Pillow/PIL/ImageOps.py
Brian Crowell a7e3b2e47b py3k: The big push
There are two main issues fixed with this commit:

* bytes vs. str: All file, image, and palette data are now handled as
  bytes. A new _binary module consolidates the hacks needed to do this
  across Python versions. tostring/fromstring methods have been renamed to
  tobytes/frombytes, but the Python 2.6/2.7 versions alias them to the old
  names for compatibility. Users should move to tobytes/frombytes.

  One other potentially-breaking change is that text data in image files
  (such as tags, comments) are now explicitly handled with a specific
  character encoding in mind. This works well with the Unicode str in
  Python 3, but may trip up old code expecting a straight byte-for-byte
  translation to a Python string. This also required a change to Gohlke's
  tags tests (in Tests/test_file_png.py) to expect Unicode strings from
  the code.

* True div vs. floor div: Many division operations used the "/" operator
  to do floor division, which is now the "//" operator in Python 3. These
  were fixed.

As of this commit, on the first pass, I have one failing test (improper
handling of a slice object in a C module, test_imagepath.py) in Python 3,
and three that that I haven't tried running yet (test_imagegl,
test_imagegrab, and test_imageqt). I also haven't tested anything on
Windows. All but the three skipped tests run flawlessly against Pythons
2.6 and 2.7.
2013-01-10 08:46:56 -06:00

441 lines
13 KiB
Python

#
# The Python Imaging Library.
# $Id$
#
# standard image operations
#
# History:
# 2001-10-20 fl Created
# 2001-10-23 fl Added autocontrast operator
# 2001-12-18 fl Added Kevin's fit operator
# 2004-03-14 fl Fixed potential division by zero in equalize
# 2005-05-05 fl Fixed equalize for low number of values
#
# Copyright (c) 2001-2004 by Secret Labs AB
# Copyright (c) 2001-2004 by Fredrik Lundh
#
# See the README file for information on usage and redistribution.
#
from . import Image
import operator
from functools import reduce
##
# (New in 1.1.3) The <b>ImageOps</b> 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.
#
# @since 1.1.3
##
#
# helpers
def _border(border):
if isinstance(border, tuple):
if len(border) == 2:
left, top = right, bottom = border
elif len(border) == 4:
left, top, right, bottom = border
else:
left = top = right = bottom = border
return left, top, right, bottom
def _color(color, mode):
if Image.isStringType(color):
from . import ImageColor
color = ImageColor.getcolor(color, mode)
return color
def _lut(image, lut):
if image.mode == "P":
# FIXME: apply to lookup table, not image data
raise NotImplementedError("mode P support coming soon")
elif image.mode in ("L", "RGB"):
if image.mode == "RGB" and len(lut) == 256:
lut = lut + lut + lut
return image.point(lut)
else:
raise IOError("not supported for this image mode")
#
# actions
##
# Maximize (normalize) image contrast. This function calculates a
# histogram of the input image, removes <i>cutoff</i> 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).
#
# @param image The image to process.
# @param cutoff How many percent to cut off from the histogram.
# @param ignore The background pixel value (use None for no background).
# @return An image.
def autocontrast(image, cutoff=0, ignore=None):
"Maximize image contrast, based on histogram"
histogram = image.histogram()
lut = []
for layer in range(0, len(histogram), 256):
h = histogram[layer:layer+256]
if ignore is not None:
# get rid of outliers
try:
h[ignore] = 0
except TypeError:
# assume sequence
for ix in ignore:
h[ix] = 0
if cutoff:
# cut off pixels from both ends of the histogram
# get number of pixels
n = 0
for ix in range(256):
n = n + h[ix]
# remove cutoff% pixels from the low end
cut = n * cutoff // 100
for lo in range(256):
if cut > h[lo]:
cut = cut - h[lo]
h[lo] = 0
else:
h[lo] = h[lo] - cut
cut = 0
if cut <= 0:
break
# remove cutoff% samples from the hi end
cut = n * cutoff // 100
for hi in range(255, -1, -1):
if cut > h[hi]:
cut = cut - h[hi]
h[hi] = 0
else:
h[hi] = h[hi] - cut
cut = 0
if cut <= 0:
break
# find lowest/highest samples after preprocessing
for lo in range(256):
if h[lo]:
break
for hi in range(255, -1, -1):
if h[hi]:
break
if hi <= lo:
# don't bother
lut.extend(list(range(256)))
else:
scale = 255.0 / (hi - lo)
offset = -lo * scale
for ix in range(256):
ix = int(ix * scale + offset)
if ix < 0:
ix = 0
elif ix > 255:
ix = 255
lut.append(ix)
return _lut(image, lut)
##
# Colorize grayscale image. The <i>black</i> and <i>white</i>
# arguments should be RGB tuples; 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.
#
# @param image The image to colourize.
# @param black The colour to use for black input pixels.
# @param white The colour to use for white input pixels.
# @return An image.
def colorize(image, black, white):
"Colorize a grayscale image"
assert image.mode == "L"
black = _color(black, "RGB")
white = _color(white, "RGB")
red = []; green = []; blue = []
for i in range(256):
red.append(black[0]+i*(white[0]-black[0])//255)
green.append(black[1]+i*(white[1]-black[1])//255)
blue.append(black[2]+i*(white[2]-black[2])//255)
image = image.convert("RGB")
return _lut(image, red + green + blue)
##
# Remove border from image. The same amount of pixels are removed
# from all four sides. This function works on all image modes.
#
# @param image The image to crop.
# @param border The number of pixels to remove.
# @return An image.
# @see Image#Image.crop
def crop(image, border=0):
"Crop border off image"
left, top, right, bottom = _border(border)
return image.crop(
(left, top, image.size[0]-right, image.size[1]-bottom)
)
##
# Deform the image.
#
# @param image The image to deform.
# @param deformer A deformer object. Any object that implements a
# <b>getmesh</b> method can be used.
# @param resample What resampling filter to use.
# @return An image.
def deform(image, deformer, resample=Image.BILINEAR):
"Deform image using the given deformer"
return image.transform(
image.size, Image.MESH, deformer.getmesh(image), resample
)
##
# 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.
#
# @param image The image to equalize.
# @param mask An optional mask. If given, only the pixels selected by
# the mask are included in the analysis.
# @return An image.
def equalize(image, mask=None):
"Equalize image histogram"
if image.mode == "P":
image = image.convert("RGB")
h = image.histogram(mask)
lut = []
for b in range(0, len(h), 256):
histo = [_f for _f in h[b:b+256] if _f]
if len(histo) <= 1:
lut.extend(list(range(256)))
else:
step = (reduce(operator.add, histo) - histo[-1]) // 255
if not step:
lut.extend(list(range(256)))
else:
n = step // 2
for i in range(256):
lut.append(n // step)
n = n + h[i+b]
return _lut(image, lut)
##
# Add border to the image
#
# @param image The image to expand.
# @param border Border width, in pixels.
# @param fill Pixel fill value (a colour value). Default is 0 (black).
# @return An image.
def expand(image, border=0, fill=0):
"Add border to image"
left, top, right, bottom = _border(border)
width = left + image.size[0] + right
height = top + image.size[1] + bottom
out = Image.new(image.mode, (width, height), _color(fill, image.mode))
out.paste(image, (left, top))
return out
##
# Returns a sized and cropped version of the image, cropped to the
# requested aspect ratio and size.
# <p>
# The <b>fit</b> function was contributed by Kevin Cazabon.
#
# @param size The requested output size in pixels, given as a
# (width, height) tuple.
# @param method What resampling method to use. Default is Image.NEAREST.
# @param bleed Remove a border around the outside of 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).
# @param centering Control the cropping position. Use (0.5, 0.5) for
# center cropping (e.g. if cropping the width, take 50% off of the
# left side, and therefore 50% off the right side). (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).
# @return An image.
def fit(image, size, method=Image.NEAREST, bleed=0.0, centering=(0.5, 0.5)):
"""
This method returns a sized and cropped version of the image,
cropped to the aspect ratio and size that you request.
"""
# by Kevin Cazabon, Feb 17/2000
# kevin@cazabon.com
# http://www.cazabon.com
# ensure inputs are valid
if not isinstance(centering, list):
centering = [centering[0], centering[1]]
if centering[0] > 1.0 or centering[0] < 0.0:
centering [0] = 0.50
if centering[1] > 1.0 or centering[1] < 0.0:
centering[1] = 0.50
if bleed > 0.49999 or bleed < 0.0:
bleed = 0.0
# calculate the area to use for resizing and cropping, subtracting
# the 'bleed' around the edges
# number of pixels to trim off on Top and Bottom, Left and Right
bleedPixels = (
int((float(bleed) * float(image.size[0])) + 0.5),
int((float(bleed) * float(image.size[1])) + 0.5)
)
liveArea = (
bleedPixels[0], bleedPixels[1], image.size[0] - bleedPixels[0] - 1,
image.size[1] - bleedPixels[1] - 1
)
liveSize = (liveArea[2] - liveArea[0], liveArea[3] - liveArea[1])
# calculate the aspect ratio of the liveArea
liveAreaAspectRatio = float(liveSize[0])/float(liveSize[1])
# calculate the aspect ratio of the output image
aspectRatio = float(size[0]) / float(size[1])
# figure out if the sides or top/bottom will be cropped off
if liveAreaAspectRatio >= aspectRatio:
# liveArea is wider than what's needed, crop the sides
cropWidth = int((aspectRatio * float(liveSize[1])) + 0.5)
cropHeight = liveSize[1]
else:
# liveArea is taller than what's needed, crop the top and bottom
cropWidth = liveSize[0]
cropHeight = int((float(liveSize[0])/aspectRatio) + 0.5)
# make the crop
leftSide = int(liveArea[0] + (float(liveSize[0]-cropWidth) * centering[0]))
if leftSide < 0:
leftSide = 0
topSide = int(liveArea[1] + (float(liveSize[1]-cropHeight) * centering[1]))
if topSide < 0:
topSide = 0
out = image.crop(
(leftSide, topSide, leftSide + cropWidth, topSide + cropHeight)
)
# resize the image and return it
return out.resize(size, method)
##
# Flip the image vertically (top to bottom).
#
# @param image The image to flip.
# @return An image.
def flip(image):
"Flip image vertically"
return image.transpose(Image.FLIP_TOP_BOTTOM)
##
# Convert the image to grayscale.
#
# @param image The image to convert.
# @return An image.
def grayscale(image):
"Convert to grayscale"
return image.convert("L")
##
# Invert (negate) the image.
#
# @param image The image to invert.
# @return An image.
def invert(image):
"Invert image (negate)"
lut = []
for i in range(256):
lut.append(255-i)
return _lut(image, lut)
##
# Flip image horizontally (left to right).
#
# @param image The image to mirror.
# @return An image.
def mirror(image):
"Flip image horizontally"
return image.transpose(Image.FLIP_LEFT_RIGHT)
##
# Reduce the number of bits for each colour channel.
#
# @param image The image to posterize.
# @param bits The number of bits to keep for each channel (1-8).
# @return An image.
def posterize(image, bits):
"Reduce the number of bits per color channel"
lut = []
mask = ~(2**(8-bits)-1)
for i in range(256):
lut.append(i & mask)
return _lut(image, lut)
##
# Invert all pixel values above a threshold.
#
# @param image The image to posterize.
# @param threshold All pixels above this greyscale level are inverted.
# @return An image.
def solarize(image, threshold=128):
"Invert all values above threshold"
lut = []
for i in range(256):
if i < threshold:
lut.append(i)
else:
lut.append(255-i)
return _lut(image, lut)
# --------------------------------------------------------------------
# PIL USM components, from Kevin Cazabon.
def gaussian_blur(im, radius=None):
""" PIL_usm.gblur(im, [radius])"""
if radius is None:
radius = 5.0
im.load()
return im.im.gaussian_blur(radius)
gblur = gaussian_blur
def unsharp_mask(im, radius=None, percent=None, threshold=None):
""" PIL_usm.usm(im, [radius, percent, threshold])"""
if radius is None:
radius = 5.0
if percent is None:
percent = 150
if threshold is None:
threshold = 3
im.load()
return im.im.unsharp_mask(radius, percent, threshold)
usm = unsharp_mask