sp.ssim

Module providing functionality to implement Structural Similarity Image Quality Assessment. Based on original paper by Z. Whang “Image Quality Assessment: From Error Visibility to Structural Similarity” IEEE Transactions on Image Processing Vol. 13. No. 4. April 2004.

Example Usage

ssim

import pylab
import numpy
from sp import ssim
from PIL import Image
einstein = numpy.asarray(Image.open('./imgs/einstein.tif'))
meanshift = numpy.asarray(Image.open('./imgs/meanshift.tif'))
contrast = numpy.asarray(Image.open('./imgs/contrast.tif'))
impulse = numpy.asarray(Image.open('./imgs/impulse.tif'))
blur = numpy.asarray(Image.open('./imgs/blur.tif'))
jpg = numpy.asarray(Image.open('./imgs/jpg.tif'))
einstein_ssim = ssim.ssim(einstein, einstein)
meanshift_ssim = ssim.ssim(einstein, meanshift)
contrast_ssim = ssim.ssim(einstein, contrast)
impulse_ssim = ssim.ssim(einstein, impulse)
blur_ssim = ssim.ssim(einstein, blur)
jpg_ssim = ssim.ssim(einstein, jpg)
pylab.figure()
pylab.subplot(2, 3, 1)
pylab.title('Original\n SSIM %.3f' % einstein_ssim.mean())
pylab.imshow(einstein, cmap=pylab.gray())
pylab.subplot(2, 3, 2)
pylab.title('Mean-Shifted\n SSIM %.3f' % meanshift_ssim.mean())
pylab.imshow(meanshift, cmap=pylab.gray())
pylab.subplot(2, 3, 3)
pylab.title('Contrast-Adjusted\n SSIM %.3f' % contrast_ssim.mean())
pylab.imshow(contrast, cmap=pylab.gray())
pylab.subplot(2, 3, 4)
pylab.title('Impulse-Noise\n SSIM %.3f' % impulse_ssim.mean())
pylab.imshow(impulse, cmap=pylab.gray())
pylab.subplot(2, 3, 5)
pylab.title('Blur\n SSIM %.3f' % blur_ssim.mean())
pylab.imshow(blur, cmap=pylab.gray())
pylab.subplot(2, 3, 6)
pylab.title('JPG\n SSIM %.3f' % jpg_ssim.mean())
pylab.imshow(jpg, cmap=pylab.gray())
pylab.show()

(Source code, png, hires.png, pdf)

import numpy
import pylab
from sp import ssim
from PIL import Image
original = numpy.asarray(Image.open('./imgs/building.tif'))
jpg = numpy.asarray(Image.open('./imgs/building_jpg.tif'))
ssim_map = ssim.ssim(original, jpg)
abs_map = numpy.abs(original.astype('float64')-jpg.astype('float64'))
pylab.figure()
pylab.subplot(2, 2, 1)
pylab.title('Original')
pylab.imshow(original, cmap=pylab.gray())
pylab.subplot(2, 2, 2)
pylab.title('JPEG compressed')
pylab.imshow(jpg, cmap=pylab.gray())
pylab.subplot(2, 2, 3)
pylab.title('Absolute error map')
pylab.imshow(abs_map.astype('uint8'), cmap=pylab.gray())
pylab.subplot(2, 2, 4)
pylab.title('SSIM index map')
pylab.imshow(ssim_map, cmap=pylab.gray())
pylab.show()

(Source code, png, hires.png, pdf)

msssim

import numpy
import pylab
from sp import ssim
from PIL import Image
einstein = numpy.asarray(Image.open('./imgs/einstein.tif'))
impulse = numpy.asarray(Image.open('./imgs/impulse.tif'))
ms_ssim = ssim.msssim(original, impulse)
pylab.figure()
pylab.subplot(1, 2, 1)
pylab.title('Original')
pylab.imshow(original, cmap=pylab.gray())
pylab.subplot(1, 2, 2)
pylab.title('Impulse-Noise\n %.3f' % ms_ssim)
pylab.imshow(impulse, cmap=pylab.gray())
pylab.show()

(Source code, png, hires.png, pdf)

See the individual methods below for further details.

Functions

sp.ssim.ssim(img1, img2, cs_map=False)

Return the Structural Similarity Map corresponding to input images img1 and img2 (images are assumed to be uint8)

This function attempts to mimic precisely the functionality of ssim.m a MATLAB provided by the author’s of SSIM https://ece.uwaterloo.ca/~z70wang/research/ssim/ssim_index.m

sp.ssim.msssim(img1, img2)

This function implements Multi-Scale Structural Similarity (MSSSIM) Image Quality Assessment according to Z. Wang’s “Multi-scale structural similarity for image quality assessment” Invited Paper, IEEE Asilomar Conference on Signals, Systems and Computers, Nov. 2003

Author’s MATLAB implementation:- http://www.cns.nyu.edu/~lcv/ssim/msssim.zip