当把区域粘回图像时，指定的区域大小必须和区域图像的大小相同。此外，区域不能超出图像的边界。然而，原始图像的模式和区域图像的模式不必相同。如果不相同，区域图像的模式会在粘贴前被自动转换 (细节请查看后面有关颜色变换 的章节)。

这里有另一个例子：

2.4. 滚动一幅图像

   1 def roll(image, delta):
   2     "Roll an image sideways"
   3 
   4     xsize, ysize = image.size
   5 
   6     delta = delta % xsize
   7     if delta == 0: return image
   8 
   9     part1 = image.crop((0, 0, delta, ysize))
  10     part2 = image.crop((delta, 0, xsize, ysize))
  11     image.paste(part2, (0, 0, xsize-delta, ysize))
  12     image.paste(part1, (xsize-delta, 0, xsize, ysize))
  13 
  14     return image

更高级的技巧是，paste方法可以带一个透明掩模作为可选参数。在这个掩模中，像素值255 代表被粘贴的图像在那个位置上是不透明的。 (就是说，此处显示被粘贴的图像上的值。)像素值 0 表示被粘贴的图像是完全透明的。在它们之间的值表示不同程度的透明度。

Python Imaging Library 还允许对一幅多通道图像（比如RGB图像）的单个通道进行操作。split方法能够创建一组新的图像，每一幅都是原来多通道图像的一个通道。merge函数以一个模式和一组图像的元组为参数，把这些图像组成一幅新图像。下面的例子实现交换一幅RGB图像的三个通道：

2.5. 分离与合并通道

   1 r, g, b = im.split()
   2 im = Image.merge("RGB", (b, g, r))

3. 几何变换

Image 类包含resize 和 rotate 方法来缩放和旋转图像。前者带一个tuple类型的参数来表示新的图像大小，后者带一个逆时针旋转的角度值作为参数。

3.1. 简单的几何变换

   1 out = im.resize((128, 128))
   2 out = im.rotate(45) # degrees counter-clockwise

如果要将图像旋转90度的整数倍，可以使用rotate 或者transpose 方法。后者还可以用来水平或者垂直镜像一幅图像。

3.2. transpose图像

   1 out = im.transpose(Image.FLIP_LEFT_RIGHT)
   2 out = im.transpose(Image.FLIP_TOP_BOTTOM)
   3 out = im.transpose(Image.ROTATE_90)
   4 out = im.transpose(Image.ROTATE_180)
   5 out = im.transpose(Image.ROTATE_270)

There's no difference in performance or result between transpose(ROTATE) and corresponding rotate operations.

一个更通用的变换方法是 transform，在参考手册中有对它的详细叙述。

4. 颜色变换

Python Imaging Library提供convert函数，可以将图像在不同的像素格式间转换。

4.1. 转换图像颜色模式

   1     im = Image.open("lena.ppm").convert("L")

库支持在所有支持的颜色模式和"L"以及"RGB"之间的直接转换。其他颜色模式之间的转换要借助于中间图像模式（通常是"RGB" 模式）。

5. 图像增强

Python Imaging Library提供一系列的函数和模块来进行图像增强。

5.1. 滤波器

ImageFilter 模块中包含一些预定义的增强滤波器，用filter 方法来使用滤波器。

5.1.1. 使用滤波器

   1 import ImageFilter
   2 out = im.filter(ImageFilter.DETAIL)

5.2. 点操作

point 方法可以对图像的像素值进行变换（比如对比度变换）。在大多数场合，使用函数对象（带一个参数）作为参数传递给point方法。每一个像素使用这个函数对象进行变换：

5.2.1. 使用点变换

   1 # multiply each pixel by 1.2
   2 out = im.point(lambda i: i * 1.2)

用上面的技巧，你可以对图像用任何简单的表达式进行变换。你还可以结合使用point 和paste 方法来有选择的改变一幅图像：

5.2.2. 处理单个通道

   1 # split the image into individual bands
   2 source = im.split()
   3 
   4 R, G, B = 0, 1, 2
   5 
   6 # select regions where red is less than 100
   7 mask = source[R].point(lambda i: i < 100 and 255)
   8 
   9 # process the green band
  10 out = source[G].point(lambda i: i * 0.7)
  11 
  12 # paste the processed band back, but only where red was < 100
  13 source[G].paste(out, None, mask)
  14 
  15 # build a new multiband image
  16 im = Image.merge(im.mode, source)

注意用来创建mask的语法：

   1     imout = im.point(lambda i: expression and 255)

Python 只计算一个逻辑表达式的一部分，只要能确定表达式的结果其他部分就不进行计算了，并把最后计算得到的值作为表达式的值返回。因此，如果上述expression是false(0)，Python就不会检查第二个参数，因此返回0，否则返回255。

5.3. 增强

对于更多更高级的图像增强，可以使用ImageEnhance 模块。一旦从图像上创建了增强对象，你就可以尝试采用各种不同的参数进行快速的增强处理了。

你能通过这样的方法来调整图像的对比度、亮度、色彩平衡和锐度。

5.3.1. 增强图像

   1 import ImageEnhance
   2 
   3 enh = ImageEnhance.Contrast(im)
   4 enh.enhance(1.3).show("30% more contrast")

6. 图像序列

Python Imaging Library 包含对于图像序列 (也称作动画 格式)的基本支持。支持的序列格式包括 FLI/FLC, GIF, 和一些试验性的格式。TIFF 文件也能包含超过一帧的图像。

当你打开一个序列文件时，PIL 会自动加载序列中的第一帧。你可以使用seek 和tell 方法在不同帧之间移动：

6.1. 读取图像序列

   1 import Image
   2 
   3 im = Image.open("animation.gif")
   4 im.seek(1) # skip to the second frame
   5 
   6 try:
   7     while 1:
   8         im.seek(im.tell()+1)
   9         # do something to im
  10 except EOFError:
  11     pass # end of sequence

正如这个例子所示，当序列结束时，你会得到一个EOFError 异常。

注意，当前版本库的绝大多数驱动只允许你移动到下一帧(如上面例子所示)。如果要回到文件的开头，你可能必须重新打开它。

下面的迭代类让你能够使用for循环来迭代图像序列：

6.2. 一个序列迭代类

   1 class ImageSequence:
   2     def __init__(self, im):
   3         self.im = im
   4     def __getitem__(self, ix):
   5         try:
   6             if ix:
   7                 self.im.seek(ix)
   8             return self.im
   9         except EOFError:
  10             raise IndexError # end of sequence
  11 
  12 for frame in ImageSequence(im):
  13     # ...do something to frame...

7. Postscript格式打印

Python Imaging Library提供将图像、文字和图形输出到Postscript打印机的功能。这是一个简单的例子：

7.1. Drawing Postscript

   1 import Image
   2 import PSDraw
   3 
   4 im = Image.open("lena.ppm")
   5 title = "lena"
   6 box = (1*72, 2*72, 7*72, 10*72) # in points
   7 
   8 ps = PSDraw.PSDraw() # default is sys.stdout
   9 ps.begin_document(title)
  10  
  11 # draw the image (75 dpi)
  12 ps.image(box, im, 75)
  13 ps.rectangle(box) 
  14 
  15 # draw centered title
  16 ps.setfont("HelveticaNarrow-Bold", 36)
  17 w, h, b = ps.textsize(title)
  18 ps.text((4*72-w/2, 1*72-h), title) 
  19 
  20 ps.end_document()

8. 更多关于读取图像

前面叙述过，Image模块的open函数用来打开一个图像文件。在大多数情况，你只用简单的把文件名传给它就可以了：

   1 im = Image.open("lena.ppm")

如果一切正常，结果是一个Image 对象。否则，会抛出一个IOError 异常。

你可以使用一个类似文件的对象来代替文件名。这个对象必须实现read、 seek 和 tell 方法，并以二进制方式打开。 从一个打开的文件读取

   1 fp = open("lena.ppm", "rb")
   2 im = Image.open(fp)

要从字符串数据中读取一幅图像，可以使用StringIO 类： 从一个字符串读取

   1 import StringIO
   2 
   3 im = Image.open(StringIO.StringIO(buffer))

注意库在读取图像头之前，会先移动到文件头 (用seek(0))。另外，在图像数据被读取 (通过 load 方法)以后，seek方法也会被调用。如果图像文件被嵌在一个更大的文件里面，比如tar文件，你可以使用ContainerIO 或者TarIO 模块来访问它。 从一个tar压缩文档读取

   1 import TarIO
   2 
   3 fp = TarIO.TarIO("Imaging.tar", "Imaging/test/lena.ppm")
   4 im = Image.open(fp)

8.1. 控制解码器

一些解码器允许你在从文件读取图像的同时对图像进行操作。这个特性常常被用来在创建缩略图（创建缩略图的速度通常比缩略图的质量更重要）或者打印到一个黑白激光打印机（只需要图像的灰度信息）时加速图像的解码。

draft 方法能够操作一个没有被载入数据的图像对象，使得它能够尽可能与需要的模式和大小相匹配。这通过重新配置图像解码器来实现。 以草稿方式读取

   1 im = Image.open(file)
   2 print "original =", im.mode, im.size
   3 
   4 im.draft("L", (100, 100))
   5 print "draft =", im.mode, im.size

这个程序可能会打印出这样的结果：

   1 original = RGB (512, 512)
   2 draft = L (128, 128)

注意，最终获得图像可能与要求的模式和大小不完全一致。如果要求生成的图像不能超过给定的大小，可以使用thumbnail方法来代替。

概念

通过图像的[image.htm#image-size-attribute size]属性可以读取图像的大小信息。大小信息由一个包含水平和垂直像素数的二元组表示。

4. 坐标系统

Python Imaging Library 使用笛卡尔像素坐标系统，原点 (0,0)在图像的左上角。 注意：坐标值对应像素的左上角，像素(0, 0)实际中心位于(0.5, 0.5)。

坐标通常以2元组(x, y)的形式传递给库。矩形则表示成4元组的形式，左上角为第一个。比如，覆盖整个800x600像素的矩形表示为(0, 0, 800, 600)。

5. 调色板

调色板模式 ("P")使用一个彩色调色板来定义每个像素的真实颜色。

6. 信息

你可以使用[image.htm#image-info-attribute info]属性在图像中添加辅助的信息。这是一个字典对象。

在读取和存储文件时如何处理这些信息是和文件的类型有关系的（查看[formats.htm 图像文件格式]这一章）。

7. 滤波器

Image 模块

2.1. new

If the scale is omitted, it defaults to 1.0. If the offset is omitted, it defaults to 0.0.

3.19. putpalette

The ImageChops module contains a number of arithmetical image operations, called channel operations ("chops"). These can be used for various purposes, including special effects, image compositions, algorithmic painting, and more.

At this time, channel operations are only implemented for 8-bit images (e.g. "L" and "RGB").

1. 函数

Most channel operations take one or two image arguments and returns a new image. Unless otherwise noted, the result of a channel operation is always clipped to the range 0 to MAX (which is 255 for all modes supported by the operations in this module).

1.1. constant

ImageChops.constant(image, value) => image

Return a layer with the same size as the given image, but filled with the given pixel value.

1.2. duplicate

ImageChops.duplicate(image) => image

Return a copy of the given image.

1.3. invert

ImageChops.invert(image) => image

Inverts an image.

• out = MAX - image

1.4. lighter

ImageChops.lighter(image1, image2) => image

Compares the two images, pixel by pixel, and returns a new image containing the lighter values.

• out = max(image1, image2)

1.5. darker

ImageChops.darker(image1, image2) => image

Compares the two images, pixel by pixel, and returns a new image containing the darker values.

• out = min(image1, image2)

1.6. difference

ImageChops.difference(image1, image2) => image

Returns the absolute value of the difference between the two images.

• out = abs(image1 - image2)

1.7. multiply

ImageChops.multiply(image1, image2) => image

Superimposes two images on top of each other. If you multiply an image with a solid black image, the result is black. If you multiply with a solid white image, the image is unaffected.

• out = image1 * image2 / MAX

1.8. screen

ImageChops.screen(image1, image2) => image

Superimposes two inverted images on top of each other.

• out = MAX - ((MAX - image1) * (MAX - image2) / MAX)

1.9. add

ImageChops.add(image1, image2, scale, offset) => image

Adds two images, dividing the result by scale and adding the offset. If omitted, scale defaults to 1.0, and offset to 0.0.

• out = (image1 + image2) / scale + offset

1.10. subtract

ImageChops.subtract(image1, image2, scale, offset) => image

Subtracts two images, dividing the result by scale and adding the offset. If omitted, scale defaults to 1.0, and offset to 0.0.

• out = (image1 - image2) / scale + offset

1.11. blend

ImageChops.blend(image1, image2, alpha) => image

Same as the blend function in the Image module.

1.12. composite

ImageChops.composite(image1, image2, mask) => image

Same as the composite function in the Image module.

1.13. offset

ImageChops.offset(xoffset, yoffset) => image

ImageChops.offset(offset) => image

Returns a copy of the image where data has been offset by the given distances. Data wraps around the edges. If yoffset is omitted, it is assumed to be equal to xoffset.

ImageColor 模块

The ImageColor module contains colour tables and converters from CSS3-style colour specifiers to RGB tuples. This module is used by Image.new and the ImageDraw module, among others.

1. Colour Names

The ImageColor module supports the following string formats:

• Hexadecimal color specifiers, given as "#rgb" or "#rrggbb". For example, "#ff0000" specifies pure red.
• RGB functions, given as "rgb(red, green, blue)" where the colour values are integers in the range 0 to 255. Alternatively, the color values can be given as three percentages (0% to 100%). For example, "rgb(255,0,0)" and "rgb(100%,0%,0%)" both specify pure red.
• Hue-Saturation-Lightness (HSL) functions, given as "hsl(hue, saturation%, lightness%)" where hue is the colour given as an angle between 0 and 360 (red=0, green=120, blue=240), saturation is a value between 0% and 100% (gray=0%, full color=100%), and lightness is a value between 0% and 100% (black=0%, normal=50%, white=100%). For example, "hsl(0,100%,50%)" is pure red.

• Common HTML colour names. The ImageColor module provides some 140 standard colour names, based on the colors supported by the X Window system and most web browsers. Colour names are case insensitive. For example, "red" and "Red" both specify pure red.

2. 函数

2.1. getrgb

getrgb(color) => (red, green, blue)

(New in 1.1.4) Convert a colour string to an RGB tuple. If the string cannot be parsed, this function raises a ValueError exception.

2.2. getcolor

getcolor(color, mode) => (red, green, blue) or integer

(New in 1.1.4) Same as getrgb, but converts the RGB value to a greyscale value if the mode is not color or a palette image. If the string cannot be parsed, this function raises a ValueError exception.

ImageDraw 模块

The ImageDraw module provide simple 2D graphics for Image objects. You can use this module to create new images, annotate or retouch existing images, and to generate graphics on the fly for web use.

1. Example

Draw a Grey Cross Over an Image

import Image, ImageDraw

im = Image.open("lena.pgm")

draw = ImageDraw.Draw(im) draw.line((0, 0) + im.size, fill=128) draw.line((0, im.size[1], im.size[0], 0), fill=128) del draw

# write to stdout im.save(sys.stdout, "PNG")

2. Concepts

2.1. Coordinates

The graphics interface uses the same coordinate system as PIL itself, with (0, 0) in the upper left corner. Colours

To specify colours, you can use numbers or tuples just as you would use with Image.new or Image.putpixel. For "1", "L", and "I" images, use integers. For "RGB" images, use a 3-tuple containing integer values. For "F" images, use integer or floating point values.

For palette images (mode "P"), use integers as colour indexes. In 1.1.4 and later, you can also use RGB 3-tuples or colour names (see below). The drawing layer will automatically assign colour indexes, as long as you don't draw with more than 256 colours.

2.2. Colour Names

In PIL 1.1.4 and later, you can also use string constants when drawing in "RGB" images. PIL supports the following string formats:

• Hexadecimal color specifiers, given as "#rgb" or "#rrggbb". For example, "#ff0000" specifies pure red.
• RGB functions, given as "rgb(red, green, blue)" where the colour values are integers in the range 0 to 255. Alternatively, the color values can be given as three percentages (0% to 100%). For example, "rgb(255,0,0)" and "rgb(100%,0%,0%)" both specify pure red.
• Hue-Saturation-Lightness (HSL) functions, given as "hsl(hue, saturation%, lightness%)" where hue is the colour given as an angle between 0 and 360 (red=0, green=120, blue=240), saturation is a value between 0% and 100% (gray=0%, full color=100%), and lightness is a value between 0% and 100% (black=0%, normal=50%, white=100%). For example, "hsl(0,100%,50%)" is pure red.

• Common HTML colour names. The ImageDraw provides some 140 standard colour names, based on the colors supported by the X Window system and most web browsers. Colour names are case insensitive, and may contain whitespace. For example, "red" and "Red" both specify pure red.

2.3. Fonts

PIL can use bitmap fonts or OpenType/TrueType fonts.

Bitmap fonts are stored in PIL's own format, where each font typically consists of a two files, one named .pil and the other usually named .pbm. The former contains font metrics, the latter raster data.

To load a bitmap font, use the load functions in the ImageFont module.

To load a OpenType/TrueType font, use the truetype function in the ImageFont module. Note that OpenType/TrueType support is optional, and may not be supported by all PIL builds.

3. 函数

3.1. Draw

Draw(image) => Draw instance

Creates an object that can be used to draw in the given image.

Note that the image will be modified in place.

4. 方法

4.1. arc

draw.arc(xy, start, end, options)

Draws an arc (a portion of a circle outline) between the start and end angles, inside the given bounding box.

The outline option gives the colour to use for the arc.

4.2. bitmap

draw.bitmap(xy, bitmap, options)

Draws a bitmap at the given position, using the current fill colour.

4.3. chord

draw.chord(xy, start, end, options)

Same as arc, but connects the end points with a straight line.

The outline option gives the colour to use for the chord outline. The fill option gives the colour to use for the chord interior.

4.4. ellipse

draw.ellipse(xy, options)

Draws an ellipse inside the given bounding box.

The outline option gives the colour to use for the ellipse outline. The fill option gives the colour to use for the ellipse interior.

4.5. line

draw.line(xy, options)

Draws a line between the coordinates in the xy list.

The coordinate list can be any sequence object containing either 2-tuples [ (x, y), ... ] or numeric values [ x, y, ... ]. It should contain at least two coordinates.

The fill option gives the colour to use for the line.

4.6. pieslice

draw.pieslice(xy, start, end, options)

Same as arc, but also draws straight lines between the end points and the center of the bounding box.

The outline option gives the colour to use for the pieslice outline. The fill option gives the colour to use for the pieslice interior.

4.7. point

draw.point(xy, options)

Draws points (individual pixels) at the given coordinates.

The coordinate list can be any sequence object containing either 2-tuples [ (x, y), ... ] or numeric values [ x, y, ... ].

The fill option gives the colour to use for the points.

4.8. polygon

draw.polygon(xy, options)

Draws a polygon.

The polygon outline consists of straight lines between the given coordinates, plus a straight line between the last and the first coordinate.

The coordinate list can be any sequence object containing either 2-tuples [ (x, y), ... ] or numeric values [ x, y, ... ]. It should contain at least three coordinates.

The outline option gives the colour to use for the polygon outline. The fill option gives the colour to use for the polygon interior.

4.9. rectangle

draw.rectangle(box, options)

Draws a rectangle.

The box can be any sequence object containing either 2-tuples [ (x, y), (x, y) ] or numeric values [ x, y, x, y ]. It should contain two coordinates.

Note that the second coordinate pair defines a point just outside the rectangle, also when the rectangle is not filled.

The outline option gives the colour to use for the rectangle outline. The fill option gives the colour to use for the rectangle interior.

4.10. text

draw.text(position, string, options)

Draws the string at the given position. The position gives the upper right corner of the text.

The font option is used to specify which font to use. It should be an instance of the ImageFont class, typically loaded from file using the load method in the ImageFont module.

The fill option gives the colour to use for the text.

4.11. textsize

draw.textsize(string, options) => (width, height)

Return the size of the given string, in pixels.

The font option is used to specify which font to use. It should be an instance of the ImageFont class, typically loaded from file using the load method in the ImageFont module.

5. Options

5.1. outline

outline integer or tuple

5.2. fill

fill integer or tuple

5.3. font

font ImageFont instance

6. Compatibility

The Draw class contains a constructor and a number of methods which are provided for backwards compatibility only. For this to work properly, you should either use options on the drawing primitives, or these methods. Do not mix the old and new calling conventions.

6.1. ImageDraw

ImageDraw(image) => Draw instance

(Deprecated). Same as Draw. Don't use this name in new code.

6.2. setink

draw.setink(ink)

(Deprecated). Sets the colour to use for subsequent draw and fill operations.

6.3. setfill

draw.setfill(mode)

(Deprecated). Sets the fill mode.

If the mode is 0, subsequently drawn shapes (like polygons and rectangles) are outlined. If the mode is 1, they are filled.

6.4. setfont

draw.setfont(font)

(Deprecated). Sets the default font to use for the text method.

The font argument should be an instance of the ImageFont class, typically loaded from file using the load method in the ImageFont module.

ImageEnhance 模块

The ImageEnhance module contains a number of classes that can be used for image enhancement.

1. Example

Vary the Sharpness of an Image

import ImageEnhance

enhancer = ImageEnhance.Sharpness(image)

for i in range(8):

• factor = i / 4.0 enhancer.enhance(factor).show("Sharpness %f" % factor)

See the enhancer.py demo program in the Scripts directory.

2. Interface

All enhancement classes implement a common interface, containing a single method:

enhancer.enhance(factor) => image

Returns an enhanced image. The factor is a floating point value controlling the enhancement. Factor 1.0 always returns a copy of the original image, lower factors mean less colour (brightness, contrast, etc), and higher values more. There are no restrictions on this value.

3. The Color Class

The colour enhancement class is used to adjust the colour balance of an image, in a manner similar to the controls on a colour TV set. This class implements the enhancement interface as described above.

ImageEnhance.Color(image) => Color enhancer instance

Creates an enhancement object for adjusting colour in an image. A factor of 0.0 gives a black and white image, a factor of 1.0 gives the original image.

4. The Brightness Class

The brightness enhancement class is used to control the brightness of an image.

ImageEnhance.Brightness(image) => Brightness enhancer instance

Creates an enhancement object for adjusting brightness in an image. A factor of 0.0 gives a black image, factor 1.0 gives the original image.

5. The Contrast Class

The contrast enhancement class is used to control the contrast of an image, similar to the contrast control on a TV set.

ImageEnhance.Contrast(image) => Contrast enhancer instance

Creates an enhancement object for adjusting contrast in an image. A factor of 0.0 gives an solid grey image, factor 1.0 gives the original image.

6. The Sharpness Class

The sharpness enhancement class is used to control the sharpness of an image.

ImageEnhance.Sharpness(image) => Sharpness enhancer instance

Creates an enhancement object for adjusting the sharpness of an image. The factor 0.0 gives a blurred image, 1.0 gives the original image, and a factor of 2.0 gives a sharpened image.

ImageFile 模块

The ImageFile module provides support functions for the image open and save functions.

In addition, it provides a Parser class which can be used to decode an image piece by piece (e.g. while receiving it over a network connection). This class implements the same consumer interface as the standard sgmllib and xmllib modules.

1. Example

Parse An Image

import ImageFile

fp = open("lena.pgm", "rb")

p = ImageFile.Parser()

while 1:

• s = fp.read(1024) if not s:
  • break
  p.feed(s)

im = p.close()

im.save("copy.jpg")

2. 函数

2.1. Parser

ImageFile.Parser() => Parser instance

Creates a parser object. Parsers cannot be reused.

3. 方法

3.1. feed

parser.feed(data)

Feed a string of data to the parser. This method may raise an IOError exception. close

parser.close() => image or None

Tells the parser to finish decoding. If the parser managed to decode an image, it returns an Image object. Otherwise, this method raises an IOError exception.

Note: If the file cannot be identified the parser will raise an IOError exception in the close method. If the file can be identified, but not decoded (for example, if the data is damaged, or if it uses an unsupported compression method), the parser will raise an IOError exception as soon as possible, either in feed or close.

ImageFileIO 模块

ImageFileIO 模块可以用来从套接字或者任何其它流设备读取图像。

这个模块已经被deprecated。新的代码应该使用ImageFile模块提供的Parser类来代替。

1. 函数

ImageFileIO.ImageFileIO(stream)

为刘文件对象添加缓存，以提供Image.open方法所需要的seek 和 tell方法。 流对象必须实现read 和 close 方法。

ImageFilter 模块

The ImageFilter module contains definitions for a pre-defined set of filters, which can be be used in conjuction with the filter method of the Image class.

1. Example

Filter an Image

import ImageFilter

im1 = im.filter(ImageFilter.BLUR)

im2 = im.filter(ImageFilter.MinFilter(3)) im3 = im.filter(ImageFilter.MinFilter) # same as MinFilter(3)

2. Filters

The current version of the library provides the following set of predefined image enhancement filters:

BLUR, CONTOUR, DETAIL, EDGE_ENHANCE, EDGE_ENHANCE_MORE, EMBOSS, FIND_EDGES, SMOOTH, SMOOTH_MORE, and SHARPEN.

2.1. Kernel

Kernel(size, kernel, scale=None, offset=0)

(New in 1.1.5) Create a convolution kernel of the given size. In the current version, the size must be either (3, 3) or (5, 5), and the kernel argument must be a sequence containing 9 or 25 integer or floating point weights.

If the scale argument is given, the result of applying the kernel to each pixel is divided by the scale value. The default is the sum of the kernel weights.

If the offset argument is given, the value is added to the result, after it has been divided by the scale.

2.2. RankFilter

RankFilter(size, rank)

(New in 1.1.5) Create a rank filter of the given size. For each pixel in the input image, the rank filter sorts all pixels in a (size, size) environment, and copies the rank'th value to the output image.

2.3. MinFilter

MinFilter(size=3)

(New in 1.1.5) Create a min filter of the given size. For each pixel in the input image, this filter copies the smallest pixel value from a (size, size) environment to the output image.

2.4. MedianFilter

MedianFilter(size=3)

(New in 1.1.5) Create a min filter of the given size. For each pixel in the input image, this filter copies the smallest pixel value from a (size, size) environment to the output image.

2.5. MaxFilter

MaxFilter(size=3)

(New in 1.1.5) Create a max filter of the given size. For each pixel in the input image, this filter copies the largest pixel value from a (size, size) environment to the output image.

ImageFont 模块

The ImageFont module defines a class with the same name. Instances of this class store bitmap fonts, and are used with the text method of the ImageDraw class.

PIL uses it's own font file format to store bitmap fonts. You can use the pilfont utility to convert BDF and PCF font descriptors (X window font formats) to this format.

Starting with version 1.1.4, PIL can be configured to support TrueType and OpenType fonts. For earlier version, TrueType support is only available as part of the imToolkit package.

1. 例子

Here's a simple example:

import ImageFont, ImageDraw

font = ImageFont.load("arial.pil")

draw = ImageDraw.Draw(image) draw.text((10, 10), "hello", font=font)

2. 函数

2.1. load

ImageFont.load(file) => Font instance

Loads a font from the given file, and returns the corresponding font object. If this function fails, it raises an IOError exception.

2.2. load_path

ImageFont.load_path(file) => Font instance

Same as load, but searches for the file along sys.path if it's not found in the current directory.

2.3. truetype

ImageFont.truetype(file, size) => Font instance

Load a TrueType or OpenType font file, and create a font object. This function loads a font object from the given file, and creates a font object for a font of the given size.

On Windows, if the given file name does not exist, the loader also looks in Windows fonts directory.

This function requires the _imagingft service.

2.4. load_default

ImageFont.load_default() => Font instance

(New in 1.1.4) Load a "better than nothing" default font.

3. 方法

Font objects must implement the following methods, which are used by the ImageDraw layer.

3.1. getsize

font.getsize(text) => (width, height)

Returns the width and height of the given text, as a 2-tuple.

3.2. getmask

font.getmask(text) => Image object

Returns a bitmap for the text. The bitmap should be an internal PIL storage memory instance (as defined by the Image.core interface module).

If the font uses antialiasing, the bitmap should have mode "L" and use a maximum value of 255. Otherwise, it should have mode "1".

ImageGrab 模块

The ImageGrab module can be used to copy the contents of the screen or the clipboard to a PIL image memory.

The current version works on Windows only.

1. 函数

1.1. grab

ImageGrab.grab() => image

ImageGrab.grab(bbox) => image

(New in 1.1.3) Take a snapshot of the screen, and return an "RGB" image. The bounding box argument can be used to copy only a part of the screen.

1.2. grabclipboard

ImageGrab.grabclipboard() => image or list of strings or None

(New in 1.1.4) Take a snapshot of the clipboard contents, and return an image object or a list of file names. If the clipboard doesn't contain image data, this function returns None.

You can use isinstance to check if the function returned a valid image object, or something else:

   1 im = ImageGrab.grabclipboard()
   2 
   3 if isinstance(im, Image.Image):
   4     ... got an image ...
   5 elif im:
   6    for filename in im:
   7        try:
   8            im = Image.open(filename)
   9        except IOError:
  10            pass # ignore this file
  11        else:
  12            ... got an image ...
  13 else:
  14     ... clipboard empty ...

ImageOps 模块

(New in 1.1.3) The ImageOps module contains a number of 'ready-made' image processing operations. This module is somewhat experimental, and most operators only work on L and RGB images.

1. 函数

1.1. autocontrast

ImageOps.autocontrast(image, cutoff=0) => image

Maximize (normalize) image contrast. This function calculates a histogram of the input image, removes cutoff percent of the lightest and darkest pixels from the histogram, and remaps the image so that the darkest pixel becomes black (0), and the lightest becomes white (255).

1.2. colorize

ImageOps.colorize(image, black, white) => image

Colorize grayscale image. The black and white arguments should be RGB tuples or color names; this function calculates a colour wedge mapping all black pixels in the source image to the first colour, and all white pixels to the second colour.

1.3. crop

ImageOps.crop(image, border=0) => image

Remove border pixels from all four edges. This function works on all image modes.

1.4. deform

ImageOps.deform(image, deformer) => image

Deform the image using the given deformer object.

1.5. equalize

ImageOps.equalize(image) => image

Equalize the image histogram. This function applies a non-linear mapping to the input image, in order to create a uniform distribution of grayscale values in the output image.

1.6. expand

ImageOps.expand(image, border=0, fill=0) => image

Add border pixels of border to the image, at all four edges.

1.7. fit

ImageOps.fit(image, size, method, bleed, centering) => image

Returns a sized and cropped version of the image, cropped to the requested aspect ratio and size. The size argument is the requested output size in pixels, given as a (width, height) tuple.

The method argument is what resampling method to use. The default is Image.NEAREST (nearest neighbour).

The bleed argument allows you to remove a border around the outside the image (from all four edges). The value is a decimal percentage (use 0.01 for one percent). The default value is 0 (no border).

The centering argument is used to control the cropping position. (0.5, 0.5) is center cropping (i.e. if cropping the width, take 50% off of the left side (and therefore 50% off the right side), and same with top/bottom).

(0.0, 0.0) will crop from the top left corner (i.e. if cropping the width, take all of the crop off of the right side, and if cropping the height, take all of it off the bottom).

(1.0, 0.0) will crop from the bottom left corner, etc. (i.e. if cropping the width, take all of the crop off the left side, and if cropping the height take none from the top (and therefore all off the bottom)).

The fit function was contributed by Kevin Cazabon.

1.8. flip

ImageOps.flip(image) => image

Flip the image vertically (top to bottom).

1.9. grayscale

ImageOps.grayscale(image) => image

Convert the image to grayscale.

1.10. invert

ImageOps.invert(image) => image

Invert (negate) the image.

1.11. mirror

ImageOps.mirror(image) => image

Flip image horizontally (left to right).

1.12. posterize

ImageOps.posterize(image, bits) => image

Reduce the number of bits for each colour channel.

1.13. solarize

ImageOps.solarize(image, threshold=128) => image

Invert all pixel values above the given threshold.

ImagePath 模块

ImagePath 模块用来存储和操作二维向量数据。路径对象可以被传到ImageDraw模块中的方法中。

1. 函数

1.1. Path

ImagePath.Path(coordinates) => Path instance

创建一个路径对象。坐标表coordinates可以是任何包含2元组 [ (x, y), ... ] 或者数字值 [ x, y, ... ]的序列对象。

ImagePalette 模块

要使用ImagePalette 类和相关的函数，要导入ImagePalette 模块。

1. 例子

将调色板添加到图像

import ImagePalette

im.putpalette(ImagePalette.ImagePalette("RGB"))

2. 类

2.1. ImagePalette

ImagePalette.ImagePalette(mode="RGB") => palette instance

这个构造函数创建一个调色板，将"P"模式映射到指定的模式。调色板被初始化成线性灰度的斜面（译注：索引值0映射为灰度0，索引值1映射为灰度1，以此类推）。

ImageSequence 模块

ImageSequence 模块包含让你能够跌代一个图像序列的包装类。

1. 函数

1.1. Iterator

ImageSequence.Iterator(image) => Iterator instance

创建一个Iterator 对象让你可以循环遍历一个序列中的所有帧。

2. 方法

Iterator 类实现 [] 运算符。

2.1. Operator []

你可以用大于等于0的整数作参数调用这个运算符。如果没有足够的帧，跌代子会抛出一个IndexError异常。

ImageStat 模块

The ImageStat module calculates global statistics for an image, or for a region of an image.

1. 函数

1.1. Stat

ImageStat.Stat(image) => Stat instance

ImageStat.Stat(image, mask) => Stat instance

Calculates statistics for the give image. If a mask is included, only the regions covered by that mask are included in the statistics.

ImageStat.Stat(list) => Stat instance

Same as above, but calculates statistics for a previously calculated histogram.

2. Attributes

The following attributes contain a sequence with one element for each layer in the image. All attributes are lazily evaluated; if you don't need a value, it won't be calculated.

2.1. extrema

stat.extrema

(Attribute). Get min/max values for each band in the image.

2.2. count

stat.count

(Attribute). Get total number of pixels.

2.3. sum

stat.sum

(Attribute). Get sum of all pixels.

2.4. sum2

stat.sum2

(Attribute). Squared sum of all pixels.

2.5. pixel

stat.mean

(Attribute). Average pixel level.

2.6. median

stat.median

(Attribute). Median pixel level.

2.7. rms

stat.rms

(Attribute). RMS (root-mean-square).

2.8. var

stat.var

(Attribute). Variance.

2.9. stddev

stat.stddev

(Attribute). Standard deviation.

ImageTk 模块

The ImageTk module contains support to create and modify Tkinter BitmapImage and PhotoImage objects.

For examples, see the demo programs in the Scripts directory.

1. The BitmapImage Class

ImageTk.BitmapImage(image, options) => BitmapImage instance

Create a Tkinter-compatible bitmap image, which can be used everywhere Tkinter expects an image object.

The given image must have mode "1". Pixels having value 0 are treated as transparent. Options, if any, are passed to Tkinter. The most commonly used option is foreground, which is used to specify the colour for the non-transparent parts. See the Tkinter documentation for information on how to specify colours.

2. The PhotoImage Class

ImageTk.PhotoImage(image) => PhotoImage instance

Creates a Tkinter-compatible photo image, which can be used everywhere Tkinter expects an image object. If the image is an RGBA image, pixels having alpha 0 are treated as transparent.

ImageTk.PhotoImage(mode, size) => PhotoImage instance

Same as above, but creates an empty (transparent) photo image object. Use paste to copy image data to this object.

photo.paste(image, box)

Pastes an image into the photo image. The box is a 4-tuple defining the left, upper, right, and lower pixel coordinate. If the box is omitted, or None, all of the image is assumed. In all cases, the size of the pasted image must match the size of the region. If the image mode does not match the photo image mode, conversions are automatically applied.

ImageWin 模块

The ImageWin module contains support to create and display images on Windows. It can be used with PythonWin as well as other user interface toolkits that provides access to Windows device contexts.

1. Dib 类

1.1. Dib

ImageWin.Dib(image) => Dib instance

ImageWin.Dib(mode, size) => Dib instance

This constructor creates a Windows bitmap from a PIL image, or from a given mode and size.

The mode can be one of "1", "L", or "RGB".

If the display requires a palette, this constructor creates a suitable palette and associates it with the image. For an "L" image, 128 greylevels are allocated. For an "RGB" image, a 6x6x6 colour cube is used, together with 20 greylevels.

To make sure that palettes work properly under Windows, you must call the palette method upon certain events from Windows. See the method descriptions below.

2. 方法

2.1. expose

dib.expose(hdc)

Expose (draw) the image using the given device context handle. The handle is an integer representing a Windows HDC handle.

In PythonWin, you can use the GetHandleAttrib method of the CDC class to get a suitable handle.

2.2. draw

dib.draw(hdc, destination)

dib.draw(hdc, destination, source)

Same as expose, but allows you to specify where to draw the image, and what part of it to draw.

The destination and source areas are given as 4-tuple rectangles. If the source is omitted, the entire image is copied. If the source and the destination have different sizes, the image is resized as necessary.

2.3. palette

dib.palette(handle)

Installs the palette associated with the image in the given device context. The handle argument is an integer representing a Windows HDC handle.

This method should be called upon QUERYNEWPALETTE and PALETTECHANGED events from Windows. If this method returns a non-zero value, one or more display palette entries were changed since the last updated, and the image should be redrawn.

2.4. paste

dib.paste(image, bbox)

Paste an image into the bitmap image. The bbox argument is a 4-tuple defining the left, upper, right, and lower pixel coordinate. If None is used instead of a tuple, the entire image is copied. In all cases, the size of the pasted image must match the size of the region. If the image mode does not match the bitmap mode, conversions are automatically applied.

PSDraw 模块

The PSDraw module provides simple print support for Postscript printers. You can print text, graphics and images through this module.

1. Classes

1.1. PSDraw

PSDraw.PSDraw(file) => PSDraw instance

Sets up printing to the given file. If file is omitted, sys.stdout is assumed.

2. PSDraw 方法

2.1. begin

ps.begin_document()

Sets up printing of a document.

2.2. end

ps.end_document()

Ends printing.

2.3. line

ps.line(from, to)

Draws a line between the two points. Coordinates are given in Postscript point coordinates (72 points per inch, (0, 0) is the lower left corner of the page).

2.4. rectangle

ps.rectangle(box)

Draws a rectangle.

2.5. text

ps.text(position, text)

ps.text(position, text, alignment)

Draws text at the given position. You must use setfont before calling this method.

2.6. setfont

ps.setfont(font, size)

Selects which font to use. The font argument is a Postscript font name, the size argument is given in points.

2.7. setink

ps.setink(ink)

Selects the pixel value to use with subsequent operations.

2.8. setfill

ps.setfill(onoff)

Selects if subsequent rectangle operations should draw filled rectangles or just outlines.

ImageCrackCode 模块 (PIL Plus)

The ImageCrackCode module allows you to detect and measure features in an image. This module is only available in the PIL Plus package.

1. 函数

1.1. CrackCode

CrackCode(image, position) => CrackCode instance

Identifies a feature in the given image. If the position is omitted, the constructor searches from the top left corner.

2. 方法 and attributes

2.1. area

cc.area

(attribute). The feature area, in pixels.

2.2. bbox

cc.bbox

(attribute). The bounding box, given as a 4-tuple (left, upper, right, lower).

2.3. caliper

cc.caliper

(attribute). The caliper size, given as a 2-tuple (height, width).

2.4. centroid

cc.centroid

(attribute). The center of gravity.

2.5. edge

cc.edge

(attribute). True if the feature touches the edges of the image, zero otherwise.

2.6. links

cc.links

(attribute). The number of links in the crack code chain.

2.7. offset

cc.offset

(attribute). The offset from the upper left corner of the image, to the feature's bounding box,

2.8. start

cc.start

(attribute). The first coordinate in the crack code chain.

2.9. top

cc.top

(attribute). The topmost coordinate in the crack code chain.

2.10. hit

cc.hit(xy) => flag

Check if the given point is inside this feature.

2.11. topath

cc.topath(xy) => path

Return crack code outline as an ImagePath object.

2.12. getmask

cc.getmask() => image

Get filled feature mask, as an image object.

2.13. getoutline

cc.getoutline() => image

Get feature outline, as an image object.

ImageMath 模块 (PIL Plus)

ImageMath 模块可以用来计算"图像表达式"。这个模块仅提供一个eval函数，它带一个表达式串以及一幅或多幅图像作参数。

这个模块仅在 PIL Plus 包中可用。

1. 例子

2. 函数

2.1. eval

eval(expression, environment) => image or value

在给定的environment中计算表达式，并返回结果。结果是一幅图像或者是单个数值(整数、浮点数或一个像素元组)。

3. 表达式语法

表达式是标准的Python表达式，但它们是在非标准的环境中计算的。你可以像平常一样使用一般的PIL方法，也可以使用下列运算符和函数：

3.1. 运算符

3.2. 内建函数

第三部分：工具手册

pildriver 工具

The pildriver tool gives access to most PIL functions from your operating system's command-line interface.

$ pildriver program

When called as a script, the command-line arguments are passed to a PILDriver instance (see below). If there are no command-line arguments, the module runs an interactive interpreter, each line of which is split into space-separated tokens and passed to the execute method.

The pildriver tool was contributed by Eric S. Raymond.

1. 例子

The following example loads test.png, crops out a portion of its upper-left-hand corner and displays the cropped portion:

$ pildriver show crop 0 0 200 300 open test.png

The following example loads test.tiff, rotates it 30 degrees, and saves the result as rotated.png (in PNG format):

$ pildriver save rotated.png rotate 30 open test.tiff

2. The PILDriver Class

The pildriver module provides a single class called PILDriver.

An instance of the PILDriver class is essentially a software stack machine (Polish-notation interpreter) for sequencing PIL image transformations. The state of the instance is the interpreter stack.

The only method one will normally invoke after initialization is the execute method. This takes an argument list of tokens, pushes them onto the instance's stack, and then tries to clear the stack by successive evaluation of PILdriver operators. Any part of the stack not cleaned off persists and is part of the evaluation context for the next call of the execute method.

PILDriver doesn't catch any exceptions, on the theory that these are actually diagnostic information that should be interpreted by the calling code.

3. 方法

In the method descriptions below, each line lists a command token, followed by <>-enclosed arguments which describe how the method interprets the entries on the stack. Each argument specification begins with a type specification: either int, float, string, or image.

All operations consume their arguments off the stack (use dup to keep copies around). Use verbose 1 to see the stack state displayed before each operation.

add <image:pic1> <image:pic2> <int:offset> <float:scale>

• Pop the two top images, produce the scaled sum with offset.

blend <image:pic1> <image:pic2> <float:alpha>

• Replace two images and an alpha with the blended image.

brightness <image:pic1>

• Enhance brightness in the top image.

clear

• Clear the stack.

color <image:pic1>

• Enhance colour in the top image.

composite <image:pic1> <image:pic2> <image:mask>

• Replace two images and a mask with their composite.

contrast <image:pic1>

• Enhance contrast in the top image.

convert <string:mode> <image:pic1>

• Convert the top image to the given mode.

copy <image:pic1>

• Make and push a true copy of the top image.

crop <int:left> <int:upper> <int:right> <int:lower> <image:pic1>

• Crop and push a rectangular region from the current image.

darker <image:pic1> <image:pic2>

• Pop the two top images, push an image of the darker pixels of both.

difference <image:pic1> <image:pic2>

• Pop the two top images, push the difference image

draft <string:mode> <int:xsize> <int:ysize>

• Configure the loader for a given mode and size.

dup

• Duplicate the top-of-stack item.

filter <string:filtername> <image:pic1>

• Process the top image with the given filter.

format <image:pic1>

• Push the format of the top image onto the stack.

getbbox

• Push left, upper, right, and lower pixel coordinates of the top image.

extrema

• Push minimum and maximum pixel values of the top image.

invert <image:pic1>

• Invert the top image.

lighter <image:pic1> <image:pic2>

• Pop the two top images, push an image of the lighter pixels of both.

merge <string:mode> <image:pic1> [<image:pic2> [<image:pic3> [<image:pic4>]]]

• Merge top-of stack images in a way described by the mode.

mode <image:pic1>

• Push the mode of the top image onto the stack.

multiply <image:pic1> <image:pic2>

• Pop the two top images, push the multiplication image.

new <int:xsize> <int:ysize> <int:color>:

• Create and push a greyscale image of given size and colour.

offset <int:xoffset> <int:yoffset> <image:pic1>

• Offset the pixels in the top image.

open <string:filename>

• Open the indicated image, read it, push the image on the stack.

paste <image:figure> <int:xoffset> <int:yoffset> <image:ground>

• Paste figure image into ground with upper left at given offsets.

pop

• Discard the top element on the stack.

resize <int:xsize> <int:ysize> <image:pic1>

• Resize the top image.

rotate <int:angle> <image:pic1>

• Rotate image through a given angle

save <string:filename> <image:pic1>

• Save image with default options.

save2 <string:filename> <string:options> <image:pic1>

• Save image with specified options.

screen <image:pic1> <image:pic2>

• Pop the two top images, superimpose their inverted versions.

sharpness <image:pic1>

• Enhance sharpness in the top image.

show <image:pic1>

• Display and pop the top image.

size <image:pic1>

• Push the image size on the stack as (y, x).

subtract <image:pic1> <image:pic2> <int:offset> <float:scale>

• Pop the two top images, produce the scaled difference with offset.

swap

• Swap the top-of-stack item with the next one down.

thumbnail <int:xsize> <int:ysize> <image:pic1>

• Modify the top image in the stack to contain a thumbnail of itself.

transpose <string:operator> <image:pic1>

• Transpose the top image.

verbose <int:num>

• Set verbosity flag from top of stack.

4. pilconvert 工具

pilconvert 工具把图像从一种格式转换成另一种格式。输出格式使用了 -c 选项来显式指定。如果没有指定，由目标的扩展名来决定。

$ pilconvert lena.tif lena.png $ pilconvert -c JPEG lena.tif lena.tmp

5. pilfile 工具

pilfile 工具能够识别图像文件，对于能够识别的图像，它能显示图像的格式、大小和模式。

$ pilfile *.tif lena.tif: TIFF 128x128 RGB

使用-i 选项来显示info属性。使用-t 选项来显示tile descriptor (which contains information used to load the image)(译注：PIL1.1.5并无此选项)。

6. pilfont 工具

pilfont 工具将BDF 或 PCF 点阵字体文件转换成PIL的ImageFont模块能够使用的格式。

$ pilfont *.pdf

7. pilprint 工具

pilprint 工具用来打印图像到任何PostScript level 1 打印机。图像会被打印在页面的中央，上面有文件名(除掉路径和扩展名)。输出会被写到标准输出。

$ pilprint lena.tif | lpr -h

你可以使用-p 选项来直接通过 lpr 和-c 选项打印到一个彩色打印机 (如果不是彩色打印机，彩色图像会在打印前先被转换成灰度图象)。

附录

1. 软件许可证

Python Imaging Library：

• Copyright © 1997-2003 by Secret Labs AB Copyright © 1995-2003 by Fredrik Lundh

取得、使用并/或拷贝这个软件以及/或者与此相关的文档，就意味着你阅读、理解以下条款，并会照做：

只要在所有的拷贝中出现上述保全信息，并在相应的文档中保留上述版权信息和这个许可信息， Secret Labs AB或者作者的名字没有出现在未被事先明确、书面许可的广告或与软件发布相关的公共宣传信息中，则允许为了任何目的使用、拷贝和发布这个软件及其相关的文档并不需要支付费用。

By obtaining, using, and/or copying this software and/or its associated documentation, you agree that you have read, understood, and will comply with the following terms and conditions:

Permission to use, copy, modify, and distribute this software and its associated documentation for any purpose and without fee is hereby granted, provided that the above copyright notice appears in all copies, and that both that copyright notice and this permission notice appear in supporting documentation, and that the name of Secret Labs AB or the author not be used in advertising or publicity pertaining to distribution of the software without specific, written prior permission.

SECRET LABS AB AND THE AUTHOR DISCLAIMS ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL SECRET LABS AB OR THE AUTHOR BE LIABLE FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.

2. 技术支持

Patches, fixes, updates, and new utilities are welcome. If you stumble upon files that the library does not handle as expected, post a note to the Image SIG mailing list (see below). If you fix such a problem and supply a patch, you may send me the image file anyway so I don't mess things up again in later revisions.

Ideas on formats and features that should be added, sample files, and other contributions are also welcome.

For all sorts of updates, including information on commercial support and extensions to PIL, check the PIL product page, at:

• http://www.pythonware.com/products/pil

You may also find information related to PIL at http://www.pythonware.com or via the Python home page http://www.python.org

For support and general questions, send e-mail to the Python Image SIG mailing list:

• [email protected]

You can join the Image SIG by sending a mail to [email protected]. Put subscribe in the message body to automatically subscribe to the list, or help to get additional information.

Alternatively, you can use the Python mailing list, [email protected], or the newsgroup comp.lang.python.

3. 图像文件格式

Python Imaging Library 支持很广泛的光栅文件格式。PIL库差不多能识别和读取30种文件格式。输出格式的支持没有这么广泛，但最常用的交换和展示的格式还是支持的。

open 函数能够从文件的内容识别文件的格式，而不是根据文件名来识别。但是save 方法会查看文件名来决定使用哪种格式，除非文件格式被显示的指定。 格式描述 BMP

PIL 能够读写包含"1", "L", "P", 或者"RGB"类型数据的Windows 和 OS/2 的BMP文件。16色图像按照"P"模式处理。不支持行程编码（Run-length encoding）。

open 方法设置以下info 属性：

compression

• 如果文件使用行程编码，则此属性设置成"bmp_rle"。

CUR (只读)

CUR 是在Windows中用来存储指针的格式。 CUR 解码器能读取最大可能的指针。不支持指针动画指针。 DCX (只读)

DCX is a container file format for PCX files, defined by Intel. The DCX format is commonly used in fax applications. The DCX decoder files containing "1", "L", "P", or "RGB" data. Only the first image is read. EPS (只写)

The library identifies EPS files containing image data. It can also write EPS images. FLI, FLC (只读)

The library reads Autodesk FLI and FLC animations.

open 方法会设置以下info 属性：

duration

• The delay (in milliseconds) between each frame.

FPX (只读)

The library reads Kodak FlashPix files. In the current version, only the highest resolution image is read from the file, and the viewing transform is not taken into account.

Note: To enable full FlashPix support, you need to build and install the IJG JPEG library before building the Python Imaging Library. See the distribution README for details. GBR (只读)

The GBR decoder reads GIMP brush files.

open 方法会设置以下info 属性：

description

• The brush name.

GD (只读)

The library reads GD uncompressed files. Note that this file format cannot be automatically identified, so you must use the open function in the GdImageFile module to read such a file.

open 方法会设置以下info 属性：

transparency

• Transparency colour index. This key is omitted if the image is not transparent.

GIF

库能够读取 GIF87a 和GIF89a 版本的 GIF文件格式。保存文件时库会使用行程编码的 GIF87a 格式文件。 注意 GIF文件总是作为调色板模式 ("P")图像被读取的。

open 方法会设置以下info 属性：

version

• 版本（("GIF87a" 或 "GIF89a"）。

transparency

• 透明色的索引。如果图像是不透明的，这个键值被忽略。

ICO (只读)

ICO is used to store icons on Windows. The largest available icon is read. IM

IM is a format used by LabEye and other applications based on the IFUNC image processing library. The library reads and writes most uncompressed interchange versions of this format.

IM is the only format that can store all internal PIL formats. IMT (只读)

The library reads Image Tools images containing "L" data. JPEG

库能读取包含"L", "RGB", 或 "CMYK"模式数据的 JPEG, JFIF,和 Adobe JPEG文件，能够输出标准格式或者progressive的JFIF文件。

Using the draft method, you can speed things up by converting "RGB" images to "L", and resize images to 1/2, 1/4 or 1/8 of their original size while loading them. The draft method also configures the JPEG decoder to trade some quality for speed.

open 方法会设置以下info 属性：

jfif

• JFIF application marker found. If the file is not a JFIF file, this key is not present.

adobe

• Adobe application marker found. If the file is not an Adobe JPEG file, this key is not present.

progression

• Indicates that this is a progressive JPEG file.

The save method supports the following options:

quality

• The image quality, on a scale from 1 (worst) to 100 (best). The default is 75.

optimize

• If present, indicates that the encoder should make an extra pass over the image in order to select optimal encoder settings.

progression

• If present, indicates that this image should be stored as a progressive JPEG file.

Note: To enable JPEG support, you need to build and install the IJG JPEG library before building the Python Imaging Library. See the distribution README for details. MIC (只读)

The library identifies and reads Microsoft Image Composer (MIC) files. When opened, the first sprite in the file is loaded. You can use seek and tell to read other sprites from the file. MCIDAS (只读)

The library identifies and reads 8-bit McIdas area files. MPEG (只能识别)

The library identifies MPEG files. MSP

The library identifies and reads MSP files from Windows 1 and 2. The library writes uncompressed (Windows 1) versions of this format. PCD (只读)

The library reads PhotoCD files containing "RGB" data. By default, the 768x512 resolution is read. You can use the draft method to read the lower resolution versions instead, thus effectively resizing the image to 384x256 or 192x128. Higher resolutions cannot be read by the Python Imaging Library. PCX

The library reads and writes PCX files containing "1", "L", "P", or "RGB" data. PDF (只写)

The library can write PDF (Acrobat) images. Such images are written as binary PDF 1.1 files, using either JPEG or HEX encoding depending on the image mode (and whether JPEG support is available or not). PNG

The library identifies, reads, and writes PNG files containing "1", "L", "P", "RGB", or "RGBA" data. Interlaced files are currently not supported.

open 方法会设置以下info 属性：

gamma

• Gamma值，以浮点数的形式给出。

transparency

• 透明色索引。如果图像不是透明调色板图，则此键值被忽略。

save 方法支持以下选项：

optimize

• If present, instructs the PNG writer to make the output file as small as possible. This includes extra processing in order to find optimal encoder settings.

Note: To enable PNG support, you need to build and install the ZLIB compression library before building the Python Imaging Library. See the distribution README for details. PPM

库能够读写包含 "1", "L" 或 "RGB" 模式数据的 PBM, PGM 和 PPM 文件。 PSD (只读)

The library identifies and reads PSD files written by Adobe Photoshop 2.5 and 3.0. SGI (只读)

The library reads uncompressed "L" and "RGB" files. This driver is highly experimental. SUN (只读)

The library reads uncompressed "1", "P", "L" and "RGB" files. TGA (只读)

The library reads 24- and 32-bit uncompressed and run-length encoded TGA files. TIFF

The library reads and writes TIFF files containing "1", "L", "RGB", or "CMYK" data. It reads both striped and tiled images, pixel and plane interleaved multi-band images, and either uncompressed, or Packbits, LZW, or JPEG compressed images. In the current version, PIL always writes uncompressed TIFF files.

open 方法会设置以下info 属性：

compression

• 压缩模式。

In addition, the tag attribute contains a dictionary of decoded TIFF fields. Values are stored as either strings or tuples. Note that only short, long and ASCII tags are correctly unpacked by this release. XBM

库能够读写 X bitmap 文件 (模式"1")。 XPM (只读)

库能够读取256色或者更少颜色的X pixmap 文件 (模式"P")。

open 方法会设置以下info 属性：

transparency

• 透明色索引。如果图像不是透明的，则此键值被忽略。

文件扩展名

Python Imaging Library 会把一些文件扩展名同特定的文件类型相关联。open 函数根据文件的内容来识别文件，而不是名字。但是save 方法会查看文件名来决定使用哪种格式，除非显示的指定格式。 类型 扩展名 BMP ".bmp", ".dib" CUR ".cur" DCX ".dcx" EPS ".eps", ".ps" FLI ".fli", ".flc" FPX ".fpx" GBR ".gbr" GD ".gd" GIF ".gif" ICO ".ico" IM ".im" JPEG ".jpg", ".jpe", ".jpeg" MIC ".mic" MSP ".msp" PCD ".pcd" PCX ".pcx" PDF ".pdf" PNG ".png" PPM ".pbm", ".pgm", ".ppm" PSD ".psd" SGI ".bw", ".rgb", ".cmyk" SUN ".ras" TGA ".tga" TIFF ".tif", ".tiff" XBM ".xbm" XPM ".xpm"

要注意的是，PIL库并不支持所有这些格式的保存操作。

4. 编写自己的文件解码器

The Python Imaging Library uses a plug-in model which allows you to add your own decoders to the library, without any changes to the library itself. Such plug-ins have names like XxxImagePlugin.py, where Xxx is a unique format name (usually an abbreviation).

A decoder plug-in should contain a decoder class, based on the ImageFile base class defined in the module with the same name. This class should provide an _open method, which reads the file header and sets up at least the mode and size attributes. To be able to load the file, the method must also create a list of tile descriptors. The class must be explicitly registered, via a call to the Image module.

For performance reasons, it is important that the _open method quickly rejects files that do not have the appropriate contents. Example

The following plug-in supports a simple format, which has a 128-byte header consisting of the words "SPAM" followed by the width, height, and pixel size in bits. The header fields are separated by spaces. The image data follows directly after the header, and can be either bi-level, greyscale, or 24-bit true colour. File: SpamImagePlugin.py

   1 import Image, ImageFile
   2 import string
   3 
   4 class SpamImageFile(ImageFile.ImageFile):
   5 
   6     format = "SPAM"
   7     format_description = "Spam raster image"
   8 
   9     def _open(self):
  10 
  11         # check header
  12         header = self.fp.read(128)
  13         if header[:4] != "SPAM":
  14             raise SyntaxError, "not a SPAM file"
  15 
  16         header = string.split(header)
  17 
  18         # size in pixels (width, height)
  19         self.size = int(header[1]), int(header[2])
  20 
  21         # mode setting
  22         bits = int(header[3])
  23         if bits == 1:
  24             self.mode = "1"
  25         elif bits == 8:
  26             self.mode = "L"
  27         elif bits == 24:
  28             self.mode = "RGB"
  29         else:
  30             raise SyntaxError, "unknown number of bits"
  31 
  32         # data descriptor
  33         self.tile = [
  34             ("raw", (0, 0) + self.size, 128, (self.mode, 0, 1))
  35         ]
  36 
  37 Image.register_open("SPAM", SpamImageFile)
  38         
  39 Image.register_extension("SPAM", ".spam")
  40 Image.register_extension("SPAM", ".spa") # dos version

The format handler must always set the size and mode attributes. If these are not set, the file cannot be opened. To simplify the decoder, the calling code considers exceptions like SyntaxError, KeyError, and IndexError, as a failure to identify the file.

Note that the decoder must be explicitly registered using the register_open function in the Image module. Although not required, it is also a good idea to register any extensions used by this format. The Tile Attribute

To be able to read the file as well as just identifying it, the tile attribute must also be set. This attribute consists of a list of tile descriptors, where each descriptor specifies how data should be loaded to a given region in the image. In most cases, only a single descriptor is used, covering the full image.

The tile descriptor is a 4-tuple with the following contents:

• (decoder, region, offset, parameters)

The fields are used as follows:

decoder

• Specifies which decoder to use. The "raw" decoder used here supports uncompressed data, in a variety of pixel formats. For more information on this decoder, see the description below.

region

• A 4-tuple specifying where to store data in the image.

offset

• Byte offset from the beginning of the file to image data.

parameters

• Parameters to the decoder. The contents of this field depends on the decoder specified by the first field in the tile descriptor tuple. If the decoder doesn't need any parameters, use None for this field.

Note that the tile attribute contains a list of tile descriptors, not just a single descriptor. The Raw Decoder

The raw decoder is used to read uncompressed data from an image file. It can be used with most uncompressed file formats, such as PPM, BMP, uncompressed TIFF, and many others. To use the raw decoder with the fromstring function, use the following syntax:

• image = fromstring(
  • mode, size, data, "raw", raw mode, stride, orientation )

When used in a tile descriptor, the parameter field should look like:

• (raw mode, stride, orientation)

The fields are used as follows:

raw mode

• The pixel layout used in the file, and is used to properly convert data to PIL's internal layout. For a summary of the available formats, see the table below.

stride

• The distance in bytes between two consecutive lines in the image. If 0, the image is assumed to be packed (no padding between lines). If omitted, the stride defaults to 0.

orientation

• Whether the first line in the image is the top line on the screen (1), or the bottom line (-1). If omitted, the orientation defaults to 1.

The raw mode field is used to determine how the data should be unpacked to match PIL's internal pixel layout. PIL supports a large set of raw modes; for a complete list, see the table in the Unpack.c module. The following table describes some commonly used raw modes: mode description "1" 1-bit bilevel, stored with the leftmost pixel in the most significant bit. 0 means black, 1 means white. "1;I" 1-bit inverted bilevel, stored with the leftmost pixel in the most significant bit. 0 means white, 1 means black. "1;R" 1-bit reversed bilevel, stored with the leftmost pixel in the least significant bit. 0 means black, 1 means white. "L" 8-bit greyscale. 0 means black, 255 means white. "L;I" 8-bit inverted greyscale. 0 means white, 255 means black. "P" 8-bit palette-mapped image. "RGB" 24-bit true colour, stored as (red, green, blue). "BGR" 24-bit true colour, stored as (blue, green, red). "RGBX" 24-bit true colour, stored as (blue, green, red, pad). "RGB;L" 24-bit true colour, line interleaved (first all red pixels, the all green pixels, finally all blue pixels).

Note that for the most common cases, the raw mode is simply the same as the mode.

The Python Imaging Library supports many other decoders, including JPEG, PNG, and PackBits. For details, see the decode.c source file, and the standard plug-in implementations provided with the library. Decoding Floating Point Data

PIL provides some special mechanisms to allow you to load a wide variety of formats into a mode "F" (floating point) image memory.

You can use the "raw" decoder to read images where data is packed in any standard machine data type, using one of the following raw modes: mode description "F" 32-bit native floating point. "F;8" 8-bit unsigned integer. "F;8S" 8-bit signed integer. "F;16" 16-bit little endian unsigned integer. "F;16S" 16-bit little endian signed integer. "F;16B" 16-bit big endian unsigned integer. "F;16BS" 16-bit big endian signed integer. "F;16N" 16-bit native unsigned integer. "F;16NS" 16-bit native signed integer. "F;32" 32-bit little endian unsigned integer. "F;32S" 32-bit little endian signed integer. "F;32B" 32-bit big endian unsigned integer. "F;32BS" 32-bit big endian signed integer. "F;32N" 32-bit native unsigned integer. "F;32NS" 32-bit native signed integer. "F;32F" 32-bit little endian floating point. "F;32BF" 32-bit big endian floating point. "F;32NF" 32-bit native floating point. "F;64F" 64-bit little endian floating point. "F;64BF" 64-bit big endian floating point. "F;64NF" 64-bit native floating point. The Bit Decoder

If the raw decoder cannot handle your format, PIL also provides a special "bit" decoder that can be used to read various packed formats into a floating point image memory.

To use the bit decoder with the fromstring function, use the following syntax:

• image = fromstring(
  • mode, size, data, "bit", bits, pad, fill, sign, orientation )

When used in a tile descriptor, the parameter field should look like:

• (bits, pad, fill, sign, orientation)

The fields are used as follows:

bits

• Number of bits per pixel (2-32). No default.

pad

• Padding between lines, in bits. This is either 0 if there is no padding, or 8 if lines are padded to full bytes. If omitted, the pad value defaults to 8.

fill

• Controls how data are added to, and stored from, the decoder bit buffer.

fill=0

• Add bytes to the msb end of the decoder buffer; store pixels from the msb end.

fill=1

• Add bytes to the lsb end of the decoder buffer; store pixels from the msb end.

fill=2

• Add bytes to the msb end of the decoder buffer; store pixels from the lsb end.

fill=3

• Add bytes to the lsb end of the decoder buffer; store pixels from the lsb end. If omitted, the fill order defaults to 0.

sign

• If non-zero, bit fields are sign extended. If zero or omitted, bit fields are unsigned.

orientation

• Whether the first line in the image is the top line on the screen (1), or the bottom line (-1). If omitted, the orientation defaults to 1.

译注：中英文术语对照表

• affine 仿射
• alpha 透明度
• band 通道
• class 类
• decoder 解码器
• dictionary 字典
• draft 草稿
• enhance 增强
• exception 异常
• filter 滤波器
• function 函数
• histogram 直方图
• image 图像
• import 导入
• iterate 迭代
• iterator 跌代子
• list 列表
• mask 掩模
• method 方法
• mode 模式
• object 对象
• operator 运算符
• palette 调色板
• pixel 像素
• raster 光栅
• rotate 旋转
• sequence 序列
• size 大小
• socket 套接字
• thumbnails 缩略图
• threshold 阈值
• tuple 元组
• zoom 缩放