deviation of the image.

from wand.image import Image

with Image(filename='input.jpg') as img:
    mean, stddev = img.mean_channel()

Parameters:: channel (str) – Select which color channel to evaluate. See CHANNELS. Default 'default_channels'.
Returns:: Tuple of mean & standard_deviation values. The mean value will be between 0.0 & quantum_range
Return type:: tuple

New in version 0.5.3.

mean_shift(width, height, color_distance=0.1)¶

Recalculates pixel value by comparing neighboring pixels within a color distance, and replacing with a mean value. Works best with Gray, YCbCr, YIQ, or YUV colorspaces.

Warning

This class method is only available with ImageMagick 7.0.8-41, or greater.

Parameters:

width (numbers.Integral) – Size of the neighborhood window in pixels.
height (numbers.Integral) – Size of the neighborhood window in pixels.
color_distance (numbers.Real) – Include pixel values within this color distance.

Raises:

WandLibraryVersionError – If system’s version of ImageMagick does not support this method.

New in version 0.5.5.

merge_layers(method)¶

Composes all the image layers from the current given image onward to produce a single image of the merged layers.

The initial canvas’s size depends on the given ImageLayerMethod, and is initialized using the first images background color. The images are then composited onto that image in sequence using the given composition that has been assigned to each individual image. The method must be set with a value from IMAGE_LAYER_METHOD that is acceptable to this operation. (See ImageMagick documentation for more details.)

Parameters:: method (str) – the method of selecting the size of the initial canvas.

New in version 0.4.3.

property minima¶: (numbers.Real) The minimum quantum value within the image. Value between 0.0 and quantum_range

Tip

If you want both maxima & minima, it would be faster to call range_channel() directly.

New in version 0.5.3.

minimum_bounding_box(orientation=None)¶

Find the minimum bounding box within the image. Use properties fuzz & background_color to influence bounding box thresholds.

from wand.image import Image
from wand.drawing import Drawing

with Image(filename='kdf_black.png') as img:
    img.fuzz = img.quantum_range * 0.1
    img.background_color = 'black'
    mbr = img.minimum_bounding_box()
    with Drawing() as ctx:
        ctx.fill_color = 'transparent'
        ctx.stroke_color = 'red'
        ctx.polygon(points=mbr['points'])
        ctx.fill_color = 'red'
        ctx.stroke_color = 'transparent'
        ctx.text(1, 10, '{0:.4g}°'.format(mbr['angle']))
        ctx(img)
    img.save(filename='kdf_black_mbr.png')

Note

Requires ImageMagick-7.0.10 or later.

Parameters:: orientation (str) – sets the image orientation. Values can be 'landscape', or 'portrait'.
Returns:: a directory of MBR properties & corner points.
Return type:: dict { “points”: list [ tuple ( float, float ) ], “area”: float, “width”: float, “height”: float, “angle”: float, “unrotate”: float }

New in version 0.6.4.

mode(width, height=None)¶

Replace each pixel with the mathematical mode of the neighboring colors. This is an alias of the statistic() method.

Parameters:

width (numbers.Integral) – Number of neighboring pixels to include in mode.
height (numbers.Integral) – Optional height of neighboring pixels, defaults to the same value as width.

New in version 0.5.4.

modulate(brightness=100.0, saturation=100.0, hue=100.0)¶

Changes the brightness, saturation and hue of an image. We modulate the image with the given brightness, saturation and hue.

Parameters:

brightness (numbers.Real) – percentage of brightness
saturation (numbers.Real) – percentage of saturation
hue (numbers.Real) – percentage of hue rotation

Raises:

ValueError – when one or more arguments are invalid

New in version 0.3.4.

morphology(method=None, kernel=None, iterations=1, channel=None)¶

Manipulate pixels based on the shape of neighboring pixels.

The method determines what type of effect to apply to matching kernel shapes. Common methods can be add/remove, or lighten/darken pixel values.

The kernel describes the shape of the matching neighbors. Common shapes are provided as “built-in” kernels. See :const`KERNEL_INFO_TYPES` for examples. The format for built-in kernels is:

label:geometry

Where label is the kernel name defined in KERNEL_INFO_TYPES, and :geometry is an optional geometry size. For example:

with Image(filename='rose:') as img:
    img.morphology(method='dilate', kernel='octagon:3x3')
    # or simply
    img.morphology(method='edgein', kernel='octagon')

Custom kernels can be applied by following a similar format:

geometry:args

Where geometry is the size of the custom kernel, and args list a comma separated list of values. For example:

custom_kernel='5x3:nan,1,1,1,nan 1,1,1,1,1 nan,1,1,1,nan'
with Image(filename='rose:') as img:
    img.morphology(method='dilate', kernel=custom_kernel)

Parameters:

method (str) – effect function to apply. See MORPHOLOGY_METHODS for a list of methods.
kernel (str) – shape to evaluate surrounding pixels. See KERNEL_INFO_TYPES for a list of built-in shapes.
iterations (numbers.Integral) – Number of times a morphology method should be applied to the image. Default 1. Use -1 for unlimited iterations until the image is unchanged by the method operator.
channel (str) – Optional color channel to target. See CHANNELS

New in version 0.5.0.

Changed in version 0.5.5: Added channel argument.

motion_blur(radius=0.0, sigma=0.0, angle=0.0, channel=None)¶

Apply a Gaussian blur along an angle direction. This simulates motion movement.

See:

Example of Motion Blur.

Parameters:

radius (numbers.Real) – Aperture size of the Gaussian operator.
sigma (numbers.Real) – Standard deviation of the Gaussian operator.
angle (numbers.Real) – Apply the effect along this angle.

New in version 0.5.4.

negate(grayscale=False, channel=None)¶

Negate the colors in the reference image.

Parameters:

grayscale (bool) – if set, only negate grayscale pixels in the image.
channel (str) – the channel type. available values can be found in the CHANNELS mapping. If None, negate all channels.

New in version 0.3.8.

noise(noise_type='uniform', attenuate=1.0, channel=None)¶

Adds noise to image.

See:

Example of Add Noise.

Parameters:

noise_type (str) – type of noise to apply. See NOISE_TYPES.
attenuate (numbers.Real) – rate of distribution. Only available in ImageMagick-7. Default is 1.0.
channel (str) – Optionally target a color channel to apply noise to. See CHANNELS.

New in version 0.5.3.

Changed in version 0.5.5: Added optional channel argument.

normalize(channel=None)¶

Normalize color channels.

Parameters:: channel (str) – the channel type. available values can be found in the CHANNELS mapping. If None, normalize all channels.

oil_paint(radius=0.0, sigma=0.0)¶

Simulates an oil painting by replace each pixel with most frequent surrounding color.

Parameters:

radius (numbers.Real) – The size of the surrounding neighbors.
sigma (numbers.Real) – The standard deviation used by the Gaussian operator. This is only available with ImageMagick-7.

New in version 0.5.4.

opaque_paint(target=None, fill=None, fuzz=0.0, invert=False, channel=None)¶

Replace any color that matches target with fill. Use fuzz to control the threshold of the target match. The invert will replace all colors but the pixels matching the target color.

Parameters:

target (wand.color.Color) – The color to match.
fill (wand.color.Color) – The color to paint with.
fuzz (class:numbers.Real) – Normalized real number between 0.0 and quantum_range. Default is 0.0.
invert (bool) – Replace all colors that do not match target. Default is False.
channel (str) – Optional color channel to target. See CHANNELS

New in version 0.5.4.

Changed in version 0.5.5: Added channel parameter.

optimize_layers()¶: Attempts to crop each frame to the smallest image without altering the animation. For best results, call Image.coalesce() before manipulating any frames. For timing accuracy, any SingleImage.delay overwrites must be applied after optimizing layers.

Note

This will only affect GIF image formats.

New in version 0.5.0.

optimize_transparency()¶: Iterates over frames, and sets transparent values for each pixel unchanged by previous frame.

Note

This will only affect GIF image formats.

New in version 0.5.0.

options = None¶: (OptionDict) The mapping of internal option settings.

New in version 0.3.0.

Changed in version 0.3.4: Added 'jpeg:sampling-factor' option.

Changed in version 0.3.9: Added 'pdf:use-cropbox' option.

ordered_dither(threshold_map='threshold', channel=None)¶

Executes a ordered-based dither operations based on predetermined threshold maps.

Map	Alias	Description
threshold	1x1	Threshold 1x1 (non-dither)
checks	2x1	Checkerboard 2x1 (dither)
o2x2	2x2	Ordered 2x2 (dispersed)
o3x3	3x3	Ordered 3x3 (dispersed)
o4x4	4x4	Ordered 4x4 (dispersed)
o8x8	8x8	Ordered 8x8 (dispersed)
h4x4a	4x1	Halftone 4x4 (angled)
h6x6a	6x1	Halftone 6x6 (angled)
h8x8a	8x1	Halftone 8x8 (angled)
h4x4o		Halftone 4x4 (orthogonal)
h6x6o		Halftone 6x6 (orthogonal)
h8x8o		Halftone 8x8 (orthogonal)
h16x16o		Halftone 16x16 (orthogonal)
c5x5b	c5x5	Circles 5x5 (black)
c5x5w		Circles 5x5 (white)
c6x6b	c6x6	Circles 6x6 (black)
c6x6w		Circles 6x6 (white)
c7x7b	c7x7	Circles 7x7 (black)
c7x7w		Circles 7x7 (white)

Parameters:

threshold_map (str) – Name of threshold dither to use, followed by optional arguments.
channel (str) – Optional argument to apply dither to specific color channel. See CHANNELS.

New in version 0.5.7.

property orientation¶: (str) The image orientation. It’s a string from ORIENTATION_TYPES list. It also can be set.

New in version 0.3.0.

property page¶

The dimensions and offset of this Wand’s page as a 4-tuple: (width, height, x, y).

with Image(filename='wizard:') as img:
    img.page = (595, 842, 0, 0)

Note that since it is based on the virtual canvas, it may not equal the dimensions of an image. See the ImageMagick documentation on the virtual canvas for more information.

This attribute can also be set by using a named papersize. For example:

with Image(filename='wizard:') as img:
    img.page = 'a4'

New in version 0.4.3.

Changed in version 0.6.4: Added support for setting by papersize.

property page_height¶: (numbers.Integral) The height of the page for this wand.

New in version 0.4.3.

property page_width¶: (numbers.Integral) The width of the page for this wand.

New in version 0.4.3.

property page_x¶: (numbers.Integral) The X-offset of the page for this wand.

New in version 0.4.3.

property page_y¶: (numbers.Integral) The Y-offset of the page for this wand.

New in version 0.4.3.

parse_meta_geometry(geometry)¶

Helper method to translate geometry format, and calculate meta-characters against image dimensions.

See “Image Geometry” definitions & examples for more info: https://imagemagick.org/script/command-line-processing.php#geometry

Parameters:: geometry (str) – user string following ImageMagick’s geometry format.
Returns:: Calculated width, height, offset-x, & offset-y.
Return type:: tuple
Raises:: ValueError – If given geometry can not be parsed.

New in version 0.5.6.

percent_escape(string_format)¶

Convenience method that expands ImageMagick’s Percent Escape characters into image attribute values.

with wand.image import Image

with Image(filename='tests/assets/sasha.jpg') as img:
    print(img.percent_escape('%f %wx%h'))
    #=> sasha.jpg 204x247

Note

Not all percent escaped values can be populated as I/O operations are managed by Python, and not the CLI utility.

Parameters:: string_format (str) – The precent escaped string to be translated.
Returns:: String of expanded values.
Return type:: str

New in version 0.5.6.

polaroid(angle=0.0, caption=None, font=None, method='undefined')¶

Creates a special effect simulating a Polaroid photo.

See:

Example of Polaroid.

Parameters:

angle (numbers.Real) – applies a shadow effect along this angle.
caption (str) – Writes a message at the bottom of the photo’s border.
font (wand.font.Font) – Specify font style.
method (str) – Interpolation method. ImageMagick-7 only.

New in version 0.5.4.

polynomial(arguments)¶

Replace image with the sum of all images in a sequence by calculating the pixel values a coefficient-weight value, and a polynomial-exponent.

For example:

with Image(filename='rose:') as img:
    img.polynomial(arguments=[0.5, 1.0])

The output image will be calculated as:

\[output = 0.5 * image ^ {1.0}\]

This can work on multiple images in a sequence by calculating across each frame in the image stack.

with Image(filename='2frames.gif') as img:
    img.polynomial(arguments=[0.5, 1.0, 0.25, 1.25])

Where the results would be calculated as:

\[output = 0.5 * frame1 ^ {1.0} + 0.25 * frame2 ^ {1.25}\]

Warning

This class method is only available with ImageMagick 7.0.8-41, or greater.

Parameters:: arguments (collections.abc.Sequence) – A list of real numbers where at least two numbers (weight & exponent) are need for each image in the sequence.
Raises:: WandLibraryVersionError – If system’s version of ImageMagick does not support this method.

New in version 0.5.5.

posterize(levels=None, dither='no')¶

Reduce color levels per channel.

Parameters:

levels (numbers.Integral) – Number of levels per channel.
dither (str) – Dither method to apply. See DITHER_METHODS.

New in version 0.5.0.

quantize(number_colors, colorspace_type='undefined', treedepth=0, dither=False, measure_error=False)¶

quantize analyzes the colors within a sequence of images and chooses a fixed number of colors to represent the image. The goal of the algorithm is to minimize the color difference between the input and output image while minimizing the processing time.

Parameters:

number_colors (numbers.Integral) – The target number of colors to reduce the image.
colorspace_type (str) – Available value can be found in the COLORSPACE_TYPES. Defaults value "undefined".
treedepth (numbers.Integral) – A value between 0 & 8 where 0 will allow ImageMagick to calculate the optimal depth with Log4(number_colors). Default value is 0.
dither (bool, or str) – Perform dither operation between neighboring pixel values. If using ImageMagick-6, this can be a value of True, or False. With ImageMagick-7, use a string from DITHER_METHODS. Default False.
measure_error (bool) – Include total quantization error of all pixels in an image & quantized value.

New in version 0.4.2.

Changed in version 0.5.9: Fixed ImageMagick-7 dither argument, and added keyword defaults.

Changed in version 0.7.0: The default value for colorspace_type argument is now set to "undefeined" to match CLI behavior.

property quantum_range¶: (:class:`numbers.Integral) The maximum value of a color channel that is supported by the imagemagick library.

New in version 0.2.0.

random_threshold(low=0.0, high=1.0, channel=None)¶

Performs a random dither to force a pixel into a binary black & white state. Each color channel operarates independently from each other.

Parameters:

low (numbers.Real) – bottom threshold. Any pixel value below the given value will be rendered “0”, or no value. Given threshold value can be between 0.0 & 1.0, or 0 & quantum_range.
high (numbers.Real) – top threshold. Any pixel value above the given value will be rendered as max quantum value. Given threshold value can be between 0.0 & 1.0, or 0 & quantum_range.
channel (str) – Optional argument to apply dither to specific color channel. See CHANNELS.

New in version 0.5.7.

range_channel(channel='default_channels')¶

Calculate the minimum and maximum of quantum values in image.

from wand.image import Image

with Image(filename='input.jpg') as img:
    minima, maxima = img.range_channel()

Parameters:: channel (str) – Select which color channel to evaluate. See CHANNELS. Default 'default_channels'.
Returns:: Tuple of minima & maxima values. Each value will be between 0.0 & quantum_range.
Return type:: tuple

New in version 0.5.3.

range_threshold(low_black=0.0, low_white=None, high_white=None, high_black=None)¶

Applies soft & hard thresholding.

For a soft thresholding, parameters should be monotonically increasing:

with Image(filename=’text.png’) as img:
img.range_threshold(0.2, 0.4, 0.6, 0.8)

For a hard thresholding, parameters should be the same:

with Image(filename=’text.png’) as img:
img.range_threshold(0.4, 0.4, 0.6, 0.6)

Warning

This class method is only available with ImageMagick 7.0.8-41, or greater.

Parameters:

low_black (numbers.Real) – Define the minimum threshold value.
low_white (numbers.Real) – Define the minimum threshold value.
high_white (numbers.Real) – Define the maximum threshold value.
high_black (numbers.Real) – Define the maximum threshold value.

Raises:

WandLibraryVersionError – If system’s version of ImageMagick does not support this method.

New in version 0.5.5.

read_mask(clip_mask=None)¶

Sets the read mask where the gray values of the clip mask are used to blend during composite operations. Call this method with a None argument to clear any previously set masks.

This method is also useful for compare() method for limiting region of interest.

Warning

This method is only available with ImageMagick-7.

Parameters:: clip_mask (BaseImage) – Image to reference as blend mask.

New in version 0.5.7.

property red_primary¶: (tuple) The chromatic red primary point for the image. With ImageMagick-6 the primary value is (x, y) coordinates; however, ImageMagick-7 has (x, y, z).

New in version 0.5.2.

region(width=None, height=None, x=None, y=None, gravity=None)¶

Extract an area of the image. This is the same as crop(), but returns a new instance of Image without altering the original source image.

from wand.image import Image

with Image(filename='input.jpg') as img:
    with img.region(width=100, height=100, x=0, y=0) as area:
        pass

Parameters:

width (numbers.Integral) – Area size on the X-axis. Default value is the image’s page_width.
height (numbers.Integral) – Area size on the Y-axis.Default value is the image’s page_height.
x (numbers.Integral) – X-axis offset. This number can be negative. Default value is the image’s page_x.
y (numbers.Integral) – Y-axis offset. This number can be negative. Default value is the image’s page_y.
region (str) – Helper attribute to set x & y offset. See GRAVITY_TYPES.

Returns:

New instance of Wand.

Return type:

Image

New in version 0.6.8.

Changed in version 0.7.0: Allow x & y offset to apply relative to gravity.

remap(affinity=None, method='no')¶

Rebuild image palette with closest color from given affinity image.

See:

Example of Remap.

Parameters:

affinity (BaseImage) – reference image.
method (str) – dither method. See DITHER_METHODS. Default is 'no' dither.

New in version 0.5.3.

property rendering_intent¶: (str) PNG rendering intent. See RENDERING_INTENT_TYPES for valid options.

New in version 0.5.4.

resample(x_res=None, y_res=None, filter='undefined', blur=1)¶

Adjust the number of pixels in an image so that when displayed at the given Resolution or Density the image will still look the same size in real world terms.

Note

This method will automatically coalesce() & resample all frames in a GIF animation. For other image formats, resample() will only effect the current image in the stack. Use iterator_reset() & iterator_next() to traverse the image stack to resample all images in a multi-layer document.

with Image(filename='input.tiff') as img:
    img.iterator_reset()
    while True:
        img.resample(128, 128)
        if not img.iterator_next():
            break

Parameters:

x_res (numbers.Real) – the X resolution (density) in the scaled image. default is the original resolution.
y_res (numbers.Real) – the Y resolution (density) in the scaled image. default is the original resolution.
filter (str, numbers.Integral) – a filter type to use for resizing. choose one in FILTER_TYPES. default is 'undefined' which means IM will try to guess best one to use.
blur (numbers.Real) – the blur factor where > 1 is blurry, < 1 is sharp. default is 1

New in version 0.4.5.

reset_coords()¶: Reset the coordinate frame of the image so to the upper-left corner is (0, 0) again (crop and rotate operations change it).

New in version 0.2.0.

reset_sequence()¶: Abstract method prototype. See wand.image.Image.reset_sequence().

New in version 0.6.0.

resize(width=None, height=None, filter='undefined', blur=1)¶

Resizes the image.

Parameters:

width (numbers.Integral) – the width in the scaled image. default is the original width
height (numbers.Integral) – the height in the scaled image. default is the original height
filter (str, numbers.Integral) – a filter type to use for resizing. choose one in FILTER_TYPES. default is 'undefined' which means IM will try to guess best one to use
blur (numbers.Real) – the blur factor where > 1 is blurry, < 1 is sharp. default is 1

Changed in version 0.2.1: The default value of filter has changed from 'triangle' to 'undefined' instead.

Changed in version 0.1.8: The blur parameter changed to take numbers.Real instead of numbers.Rational.

New in version 0.1.1.

property resolution¶: (tuple) Resolution of this image.

New in version 0.3.0.

Changed in version 0.5.8: Resolution returns a tuple of float values to match ImageMagick’s behavior.

roll(x=0, y=0)¶

Shifts all pixels over by an X/Y offset.

Parameters:

x (numbers.Integral) – Number of columns to roll over. Negative value will roll pixels from right-to-left, and positive value will roll pixels from left-to-right. Default value: 0.
y (numbers.Integral) – Number of rows to roll over. Negative value will roll pixels from bottom-to-top, and positive value will roll pixels from top-to-bottm. Default value: 0.

New in version 0.6.8.

rotate(degree, background=None, reset_coords=True)¶

Rotates the image right. It takes a background color for degree that isn’t a multiple of 90.

See:

Example of Rotation.

Parameters:

degree (numbers.Real) – a degree to rotate. multiples of 360 affect nothing
background (wand.color.Color) – an optional background color. default is transparent
reset_coords (bool) – optional flag. If set, after the rotation, the coordinate frame will be relocated to the upper-left corner of the new image. By default is True.

New in version 0.2.0: The reset_coords parameter.

New in version 0.1.8.

rotational_blur(angle=0.0, channel=None)¶

Blur an image in a radius around the center of an image.

Warning

Requires ImageMagick-6.8.8 or greater.

See:

Example of Rotational Blur.

Parameters:

angle (numbers.Real) – Degrees of rotation to blur with.
channel (str) – Optional channel to apply the effect against. See CHANNELS for a list of possible values.

Raises:

WandLibraryVersionError – If system’s version of ImageMagick does not support this method.

New in version 0.5.4.

sample(width=None, height=None)¶

Resizes the image by sampling the pixels. It’s basically quicker than resize() except less quality as a trade-off.

Parameters:

width (numbers.Integral) – the width in the scaled image. default is the original width
height (numbers.Integral) – the height in the scaled image. default is the original height

New in version 0.3.4.

property sampling_factors¶

(tuple) Factors used in sampling data streams. This can be set by given it a string "4:2:2", or tuple of numbers (2, 1, 1). However the given string value will be parsed to aspect ratio (i.e. "4:2:2" internally becomes "2,1").

Note

This property is only used by YUV, DPX, & EXR encoders. For JPEG & TIFF set "jpeg:sampling-factor" on Image.options dictionary:

with Image(filename='input.jpg') as img:
    img.options['jpeg:sampling-factor'] = '2x1'

New in version 0.6.3.

scale(columns=1, rows=1)¶

Increase image size by scaling each pixel value by given columns and rows.

Parameters:

columns (numbers.Integral) – The number of columns, in pixels, to scale the image horizontally.
rows (numbers.Integral) – The number of rows, in pixels, to scale the image vertically.

New in version 0.5.7.

property scene¶: (numbers.Integral) The scene number of the current frame within an animated image.

New in version 0.5.4.

property seed¶: (numbers.Integral) The seed for random number generator.

Warning

This property is only available with ImageMagick 7.0.8-41, or greater.

New in version 0.5.5.

selective_blur(radius=0.0, sigma=0.0, threshold=0.0, channel=None)¶

Blur an image within a given threshold.

For best effects, use a value between 10% and 50% of quantum_range

from wand.image import Image

with Image(filename='photo.jpg') as img:
    # Apply 8x3 blur with a 10% threshold
    img.selective_blur(8.0, 3.0, 0.1 * img.quantum_range)

See:

Example of Selective Blur.

Parameters:

radius (numbers.Real) – Size of gaussian aperture.
sigma (numbers.Real) – Standard deviation of gaussian operator.
threshold (numbers.Real) – Only pixels within contrast threshold are effected. Value should be between 0.0 and quantum_range.
channel (str) – Optional color channel to target. See CHANNELS

New in version 0.5.3.

Changed in version 0.5.5: Added channel argument.

sepia_tone(threshold=0.8)¶

Creates a Sepia Tone special effect similar to a darkroom chemical toning.

See:: Example of Sepia Tone.
Parameters:: threshold (numbers.Real) – The extent of the toning. Value can be between 0 & quantum_range, or 0 & 1.0. Default value is 0.8 or “80%”.

New in version 0.5.7.

sequence = None¶: (collections.abc.Sequence) The list of SingleImages that the image contains.

New in version 0.3.0.

shade(gray=False, azimuth=0.0, elevation=0.0)¶

Creates a 3D effect by simulating a light from an elevated angle.

See:

Example of Shade.

Parameters:

gray (bool) – Isolate the effect on pixel intensity. Default is False.
azimuth (numbers.Real) – Angle from x-axis.
elevation (numbers.Real) – Amount of pixels from the z-axis.

New in version 0.5.0.

shadow(alpha=0.0, sigma=0.0, x=0, y=0)¶

Generates an image shadow.

Parameters:

alpha (numbers.Real) – Ratio of transparency.
sigma (numbers.Real) – Standard deviation of the gaussian filter.
x (numbers.Integral) – x-offset.
y (numbers.Integral) – y-offset.

New in version 0.5.0.

sharpen(radius=0.0, sigma=0.0, channel=None)¶

Applies a gaussian effect to enhance the sharpness of an image.

Note

For best results, ensure radius is larger than sigma.

Defaults values of zero will have ImageMagick attempt to auto-select suitable values.

See:

Example of Sharpen.

Parameters:

radius (numbers.Real) – size of gaussian aperture.
sigma (numbers.Real) – Standard deviation of the gaussian filter.
channel (str) – Optional color channel to target. See CHANNELS.

New in version 0.5.0.

Changed in version 0.5.5: Added channel argument.

shave(columns=0, rows=0)¶

Remove pixels from the edges.

Parameters:

columns (numbers.Integral) – amount to shave off both sides of the x-axis.
rows (numbers.Integral) – amount to shave off both sides of the y-axis.

New in version 0.5.0.

shear(background='WHITE', x=0.0, y=0.0)¶

Shears the image to create a parallelogram, and fill the space created with a background color.

Parameters:

background (wand.color.Color) – Color to fill the void created by shearing the image.
x (numbers.Real) – Slide the image along the X-axis.
y (numbers.Real) – Slide the image along the Y-axis.

New in version 0.5.4.

sigmoidal_contrast(sharpen=True, strength=0.0, midpoint=0.0, channel=None)¶

Modifies the contrast of the image by applying non-linear sigmoidal algorithm.

with Image(filename='photo.jpg') as img:
    img.sigmoidal_contrast(sharpen=True,
                           strength=3,
                           midpoint=0.65 * img.quantum_range)

Parameters:

sharpen (bool) – Increase the contrast when True (default), else reduces contrast.
strength (numbers.Real) – How much to adjust the contrast. Where a value of 0.0 has no effect, 3.0 is typical, and 20.0 is extreme.
midpoint (numbers.Real) – Normalized value between 0.0 & quantum_range
channel (str) – Optional color channel to target. See CHANNELS.

New in version 0.5.4.

Changed in version 0.5.5: Added channel argument.

property signature¶: (str) The SHA-256 message digest for the image pixel stream.

New in version 0.1.9.

similarity(reference, threshold=0.0, metric='undefined')¶

Scan image for best matching reference image, and return location & similarity.

Use parameter threshold to stop subimage scanning if the matching similarity value is below the given value. This is the same as the CLI -similarity-threshold option.

This method will always return a location & the lowest computed similarity value. Users are responsible for checking the similarity value to determine if a matching location is valid. Traditionally, a similarity value greater than 0.3183099 is considered dissimilar.

from wand.image import Image

dissimilarity_threshold = 0.318
similarity_threshold = 0.05
with Image(filename='subject.jpg') as img:
    with Image(filename='object.jpg') as reference:
        location, diff = img.similarity(reference,
                                        similarity_threshold)
        if diff > dissimilarity_threshold:
            print('Images too dissimilar to match')
        elif diff <= similarity_threshold:
            print('First match @ {left}x{top}'.format(**location))
        else:
            print('Best match @ {left}x{top}'.format(**location))

Warning

This operation can be slow to complete.

Parameters:

reference (wand.image.Image) – Image to search for.
threshold (numbers.Real) – Stop scanning if reference similarity is below given threshold. Value can be between 0.0 and quantum_range. Default is 0.0.
metric (str) – specify which comparison algorithm to use. See COMPARE_METRICS for a list of values. Only used by ImageMagick-7.

Returns:

List of location & similarity value. Location being a dictionary of width, height, left, & top. The similarity value is the compare distance, so a value of 0.0 means an exact match.

Return type:

tuple (dict, numbers.Real)

New in version 0.5.4: has been added.

property size¶: (tuple) The pair of (width, height).

Note

When working with animations, or other layer-based image formats, the width & height properties are referencing the last frame read into the image stack. To get the size of the entire animated images, call Image.coalesce() method immediately after reading the image.

sketch(radius=0.0, sigma=0.0, angle=0.0)¶

Simulates a pencil sketch effect. For best results, radius value should be larger than sigma.

See:

Example of Sketch.

Parameters:

radius (numbers.Real) – size of Gaussian aperture.
sigma (numbers.Real) – standard deviation of the Gaussian operator.
angle (numbers.Real) – direction of blur.

New in version 0.5.3.

property skewness¶: (numbers.Real) The skewness of the image.

Tip

If you want both kurtosis & skewness, it would be faster to call kurtosis_channel() directly.

New in version 0.5.3.

smush(stacked=False, offset=0)¶

Appends all images together. Similar behavior to concat(), but with an optional offset between images.

Parameters:

stacked (bool) – If True, will join top-to-bottom. If False, join images from left-to-right (default).
offset (numbers.Integral) – Minimum space (in pixels) between each join.

New in version 0.5.3.

solarize(threshold=0.0, channel=None)¶

Simulates extreme overexposure.

See:

Example of Solarize.

Parameters:

threshold (numbers.Real) – between 0.0 and quantum_range.
channel (str) – Optional color channel to target. See CHANNELS

New in version 0.5.3.

Changed in version 0.5.5: Added channel argument.

sparse_color(method='barycentric', colors=None, channel_mask=7)¶

Interpolates color values between points on an image.

The colors argument should be a list of tuples containing the point’s x, y coordinate & Color value.

For example:

from wand.image import Image

colors = [
    (10, 50, 'RED'),
    (174, 32, 'YELLOW'),
    (74, 123, 'ORANGE')
]
with Image(filename='input.png') as img:
    img.sparse_color('bilinear', colors)

The available interpolate methods are:

'barycentric'
'bilinear'
'shepards'
'voronoi'
'inverse'
'manhattan'

You can control which color channels are effected by building a custom channel mask. For example:

from wand.image import Image, CHANNELS

with Image(filename='input.png') as img:
    colors = [
        (50, 50, img[50, 50]),
        (100, 50, img[100, 50]),
        (50, 75, img[50, 75]),
        (100, 100, img[100, 100])
    ]
    # Only apply Voronoi to Red & Alpha channels
    mask = CHANNELS['red'] | CHANNELS['alpha']
    img.sparse_color('voronoi', colors, channel_mask=mask)

Parameters:

method (str) – Interpolate method. See SPARSE_COLOR_METHODS
colors (abc.Iterable [ int, int, Color ]) – A list of (x, y, color) tuples.
channel_mask (numbers.Integral) – Isolate specific color channels to apply interpolation. Default to RGB channels.

New in version 0.5.3.

Changed in version 0.7.0: The colors argument switch from a dict to list to allow two, or more, points of the same Color.

splice(width=None, height=None, x=None, y=None, gravity=None)¶

Partitions image by splicing a width x height rectangle at (x, y) offset coordinate. The space inserted will be replaced by the background_color value.

Parameters:

width (numbers.Integral) – number of pixel columns.
height (numbers.Integral) – number of pixel rows.
x (numbers.Integral) – offset on the X-axis.
y (numbers.Integral) – offset on the Y-axis.

New in version 0.5.3.

Changed in version 0.7.0: Allow x & y offset to apply relative to gravity.

spread(radius=0.0, method='undefined')¶

Randomly displace pixels within a defined radius.

See:

Example of Spread.

Parameters:

radius (numbers.Real) – Distance a pixel can be displaced from source. Default value is 0.0, which will allow ImageMagick to auto select a radius.
method – Interpolation method. Only available with ImageMagick-7. See PIXEL_INTERPOLATE_METHODS.

New in version 0.5.3.

Changed in version 0.5.7: Added default value to radius.

property standard_deviation¶: (numbers.Real) The standard deviation of the image.

Tip

If you want both mean & standard_deviation, it would be faster to call mean_channel() directly.

New in version 0.5.3.

statistic(stat='undefined', width=None, height=None, channel=None)¶

Replace each pixel with the statistic results from neighboring pixel values. The width & height defines the size, or aperture, of the neighboring pixels.

See:

Example of Statistic.

Parameters:

stat (str) – The type of statistic to calculate. See STATISTIC_TYPES.
width (numbers.Integral) – The size of neighboring pixels on the X-axis.
height (numbers.Integral) – The size of neighboring pixels on the Y-axis.
channel (str) – Optional color channel to target. See CHANNELS

New in version 0.5.3.

Changed in version 0.5.5: Added optional channel argument.

stegano(watermark, offset=0)¶

Hide a digital watermark of an image within the image.

from wand.image import Image

# Embed watermark
with Image(filename='source.png') as img:
    with Image(filename='gray_watermark.png') as watermark:
        print('watermark size (for recovery)', watermark.size)
        img.stegano(watermark)
    img.save(filename='public.png')

# Recover watermark
with Image(width=w, height=h, pseudo='stegano:public.png') as img:
    img.save(filename='recovered_watermark.png')

Parameters:

watermark (wand.image.Image) – Image to hide within image.
offset (numbers.Integral) – Start embedding image after a number of pixels.

New in version 0.5.4.

strip()¶: Strips an image of all profiles and comments.

New in version 0.2.0.

swirl(degree=0.0, method='undefined')¶

Swirls pixels around the center of the image. The larger the degree the more pixels will be effected.

See:

Example of Swirl.

Parameters:

degree (numbers.Real) – Defines the amount of pixels to be effected. Value between -360.0 and 360.0.
method (str) – Controls interpolation of the effected pixels. Only available for ImageMagick-7. See PIXEL_INTERPOLATE_METHODS.

New in version 0.5.7.

texture(tile)¶

Repeat tile-image across the width & height of the image.

from wand.image import Image

with Image(width=100, height=100) as canvas:
    with Image(filename='tile.png') as tile:
        canvas.texture(tile)
    canvas.save(filename='output.png')

Parameters:: tile (Image) – image to repeat across canvas.

New in version 0.5.4.

threshold(threshold=0.5, channel=None)¶

Changes the value of individual pixels based on the intensity of each pixel compared to threshold. The result is a high-contrast, two color image. It manipulates the image in place.

Parameters:

threshold (numbers.Real) – threshold as a factor of quantum. A normalized float between 0.0 and 1.0.
channel (str) – the channel type. available values can be found in the CHANNELS mapping. If None, threshold all channels.

New in version 0.3.10.

thumbnail(width=None, height=None)¶

Changes the size of an image to the given dimensions and removes any associated profiles. The goal is to produce small low cost thumbnail images suited for display on the web.

Parameters:

width (numbers.Integral) – the width in the scaled image. default is the original width
height (numbers.Integral) – the height in the scaled image. default is the original height

New in version 0.5.4.

property ticks_per_second¶: (numbers.Integral) Internal clock for animated images. .. versionadded:: 0.5.4

tint(color=None, alpha=None)¶

Applies a color vector to each pixel in the image.

See:

Example of Tint.

Parameters:

color (Color) – Color to calculate midtone.
alpha (Color) – Determine how to blend.

New in version 0.5.3.

transform(crop='', resize='')¶

Transforms the image using MagickTransformImage(), which is a convenience function accepting geometry strings to perform cropping and resizing. Cropping is performed first, followed by resizing. Either or both arguments may be omitted or given an empty string, in which case the corresponding action will not be performed. Geometry specification strings are defined as follows:

A geometry string consists of a size followed by an optional offset. The size is specified by one of the options below, where bold terms are replaced with appropriate integer values:

scale%: Height and width both scaled by specified percentage.
scale-x%xscale-y%: Height and width individually scaled by specified percentages. Only one % symbol is needed.
width: Width given, height automagically selected to preserve aspect ratio.
xheight: Height given, width automagically selected to preserve aspect ratio.
widthxheight: Maximum values of width and height given; aspect ratio preserved.
widthxheight!: Width and height emphatically given; original aspect ratio ignored.
widthxheight>: Shrinks images with dimension(s) larger than the corresponding width and/or height dimension(s).
widthxheight<: Enlarges images with dimensions smaller than the corresponding width and/or height dimension(s).
area@: Resize image to have the specified area in pixels. Aspect ratio is preserved.
X:Y: Resize at a given aspect ratio. Common aspect ratios may include 4:3 for video/tv, 3:2 for 35mm film, 16:9 for HDTV, and 2.39:1 for cinema. Aspect ratio can be used with the crop parameter, but is only available with ImageMagick version 7.0.8 or greater.

The offset, which only applies to the cropping geometry string, is given by {+-}x{+-}y, that is, one plus or minus sign followed by an x offset, followed by another plus or minus sign, followed by a y offset. Offsets are in pixels from the upper left corner of the image. Negative offsets will cause the corresponding number of pixels to be removed from the right or bottom edge of the image, meaning the cropped size will be the computed size minus the absolute value of the offset.

For example, if you want to crop your image to 300x300 pixels and then scale it by 2x for a final size of 600x600 pixels, you can call:

image.transform('300x300', '200%')

This method is a fairly thin wrapper for the C API, and does not perform any additional checking of the parameters except insofar as verifying that they are of the correct type. Thus, like the C API function, the method is very permissive in terms of what it accepts for geometry strings; unrecognized strings and trailing characters will be ignored rather than raising an error.

Parameters:

crop (str) – A geometry string defining a subregion of the image to crop to
resize (str) – A geometry string defining the final size of the image

See also

ImageMagick Geometry Specifications: Cropping and resizing geometry for the transform method are specified according to ImageMagick’s geometry string format. The ImageMagick documentation provides more information about geometry strings.

New in version 0.2.2.

Changed in version 0.5.0: Will call crop() followed by resize() in the event that MagickTransformImage() is not available.

Deprecated since version 0.6.0: Use crop() and resize() instead.

transform_colorspace(colorspace_type)¶

Transform image’s colorspace.

Parameters:: colorspace_type (str) – colorspace_type. available value can be found in the COLORSPACE_TYPES

New in version 0.4.2.

transparent_color(color, alpha, fuzz=0, invert=False)¶

Makes the color color a transparent color with a tolerance of fuzz. The alpha parameter specify the transparency level and the parameter fuzz specify the tolerance.

Parameters:

color (wand.color.Color) – The color that should be made transparent on the image, color object
alpha (numbers.Real) – the level of transparency: 1.0 is fully opaque and 0.0 is fully transparent.
fuzz (numbers.Real) – By default target must match a particular pixel color exactly. However, in many cases two colors may differ by a small amount. The fuzz member of image defines how much tolerance is acceptable to consider two colors as the same. For example, set fuzz to 10 and the color red at intensities of 100 and 102 respectively are now interpreted as the same color for the color.
invert (bool) – Boolean to tell to paint the inverse selection.

New in version 0.3.0.

Changed in version 0.6.3: Parameter fuzz type switched from Integral to Real.

transparentize(transparency)¶

Makes the image transparent by subtracting some percentage of the black color channel. The transparency parameter specifies the percentage.

Parameters:: transparency (numbers.Real) – the percentage fade that should be performed on the image, from 0.0 to 1.0

New in version 0.2.0.

transpose()¶: Creates a vertical mirror image by reflecting the pixels around the central x-axis while rotating them 90-degrees.

New in version 0.4.1.

transverse()¶: Creates a horizontal mirror image by reflecting the pixels around the central y-axis while rotating them 270-degrees.

New in version 0.4.1.

trim(color=None, fuzz=0.0, reset_coords=False, percent_background=None, background_color=None)¶

Remove solid border from image. Uses top left pixel as a guide by default, or you can also specify the color to remove.

Parameters:

color (Color) – the border color to remove. if it’s omitted top left pixel is used by default
fuzz (numbers.Real) – Defines how much tolerance is acceptable to consider two colors as the same. Value can be between 0.0, and quantum_range.
reset_coords (bool) – Reset coordinates after trimming image. Default False.
percent_background (numbers.Real) – Sets how aggressive the trim operation will be. A value of 0.0 will trim to the minimal bounding box of all matching color, and 1.0 to the most outer edge.
background_color – Local alias to background_color, and has the same effect as defining color parameter – but much faster.

New in version 0.2.1.

Changed in version 0.3.0: Optional color and fuzz parameters.

Changed in version 0.5.2: The color parameter may accept color-compliant strings.

Changed in version 0.6.0: Optional reset_coords parameter added.

Changed in version 0.6.4: Optional percent_background & background_color parameters have been added.

property type¶

(str) The image type.

Defines image type as in IMAGE_TYPES enumeration.

It may raise ValueError when the type is unknown.

New in version 0.2.2.

unique_colors()¶: Discards all duplicate pixels, and rebuilds the image as a single row.

New in version 0.5.0.

property units¶: (str) The resolution units of this image.

unsharp_mask(radius=0.0, sigma=1.0, amount=1.0, threshold=0.0, channel=None)¶

Sharpens the image using unsharp mask filter. We convolve the image with a Gaussian operator of the given radius and standard deviation (sigma). For reasonable results, radius should be larger than sigma. Use a radius of 0 and unsharp_mask() selects a suitable radius for you.

See:

Example of Unsharp Mask.

Parameters:

radius (numbers.Real) – the radius of the Gaussian, in pixels, not counting the center pixel
sigma (numbers.Real) – the standard deviation of the Gaussian, in pixels
amount (numbers.Real) – the percentage of the difference between the original and the blur image that is added back into the original
threshold (numbers.Real) – the threshold in pixels needed to apply the difference amount.
channel (str) – Optional color channel to target. See CHANNELS

New in version 0.3.4.

Changed in version 0.5.5: Added optional channel argument.

Changed in version 0.5.7: Added default values to match CLI behavior.

vignette(radius=0.0, sigma=0.0, x=0, y=0)¶

Creates a soft vignette style effect on the image.

See:

Example of Vignette.

Parameters:

radius (numbers.Real) – the radius of the Gaussian blur effect.
sigma (numbers.Real) – the standard deviation of the Gaussian effect.
x (numbers.Integral) – Number of pixels to offset inward from the top & bottom of the image before drawing effect.
y (numbers.Integral) – Number of pixels to offset inward from the left & right of the image before drawing effect.

New in version 0.5.2.

property virtual_pixel¶: (str) The virtual pixel of image. This can also be set with a value from VIRTUAL_PIXEL_METHOD … versionadded:: 0.4.1

property wand¶: Internal pointer to the MagickWand instance. It may raise ClosedImageError when the instance has destroyed already.

watermark(image, transparency=0.0, left=0, top=0)¶

Transparentized the supplied image and places it over the current image, with the top left corner of image at coordinates left, top of the current image. The dimensions of the current image are not changed.

Parameters:

image (wand.image.Image) – the image placed over the current image
transparency (numbers.Real) – the percentage fade that should be performed on the image, from 0.0 to 1.0
left (numbers.Integral) – the x-coordinate where image will be placed
top (numbers.Integral) – the y-coordinate where image will be placed

New in version 0.2.0.

wave(amplitude=0.0, wave_length=0.0, method='undefined')¶

Creates a ripple effect within the image.

See:

Example of Wave.

Parameters:

amplitude (numbers.Real) – height of wave form.
wave_length (numbers.Real) – width of wave form.
method (str) – pixel interpolation method. Only available with ImageMagick-7. See PIXEL_INTERPOLATE_METHODS

New in version 0.5.2.

wavelet_denoise(threshold=0.0, softness=0.0)¶

Removes noise by applying a wavelet transform.

Warning

This class method is only available with ImageMagick 7.0.8-41, or greater.

See:

Example of Wavelet Denoise.

Parameters:

threshold (numbers.Real) – Smoothing limit.
softness (numbers.Real) – Attenuate of the smoothing threshold.

Raises:

WandLibraryVersionError – If system’s version of ImageMagick does not support this method.

New in version 0.5.5.

white_balance()¶: Uses LAB colorspace to apply a white balance to the image.

Note

Requires ImageMagick-7.0.10-37 or later.

New in version 0.6.4.

property white_point¶: (tuple) The chromatic white point for the image. With ImageMagick-6 the primary value is (x, y) coordinates; however, ImageMagick-7 has (x, y, z).

New in version 0.5.2.

white_threshold(threshold)¶

Forces all pixels above a given color as white. Leaves pixels below threshold unaltered.

Parameters:: threshold (Color) – Color to be referenced as a threshold.

New in version 0.5.2.

property width¶: (numbers.Integral) The width of this image.

write_mask(clip_mask=None)¶

Sets the write mask which prevents pixel-value updates to the image. Call this method with a None argument to clear any previously set masks.

Warning

This method is only available with ImageMagick-7.

Parameters:: clip_mask (BaseImage) – Image to reference as blend mask.

New in version 0.5.7.

wand.image.CHANNELS = {'all_channels': 134217727, 'alpha': 16, 'black': 8, 'blue': 4, 'composite_channels': 31, 'cyan': 1, 'default_channels': 134217727, 'gray': 1, 'gray_channels': 1, 'green': 2, 'index': 32, 'magenta': 2, 'meta': 256, 'opacity': 16, 'readmask': 64, 'red': 1, 'rgb': 7, 'rgb_channels': 7, 'sync_channels': 131072, 'true_alpha': 256, 'undefined': 0, 'write_mask': 128, 'yellow': 4}¶

(dict) The dictionary of channel types.

'undefined'
'red'
'gray'
'cyan'
'green'
'magenta'
'blue'
'yellow'
'alpha'
'opacity'
'black'
'index'
'composite_channels'
'all_channels'
'sync_channels'
'default_channels'

See also

ImageMagick Color Channels: Lists the various channel types with descriptions of each

Changed in version 0.5.5: Deprecated true_alpha, rgb_channels, and gray_channels values in favor of MagickCore channel parser.

wand.image.COLORSPACE_TYPES = ('undefined', 'cmy', 'cmyk', 'gray', 'hcl', 'hclp', 'hsb', 'hsi', 'hsl', 'hsv', 'hwb', 'lab', 'lch', 'lchab', 'lchuv', 'log', 'lms', 'luv', 'ohta', 'rec601ycbcr', 'rec709ycbcr', 'rgb', 'scrgb', 'srgb', 'transparent', 'xyy', 'xyz', 'ycbcr', 'ycc', 'ydbdr', 'yiq', 'ypbpr', 'yuv')¶

(tuple) The list of colorspaces.

'undefined'
'rgb'
'gray'
'transparent'
'ohta'
'lab'
'xyz'
'ycbcr'
'ycc'
'yiq'
'ypbpr'
'yuv'
'cmyk'
'srgb'
'hsb'
'hsl'
'hwb'
'rec601luma' - Only available with ImageMagick-6
'rec601ycbcr'
'rec709luma' - Only available with ImageMagick-6
'rec709ycbcr'
'log'
'cmy'
'luv'
'hcl'
'lch'
'lms'
'lchab'
'lchuv'
'scrgb'
'hsi'
'hsv'
'hclp'
'xyy' - Only available with ImageMagick-7
'ydbdr'

See also

ImageMagick Color Management: Describes the ImageMagick color management operations

New in version 0.3.4.

wand.image.COMPARE_METRICS = ('undefined', 'absolute', 'fuzz', 'mean_absolute', 'mean_error_per_pixel', 'mean_squared', 'normalized_cross_correlation', 'peak_absolute', 'peak_signal_to_noise_ratio', 'perceptual_hash', 'root_mean_square', 'structural_similarity', 'structural_dissimilarity')¶

(tuple) The list of compare metric types used by Image.compare() and Image.similarity() methods.

'undefined'
'absolute'
'fuzz'
'mean_absolute'
'mean_error_per_pixel'
'mean_squared'
'normalized_cross_correlation'
'peak_absolute'
'peak_signal_to_noise_ratio'
'perceptual_hash' - Available with ImageMagick-7
'root_mean_square'
'structural_similarity' - Available with ImageMagick-7
'structural_dissimilarity' - Available with ImageMagick-7

See also

ImageMagick Compare Operations

New in version 0.4.3.

Changed in version 0.5.4: - Remapped MetricType enum.

wand.image.COMPOSITE_OPERATORS = ('undefined', 'alpha', 'atop', 'blend', 'blur', 'bumpmap', 'change_mask', 'clear', 'color_burn', 'color_dodge', 'colorize', 'copy_black', 'copy_blue', 'copy', 'copy_cyan', 'copy_green', 'copy_magenta', 'copy_alpha', 'copy_red', 'copy_yellow', 'darken', 'darken_intensity', 'difference', 'displace', 'dissolve', 'distort', 'divide_dst', 'divide_src', 'dst_atop', 'dst', 'dst_in', 'dst_out', 'dst_over', 'exclusion', 'hard_light', 'hard_mix', 'hue', 'in', 'intensity', 'lighten', 'lighten_intensity', 'linear_burn', 'linear_dodge', 'linear_light', 'luminize', 'mathematics', 'minus_dst', 'minus_src', 'modulate', 'modulus_add', 'modulus_subtract', 'multiply', 'no', 'out', 'over', 'overlay', 'pegtop_light', 'pin_light', 'plus', 'replace', 'saturate', 'screen', 'soft_light', 'src_atop', 'src', 'src_in', 'src_out', 'src_over', 'threshold', 'vivid_light', 'xor', 'stereo', 'freeze', 'interpolate', 'negate', 'reflect', 'soft_burn', 'soft_dodge', 'stamp', 'rmse', 'saliency_blend', 'seamless_blend')¶

(tuple) The list of composition operators

'undefined'
'alpha' - Only available with ImageMagick-7
'atop'
'blend'
'blur'
'bumpmap'
'change_mask'
'clear'
'color_burn'
'color_dodge'
'colorize'
'copy_black'
'copy_blue'
'copy'
'copy_alpha' - Only available with ImageMagick-7
'copy_cyan'
'copy_green'
'copy_magenta'
'copy_opacity' - Only available with ImageMagick-6
'copy_red'
'copy_yellow'
'darken'
'darken_intensity'
'difference'
'displace'
'dissolve'
'distort'
'divide_dst'
'divide_src'
'dst_atop'
'dst'
'dst_in'
'dst_out'
'dst_over'
'exclusion'
'freeze' - Added with ImageMagick-7.0.10
'hard_light'
'hard_mix'
'hue'
'in'
'intensity' - Only available with ImageMagick-7
'interpolate' - Added with ImageMagick-7.0.10
'lighten'
'lighten_intensity'
'linear_burn'
'linear_dodge'
'linear_light'
'luminize'
'mathematics'
'minus_dst'
'minus_src'
'modulate'
'modulus_add'
'modulus_subtract'
'multiply'
'negate' - Added with ImageMagick-7.0.10
'no'
'out'
'over'
'overlay'
'pegtop_light'
'pin_light'
'plus'
'reflect' - Added with ImageMagick-7.0.10
'replace'
'rmse' - Added with ImageMagick-7.1.0
'saliency_blend' - Added with ImageMagick-7.1.1
'saturate'
'screen'
'seamless_blend' - Added with ImageMagick-7.1.1
'soft_burn' - Added with ImageMagick-7.0.10
'soft_dudge' - Added with ImageMagick-7.0.10
'soft_light'
'src_atop'
'src'
'src_in'
'src_out'
'src_over'
'stamp' - Added with ImageMagick-7.0.10
'stereo'
'threshold'
'vivid_light'
'xor'

Changed in version 0.3.0: Renamed from COMPOSITE_OPS to COMPOSITE_OPERATORS.

Changed in version 0.5.6: Operators have been updated to reflect latest changes in C-API. For ImageMagick-6, 'add' has been renamed to 'modulus_add', 'subtract' has been renamed to 'modulus_subtract', 'divide' has been split into 'divide_dst' & 'divide_src', and 'minus' has been split into 'minus_dst' & 'minus_src'.

See also

Compositing Images ImageMagick v6 Examples: Image composition is the technique of combining images that have, or do not have, transparency or an alpha channel. This is usually performed using the IM composite command. It may also be performed as either part of a larger sequence of operations or internally by other image operators.
ImageMagick Composition Operators: Demonstrates the results of applying the various composition composition operators.

wand.image.COMPRESSION_TYPES = ('undefined', 'b44a', 'b44', 'bzip', 'dxt1', 'dxt3', 'dxt5', 'fax', 'group4', 'jbig1', 'jbig2', 'jpeg2000', 'jpeg', 'losslessjpeg', 'lzma', 'lzw', 'no', 'piz', 'pxr24', 'rle', 'zip', 'zips')¶

(tuple) The list of Image.compression types.

'undefined'
'b44a'
'b44'
'bzip'
'dxt1'
'dxt3'
'dxt5'
'fax'
'group4'
'jbig1'
'jbig2'
'jpeg2000'
'jpeg'
'losslessjpeg'
'lzma'
'lzw'
'no'
'piz'
'pxr24'
'rle'
'zip'
'zips'

New in version 0.3.6.

Changed in version 0.5.0: Support for ImageMagick-6 & ImageMagick-7

class wand.image.ChannelDepthDict(image)¶

The mapping table of channels to their depth.

Parameters:: image (Image) – an image instance

Note

You don’t have to use this by yourself. Use Image.channel_depths property instead.

New in version 0.3.0.

class wand.image.ChannelImageDict(image)¶

The mapping table of separated images of the particular channel from the image.

Parameters:: image (Image) – an image instance

Note

You don’t have to use this by yourself. Use Image.channel_images property instead.

New in version 0.3.0.

exception wand.image.ClosedImageError¶: An error that rises when some code tries access to an already closed image.

class wand.image.ConnectedComponentObject(cc_object=None)¶

Generic Python wrapper to translate CCObjectInfo structure into a class describing objects found within an image. This class is generated by Image.connected_components() method.

New in version 0.5.5.

Changed in version 0.6.3: Added merge & metric for ImageMagick 7.0.10

Changed in version 0.6.8: Added key property for ImageMagick 7.1.0

area = None¶: (numbers.Real) Quantity of pixels that make-up the objects shape.

center_x = None¶: (numbers.Real) X offset of objects centroid.

center_y = None¶: (numbers.Real) Y offset of objects centroid.

property centroid¶: (tuple (center_x, center_y)) Center of object.

clone_from_cc_object_info(cc_object)¶: Copy data from CCObjectInfo.

clone_from_extra_70A_info(cc_object)¶: Copy the additional values from CCObjectInfo structure. This is the merge & metric properties added in ImageMagick 7.0.10.

New in version 0.6.3.

clone_from_extra_710_info(cc_object)¶: Copy additional value from CCObjectInfo structure for properties added to ImageMagick 7.1.0.

New in version 0.6.8.

height = None¶: (numbers.Integral) Height of objects minimum bounding rectangle.

left = None¶: (numbers.Integral) X offset of objects minimum bounding rectangle.

mean_color = None¶: (Color) The average color of the shape.

merge = None¶: (bool) Object merge flag. Only available after ImageMagick-7.0.10. ..versionadded:: 0.6.3

metric = None¶: (list) List of doubles used by metric. Only available after ImageMagick-7.0.10. ..versionadded:: 0.6.3

property offset¶: (tuple (left, top)) Position of objects minimum bounding rectangle.

property size¶: (tuple (width, height)) Minimum bounding rectangle.

top = None¶: (numbers.Integral) Y offset of objects minimum bounding rectangle.

width = None¶: (numbers.Integral) Width of objects minimum bounding rectangle.

wand.image.DISPOSE_TYPES = ('undefined', 'none', 'background', 'previous')¶

(tuple) The list of BaseImage.dispose types.

'undefined'
'none'
'background'
'previous'

New in version 0.5.0.

wand.image.DISTORTION_METHODS = ('undefined', 'affine', 'affine_projection', 'scale_rotate_translate', 'perspective', 'perspective_projection', 'bilinear_forward', 'bilinear_reverse', 'polynomial', 'arc', 'polar', 'depolar', 'cylinder_2_plane', 'plane_2_cylinder', 'barrel', 'barrel_inverse', 'shepards', 'resize', 'sentinel', 'rigidaffine')¶

(tuple) The list of BaseImage.distort() methods.

'undefined'
'affine'
'affine_projection'
'scale_rotate_translate'
'perspective'
'perspective_projection'
'bilinear_forward'
'bilinear_reverse'
'polynomial'
'arc'
'polar'
'depolar'
'cylinder_2_plane'
'plane_2_cylinder'
'barrel'
'barrel_inverse'
'shepards'
'resize'
'sentinel'
'rigidaffine' - Only available with ImageMagick-7.0.10, or later.

New in version 0.4.1.

wand.image.DITHER_METHODS = ('undefined', 'no', 'riemersma', 'floyd_steinberg')¶

(tuple) The list of Dither methods. Used by Image.posterize() and Image.remap() methods.

'undefined'
'no'
'riemersma'
'floyd_steinberg'

New in version 0.5.0.

wand.image.EVALUATE_OPS = ('undefined', 'abs', 'add', 'addmodulus', 'and', 'cosine', 'divide', 'exponential', 'gaussiannoise', 'impulsenoise', 'laplaciannoise', 'leftshift', 'log', 'max', 'mean', 'median', 'min', 'multiplicativenoise', 'multiply', 'or', 'poissonnoise', 'pow', 'rightshift', 'rootmeansquare', 'set', 'sine', 'subtract', 'sum', 'thresholdblack', 'threshold', 'thresholdwhite', 'uniformnoise', 'xor', 'inverse_log')¶

(tuple) The list of evaluation operators. Used by Image.evaluate() method.

'undefined'
'abs'
'add'
'addmodulus'
'and'
'cosine'
'divide'
'exponential'
'gaussiannoise'
'impulsenoise'
'inverse_log' - Added with ImageMagick-7.0.10-24
'laplaciannoise'
'leftshift'
'log'
'max'
'mean'
'median'
'min'
'multiplicativenoise'
'multiply'
'or'
'poissonnoise'
'pow'
'rightshift'
'set'
'sine'
'subtract'
'sum'
'threshold'
'thresholdblack'
'thresholdwhite'
'uniformnoise'
'xor'

See also

ImageMagick Image Evaluation Operators: Describes the MagickEvaluateImageChannel method and lists the various evaluations operators

wand.image.FILTER_TYPES = ('undefined', 'point', 'box', 'triangle', 'hermite', 'hanning', 'hamming', 'blackman', 'gaussian', 'quadratic', 'cubic', 'catrom', 'mitchell', 'jinc', 'sinc', 'sincfast', 'kaiser', 'welsh', 'parzen', 'bohman', 'bartlett', 'lagrange', 'lanczos', 'lanczossharp', 'lanczos2', 'lanczos2sharp', 'robidoux', 'robidouxsharp', 'cosine', 'spline', 'sentinel')¶

(tuple) The list of filter types. Used by Image.resample() and Image.resize() methods.

'undefined'
'point'
'box'
'triangle'
'hermite'
'hanning'
'hamming'
'blackman'
'gaussian'
'quadratic'
'cubic'
'catrom'
'mitchell'
'jinc'
'sinc'
'sincfast'
'kaiser'
'welsh'
'parzen'
'bohman'
'bartlett'
'lagrange'
'lanczos'
'lanczossharp'
'lanczos2'
'lanczos2sharp'
'robidoux'
'robidouxsharp'
'cosine'
'spline'
'sentinel'

See also

ImageMagick Resize Filters: Demonstrates the results of resampling images using the various resize filters and blur settings available in ImageMagick.

wand.image.FUNCTION_TYPES = ('undefined', 'arcsin', 'arctan', 'polynomial', 'sinusoid')¶

(tuple) The list of Image.function types.

'undefined'
'arcsin'
'arctan'
'polynomial'
'sinusoid'

wand.image.GRAVITY_TYPES = ('forget', 'north_west', 'north', 'north_east', 'west', 'center', 'east', 'south_west', 'south', 'south_east', 'static')¶

(tuple) The list of gravity types.

'forget'
'north_west'
'north'
'north_east'
'west'
'center'
'east'
'south_west'
'south'
'south_east'

New in version 0.3.0.

class wand.image.HistogramDict(image)¶

Specialized mapping object to represent color histogram. Keys are colors, and values are the number of pixels.

Parameters:: image (BaseImage) – the image to get its histogram

New in version 0.3.0.

wand.image.IMAGE_LAYER_METHOD = ('undefined', 'coalesce', 'compareany', 'compareclear', 'compareoverlay', 'dispose', 'optimize', 'optimizeimage', 'optimizeplus', 'optimizetrans', 'removedups', 'removezero', 'composite', 'merge', 'flatten', 'mosaic', 'trimbounds')¶

(tuple) The list of methods for merge_layers() and compare_layers().

'undefined'
'coalesce'
'compareany' - Only used for compare_layers().
'compareclear' - Only used for compare_layers().
'compareoverlay' - Only used for compare_layers().
'dispose'
'optimize'
'optimizeimage'
'optimizeplus'
'optimizetrans'
'removedups'
'removezero'
'composite'
'merge' - Only used for merge_layers().
'flatten' - Only used for merge_layers().
'mosaic' - Only used for merge_layers().
'trimbounds' - Only used for merge_layers().

New in version 0.4.3.

wand.image.IMAGE_TYPES = ('undefined', 'bilevel', 'grayscale', 'grayscalealpha', 'palette', 'palettealpha', 'truecolor', 'truecoloralpha', 'colorseparation', 'colorseparationalpha', 'optimize', 'palettebilevelalpha')¶

(tuple) The list of image types

'undefined'
'bilevel'
'grayscale'
'grayscalealpha' - Only available with ImageMagick-7
'grayscalematte' - Only available with ImageMagick-6
'palette'
'palettealpha' - Only available with ImageMagick-7
'palettematte' - Only available with ImageMagick-6
'truecolor'
'truecoloralpha' - Only available with ImageMagick-7
'truecolormatte' - Only available with ImageMagick-6
'colorseparation'
'colorseparationalpha' - Only available with ImageMagick-7
'colorseparationmatte' - Only available with ImageMagick-6
'optimize'
'palettebilevelalpha' - Only available with ImageMagick-7
'palettebilevelmatte' - Only available with ImageMagick-6

See also

ImageMagick Image Types: Describes the MagickSetImageType method which can be used to set the type of an image

wand.image.INTERLACE_TYPES = ('undefined', 'no', 'line', 'plane', 'partition', 'gif', 'jpeg', 'png')¶

(tuple) The list of interlace schemes.

'undefined'
'no'
'line'
'plane'
'partition'
'gif'
'jpeg'
'png'

New in version 0.5.2.

class wand.image.Image(image=None, blob=None, file=None, filename=None, pseudo=None, background=None, colorspace=None, depth=None, extract=None, format=None, height=None, interlace=None, resolution=None, sampling_factors=None, units=None, width=None)¶

An image object.

Parameters:

image (Image) – makes an exact copy of the image
blob (bytes) – opens an image of the blob byte array
file (file object) – opens an image of the file object
filename (str) – opens an image of the filename string. Additional Read Modifiers are supported.
format (str) – forces filename to buffer. format to help ImageMagick detect the file format. Used only in blob or file cases
width (numbers.Integral) – the width of new blank image or an image loaded from raw data.
height (numbers.Integral) – the height of new blank image or an image loaded from raw data.
depth (numbers.Integral) – the depth used when loading raw data.
background (wand.color.Color) – an optional background color. default is transparent
resolution (collections.abc.Sequence, numbers.Integral) – set a resolution value (dpi), useful for vectorial formats (like pdf)
colorspace (str,) – sets the stack’s default colorspace value before reading any images. See COLORSPACE_TYPES.
units (str) – paired with resolution for defining an image’s pixel density. See UNIT_TYPES.

New in version 0.1.5: The file parameter.

New in version 0.1.1: The blob parameter.

New in version 0.2.1: The format parameter.

New in version 0.2.2: The width, height, background parameters.

New in version 0.3.0: The resolution parameter.

New in version 0.4.2: The depth parameter.

Changed in version 0.4.2: The depth, width and height parameters can be used with the filename, file and blob parameters to load raw pixel data.

New in version 0.5.0: The pseudo parameter.

Changed in version 0.5.4: Read constructor no longer sets “transparent” background by default. Use the background parameter to specify canvas color when reading in image.

Changed in version 0.5.7: Added the colorspace & units parameter.

Changed in version 0.6.3: Added sampling_factors parameter for working with YUV streams.

[left:right, top:bottom]

Crops the image by its left, right, top and bottom, and then returns the cropped one.

with img[100:200, 150:300] as cropped:
    # manipulated the cropped image
    pass

Like other subscriptable objects, default is 0 or its width/height:

img[:, :]        #--> just clone
img[:100, 200:]  #--> equivalent to img[0:100, 200:img.height]

Negative integers count from the end (width/height):

img[-70:-50, -20:-10]
#--> equivalent to img[width-70:width-50, height-20:height-10]

Returns:: the cropped image
Rtype:: Image

New in version 0.1.2.

property animation¶

(bool) Whether the image is animation or not. It doesn’t only mean that the image has two or more images (frames), but all frames are even the same size. It’s about image format, not content. It’s False even if image/ico consists of two or more images of the same size.

For example, it’s False for image/jpeg, image/gif, image/ico.

If image/gif has two or more frames, it’s True. If image/gif has only one frame, it’s False.

New in version 0.3.0.

Changed in version 0.3.8: Became to accept image/x-gif as well.

artifacts = None¶: (ArtifactTree) A dict mapping to image artifacts. Similar to metadata, but used to alter behavior of various internal operations.

New in version 0.5.0.

blank(width, height, background=None)¶

Creates blank image.

Parameters:

width (numbers.Integral) – the width of new blank image.
height (numbers.Integral) – the height of new blank image.
background (wand.color.Color) – an optional background color. default is transparent

Returns:

blank image

Return type:

Image

New in version 0.3.0.

channel_depths = None¶: (ChannelDepthDict) The mapping of channels to their depth. Read only.

New in version 0.3.0.

channel_images = None¶

(ChannelImageDict) The mapping of separated channels from the image.

with image.channel_images['red'] as red_image:
    display(red_image)

clear()¶: Clears resources associated with the image, leaving the image blank, and ready to be used with new image.

New in version 0.3.0.

close()¶: Closes the image explicitly. If you use the image object in with statement, it was called implicitly so don’t have to call it.

Note

It has the same functionality of destroy() method.

compare_layers(method)¶

Generates new images showing the delta pixels between layers. Similar pixels are converted to transparent. Useful for debugging complex animations.

with img.compare_layers('compareany') as delta:
    delta.save(filename='framediff_%02d.png')

Note

May not work as expected if animations are already optimized.

Parameters:: method (str) – Can be 'compareany', 'compareclear', or 'compareoverlay'
Returns:: new image stack.
Return type:: Image

New in version 0.5.0.

convert(format)¶

Converts the image format with the original image maintained. It returns a converted image instance which is new.

with img.convert('png') as converted:
    converted.save(filename='converted.png')

Parameters:: format (str) – image format to convert to
Returns:: a converted image
Return type:: Image
Raises:: ValueError – when the given format is unsupported

New in version 0.1.6.

Changed in version 0.6.11: Call MagickSetFormat() method after MagickSetImageFormat(). This will ensure image info, magick, and filename properties are aligned.

data_url()¶

Generate a base64 data-url string from the loaded image. Useful for converting small graphics into ASCII strings for HTML/CSS web development.

Returns:: a data-url formatted string.
Return type:: str

New in version 0.6.3.

classmethod from_array(array, channel_map=None, storage=None)¶

Create an image instance from a numpy array, or any other datatype that implements __array_interface__ protocol.

import numpy
from wand.image import Image

matrix = numpy.random.rand(100, 100, 3)
with Image.from_array(matrix) as img:
    img.save(filename='noise.png')

Use the optional channel_map & storage arguments to specify the order of color channels & data size. If channel_map is omitted, this method will will guess "RGB", or "CMYK" based on array shape. If storage is omitted, this method will reference the array’s typestr value, and raise a ValueError if storage-type can not be mapped.

Float values must be normalized between 0.0 and 1.0, and signed integers should be converted to unsigned values between 0 and max value of type.

Instances of Image can also be exported to numpy arrays:

with Image(filename='rose:') as img:
    matrix = numpy.array(img)

Parameters:

array (numpy.array) – Numpy array of pixel values.
channel_map (str) – Color channel layout.
storage (str) – Datatype per pixel part.

Returns:

New instance of an image.

Return type:

Image

New in version 0.5.3.

Changed in version 0.6.0: Input array now expects the shape property to be defined as `( 'height', 'width', 'channels' )`.

image_add(image)¶

Copies a given image on to the image stack. By default, the added image will be append at the end of the stack, or immediately after the current image iterator defined by iterator_set(). Use iterator_reset() before calling this method to insert the new image before existing images on the stack.

Parameters:: image (Image) – raster to add.

New in version 0.6.7.

image_get()¶: Generate & return a clone of a single image at the current image-stack index.

New in version 0.6.7.

image_remove()¶: Remove an image from the image-stack at the current index.

New in version 0.6.7.

image_set(image)¶

Overwrite current image on the image-stack with given image.

Parameters:: image (wand.image.Image) – Wand instance of images to write to stack.

New in version 0.6.7.

image_swap(i, j)¶

Swap two images on the image-stack.

Parameters:

i (numbers.Integral) – image index to replace with j
j (numbers.Integral) – image index to replace with i

New in version 0.6.7.

make_blob(format=None)¶

Makes the binary string of the image.

Parameters:: format (str) – the image format to write e.g. 'png', 'jpeg'. it is omittable
Returns:: a blob (bytes) string
Return type:: bytes
Raises:: ValueError – when format is invalid

Changed in version 0.1.6: Removed a side effect that changes the image format silently.

New in version 0.1.5: The format parameter became optional.

New in version 0.1.1.

metadata = None¶: (Metadata) The metadata mapping of the image. Read only.

New in version 0.3.0.

property mimetype¶: (str) The MIME type of the image e.g. 'image/jpeg', 'image/png'.

New in version 0.1.7.

montage(font=None, tile=None, thumbnail=None, mode='unframe', frame=None)¶

Generates a new image containing thumbnails if each previous image read.

with Image() as img:
    for file_path in ['first.png', 'second.png', 'third.png']:
        with Image(filename=file_path) as item:
            img.options['label'] = file_path
            img.image_add(item)
    style = Font('monospace', 24, 'green')
    img.montage(font=style, tile='3x1', thumbnail='15x15')
    img.save(filename='montage.png')

Parameters:

font (Font) – Define font style to use when labeling each thumbnail. Thumbnail labeling will only be rendered if 'label' value in options dict is defined.
tile (str) – The number of thunbnails per rows & column on a page. Example: "6x4".
thumbnail (str) – Preferred image size. Montage will attempt to generate a thumbnail to match the geometry. This can also define the border size on each thumbnail. Example: "120x120x+4+3>".
mode (str) – Which effect to render. Options include "frame", "unframe", and "concatenate". Default "frame".
frame (str) – Define ornamental boarder around each thrumbnail. The color of the frame is defined by the image’s matte color. Example: "15x15+3+3".

New in version 0.6.8.

morph(number_frames=5)¶

Creates a new image by linearly interpolating pixels from the image stack.

from wand.image import Image

with Image() as img:
    img.read(filename='start.png')
    img.read(filename='end.png')
    with img.morph(32) as morph:
        morph.save(filename='animation.gif')

Parameters:: number_frames (numbers.Integral) – The number of in-between images to generate.
Returns:: New Image instance.
Return type:: Image

New in version 0.7.0.

classmethod ping(file=None, filename=None, blob=None, **kwargs)¶

Ping image header into Image() object, but without any pixel data. This is useful for inspecting image meta-data without decoding the whole image.

Parameters:

blob (bytes) – reads an image from the blob byte array
file (file object) – reads an image from the file object
filename (str) – reads an image from the filename string
resolution (collections.abc.Sequence, numbers.Integral) – set a resolution value (DPI), useful for vector formats (like PDF)
format (str) – suggest image file format when reading from a blob, or file property.

New in version 0.5.6.

profiles = None¶: (ProfileDict) The mapping of image profiles.

New in version 0.5.1.

pseudo(width, height, pseudo='xc:')¶

Creates a new image from ImageMagick’s internal protocol coders.

Parameters:

width (numbers.Integral) – Total columns of the new image.
height (numbers.Integral) – Total rows of the new image.
pseudo (str) – The protocol & arguments for the pseudo image.

New in version 0.5.0.

read(file=None, filename=None, blob=None, background=None, colorspace=None, depth=None, extract=None, format=None, height=None, interlace=None, resolution=None, sampling_factors=None, units=None, width=None)¶

Read new image into Image() object.

Parameters:

blob (bytes) – reads an image from the blob byte array
file (file object) – reads an image from the file object
filename (str) – reads an image from the filename string. Additional Read Modifiers are supported.
background (Color, str) – set default background color.
colorspace (str) – set default colorspace. See COLORSPACE_TYPES.
depth (numbers.Integral) – sets bits per color sample. Usually 8, or 16.
format (str) – sets which image decoder to read with. Helpful when reading blob data with ambiguous headers.
height (numbers.Integral) – used with width to define the canvas size. Useful for reading image streams.
interlace (str) – Defines the interlacing scheme for raw data streams. See INTERLACE_TYPES.
resolution (collections.abc.Sequence, numbers.Integral) – set a resolution value (DPI), useful for vectorial formats (like PDF)
sampling_factors (collections.abc.Sequence, str) – set up/down stampling factors for YUV data stream. Usually "4:2:2"
units (str) – used with resolution, can either be 'pixelperinch', or 'pixelpercentimeter'.
width (numbers.Integral) – used with height to define the canvas size. Useful for reading image streams.

New in version 0.3.0.

Changed in version 0.5.7: Added units parameter.

Changed in version 0.6.3: Added, or documented, optional pre-read parameters: background, colorspace, depth, format, height, interlace, sampling_factors, & width.

reset_sequence()¶: Remove any previously allocated SingleImage instances in sequence attribute.

New in version 0.6.0.

save(file=None, filename=None, adjoin=True)¶

Saves the image into the file or filename. It takes only one argument at a time.

Parameters:

file (file object) – a file object to write to
filename (str) – a filename string to write to
adjoin (bool) – write all images to a single multi-image file. Only available if file format supports frames, layers, & etc.

New in version 0.1.1.

Changed in version 0.1.5: The file parameter was added.

Changed in version 6.0.0: The adjoin parameter was added.

classmethod stereogram(left, right)¶

Create a new stereogram image from two existing images.

See:

Example of Stereogram.

Parameters:

left (wand.image.Image) – Left-eye image.
right (wand.image.Image) – Right-eye image.

New in version 0.5.4.

class wand.image.ImageProperty(image)¶

The mixin class to maintain a weak reference to the parent Image object.

New in version 0.3.0.

property image¶

(Image) The parent image.

It ensures that the parent Image, which is held in a weak reference, still exists. Returns the dereferenced Image if it does exist, or raises a ClosedImageError otherwise.

Exc:: ClosedImageError when the parent Image has been destroyed

class wand.image.Iterator(image=None, iterator=None)¶

Row iterator for Image. It shouldn’t be instantiated directly; instead, it can be acquired through Image instance:

assert isinstance(image, wand.image.Image)
iterator = iter(image)

It doesn’t iterate every pixel, but rows. For example:

for row in image:
    for col in row:
        assert isinstance(col, wand.color.Color)
        print(col)

Every row is a collections.abc.Sequence which consists of one or more wand.color.Color values.

Parameters:: image (Image) – the image to get an iterator

New in version 0.1.3.

c_clear_exception = wand.api.library.PixelClearIteratorException¶: (ctypes.CFUNCTYPE) The ctypes function that clears an exception of the resource.

Note

It is an abstract attribute that has to be implemented in the subclass.

c_destroy_resource = wand.api.library.DestroyPixelIterator¶: (ctypes.CFUNCTYPE) The ctypes function that destroys the resource.

Note

It is an abstract attribute that has to be implemented in the subclass.

c_get_exception = wand.api.library.PixelGetIteratorException¶: (ctypes.CFUNCTYPE) The ctypes function that gets an exception from the resource.

Note

It is an abstract attribute that has to be implemented in the subclass.

c_is_resource = wand.api.library.IsPixelIterator¶: (ctypes.CFUNCTYPE) The ctypes predicate function that returns whether the given pointer (that contains a resource data usually) is a valid resource.

Note

It is an abstract attribute that has to be implemented in the subclass.

clone()¶: Clones the same iterator.

wand.image.KERNEL_INFO_TYPES = ('undefined', 'unity', 'gaussian', 'dog', 'log', 'blur', 'comet', 'binomial', 'laplacian', 'sobel', 'frei_chen', 'roberts', 'prewitt', 'compass', 'kirsch', 'diamond', 'square', 'rectangle', 'octagon', 'disk', 'plus', 'cross', 'ring', 'peaks', 'edges', 'corners', 'diagonals', 'line_ends', 'line_junctions', 'ridges', 'convex_hull', 'thin_se', 'skeleton', 'chebyshev', 'manhattan', 'octagonal', 'euclidean', 'user_defined')¶

(tuple) The list of builtin kernels.

'undefined'
'unity'
'gaussian'
'dog'
'log'
'blur'
'comet'
'laplacian'
'sobel'
'frei_chen'
'roberts'
'prewitt'
'compass'
'kirsch'
'diamond'
'square'
'rectangle'
'octagon'
'disk'
'plus'
'cross'
'ring'
'peaks'
'edges'
'corners'
'diagonals'
'line_ends'
'line_junctions'
'ridges'
'convex_hull'
'thin_se'
'skeleton'
'chebyshev'
'manhattan'
'octagonal'
'euclidean'
'user_defined'
'binomial'

wand.image.MORPHOLOGY_METHODS = ('undefined', 'convolve', 'correlate', 'erode', 'dilate', 'erode_intensity', 'dilate_intensity', 'iterative_distance', 'open', 'close', 'open_intensity', 'close_intensity', 'smooth', 'edgein', 'edgeout', 'edge', 'tophat', 'bottom_hat', 'hit_and_miss', 'thinning', 'thicken', 'distance', 'voronoi')¶

(tuple) The list of morphology methods.

'undefined'
'convolve'
'correlate'
'erode'
'dilate'
'erode_intensity'
'dilate_intensity'
'distance'
'open'
'close'
'open_intensity'
'close_intensity'
'smooth'
'edgein'
'edgeout'
'edge'
'tophat'
'bottom_hat'
'hit_and_miss'
'thinning'
'thicken'
'voronoi'
'iterative_distance'

class wand.image.Metadata(image)¶

Class that implements dict-like read-only access to image metadata like EXIF or IPTC headers. Most WRITE encoders will ignore properties assigned here.

Parameters:: image (Image) – an image instance

Note

You don’t have to use this by yourself. Use Image.metadata property instead.

New in version 0.3.0.

wand.image.NOISE_TYPES = ('undefined', 'uniform', 'gaussian', 'multiplicative_gaussian', 'impulse', 'laplacian', 'poisson', 'random')¶

(tuple) The list of noise types used by Image.noise() method.

'undefined'
'uniform'
'gaussian'
'multiplicative_gaussian'
'impulse'
'laplacian'
'poisson'
'random'

New in version 0.5.3.

wand.image.ORIENTATION_TYPES = ('undefined', 'top_left', 'top_right', 'bottom_right', 'bottom_left', 'left_top', 'right_top', 'right_bottom', 'left_bottom')¶: (tuple) The list of orientation types.

New in version 0.3.0.

class wand.image.OptionDict(image)¶: Free-form mutable mapping of global internal settings.

New in version 0.3.0.

Changed in version 0.5.0: Remove key check to OPTIONS. Image properties are specific to vendor, and this library should not attempt to manage the 100+ options in a whitelist.

wand.image.PAPERSIZE_MAP = {'10x13': (720, 936), '10x14': (720, 1008), '11x17': (792, 1224), '2A0': (3370, 4768), '4A0': (4768, 6741), '4x6': (288, 432), '5x7': (360, 504), '7x9': (504, 648), '8x10': (576, 720), '9x11': (648, 792), '9x12': (648, 864), 'a0': (2384, 3370), 'a1': (1684, 2384), 'a10': (74, 105), 'a2': (1191, 1684), 'a3': (842, 1191), 'a4': (595, 842), 'a4small': (595, 842), 'a5': (420, 595), 'a6': (298, 420), 'a7': (210, 298), 'a8': (147, 210), 'a9': (105, 147), 'archC': (1296, 1728), 'archa': (648, 864), 'archb': (864, 1296), 'archd': (1728, 2592), 'arche': (2592, 3456), 'b0': (2920, 4127), 'b1': (2064, 2920), 'b10': (91, 127), 'b2': (1460, 2064), 'b3': (1032, 1460), 'b4': (729, 1032), 'b5': (516, 729), 'b6': (363, 516), 'b7': (258, 363), 'b8': (181, 258), 'b9': (127, 181), 'c0': (2599, 3676), 'c1': (1837, 2599), 'c2': (1298, 1837), 'c3': (918, 1296), 'c4': (649, 918), 'c5': (459, 649), 'c6': (323, 459), 'c7': (230, 323), 'csheet': (1224, 1584), 'dsheet': (1584, 2448), 'esheet': (2448, 3168), 'executive': (540, 720), 'flsa': (612, 936), 'flse': (612, 936), 'folio': (612, 936), 'halfletter': (396, 612), 'isob0': (2835, 4008), 'isob1': (2004, 2835), 'isob10': (88, 125), 'isob2': (1417, 2004), 'isob3': (1001, 1417), 'isob4': (709, 1001), 'isob5': (499, 709), 'isob6': (354, 499), 'isob7': (249, 354), 'isob8': (176, 249), 'isob9': (125, 176), 'jisb0': (1030, 1456), 'jisb1': (728, 1030), 'jisb2': (515, 728), 'jisb3': (364, 515), 'jisb4': (257, 364), 'jisb5': (182, 257), 'jisb6': (128, 182), 'ledger': (1224, 792), 'legal': (612, 1008), 'letter': (612, 792), 'lettersmall': (612, 792), 'monarch': (279, 540), 'quarto': (610, 780), 'statement': (396, 612), 'tabloid': (792, 1224)}¶

(dict) Map of papersize names to page sizes. Each page size is a width & height tuple at a 72dpi resolution.

from wand.image import Image, PAPERSIZE_MAP

w, h = PAPERSIZE_MAP["a4"]
with Image(width=w, height=h, background="white") as img:
    img.save(filename="a4_page.png")

New in version 0.6.4.

wand.image.PIXEL_INTERPOLATE_METHODS = ('undefined', 'average', 'average9', 'average16', 'background', 'bilinear', 'blend', 'catrom', 'integer', 'mesh', 'nearest', 'spline')¶

(tuple) List of interpolate pixel methods (ImageMagick-7 only.)

'undefined'
'average'
'average9'
'average16'
'background'
'bilinear'
'blend'
'catrom'
'integer'
'mesh'
'nearest'
'spline'

New in version 0.5.0.

class wand.image.ProfileDict(image)¶

The mapping table of embedded image profiles.

Use this to get, set, and delete whole profile payloads on an image. Each payload is a raw binary string.

For example:

with Image(filename='photo.jpg') as img:
    # Extract EXIF
    with open('exif.bin', 'wb') as payload:
        payload.write(img.profiles['exif'])
    # Import ICC
    with open('color_profile.icc', 'rb') as payload:
        img.profiles['icc'] = payload.read()
    # Remove XMP
    del imp.profiles['xmp']

See also

Embedded Image Profiles for a list of supported profiles.

New in version 0.5.1.

wand.image.RENDERING_INTENT_TYPES = ('undefined', 'saturation', 'perceptual', 'absolute', 'relative')¶

(tuple) List of rendering intent types used for Image.rendering_intent property.

'undefined'
'saturation'
'perceptual'
'absolute'
'relative'

New in version 0.5.4.

wand.image.SPARSE_COLOR_METHODS = {'barycentric': 1, 'bilinear': 7, 'inverse': 19, 'manhattan': 20, 'shepards': 16, 'undefined': 0, 'voronoi': 18}¶

(tuple) List of sparse color methods used by Image.sparse_color()

'undefined'
'barycentric'
'bilinear'
'shepards'
'voronoi'
'inverse'
'manhattan'

New in version 0.5.3.

wand.image.STATISTIC_TYPES = ('undefined', 'gradient', 'maximum', 'mean', 'median', 'minimum', 'mode', 'nonpeak', 'root_mean_square', 'standard_deviation')¶

(tuple) The list of statistic types used by Image.statistic().

'undefined'
'gradient'
'maximum'
'mean'
'median'
'minimum'
'mode'
'nonpeak'
'root_mean_square'
'standard_deviation'

New in version 0.5.3.

wand.image.STORAGE_TYPES = ('undefined', 'char', 'double', 'float', 'integer', 'long', 'quantum', 'short')¶

(tuple) The list of pixel storage types.

'undefined'
'char'
'double'
'float'
'integer'
'long'
'quantum'
'short'

New in version 0.5.0.

wand.image.UNIT_TYPES = ('undefined', 'pixelsperinch', 'pixelspercentimeter')¶

(tuple) The list of resolution unit types.

'undefined'
'pixelsperinch'
'pixelspercentimeter'

See also

ImageMagick Image Units: Describes the MagickSetImageUnits method which can be used to set image units of resolution

wand.image.VIRTUAL_PIXEL_METHOD = ('undefined', 'background', 'dither', 'edge', 'mirror', 'random', 'tile', 'transparent', 'mask', 'black', 'gray', 'white', 'horizontal_tile', 'vertical_tile', 'horizontal_tile_edge', 'vertical_tile_edge', 'checker_tile')¶

(tuple) The list of virtual_pixel types.

'undefined'
'background'
'constant' - Only available with ImageMagick-6
'dither'
'edge'
'mirror'
'random'
'tile'
'transparent'
'mask'
'black'
'gray'
'white'
'horizontal_tile'
'vertical_tile'
'horizontal_tile_edge'
'vertical_tile_edge'
'checker_tile'

New in version 0.4.1.

wand.image.manipulative(function)¶: Mark the operation manipulating itself instead of returning new one.

`wand.image` — Image objects¶

Table of Contents

Related Topics

This Page

wand.image — Image objects¶

`wand.image` — Image objects¶