The
SSIM Index for Image Quality Assessment
by Zhou Wang
The
Structural SIMilarity (SSIM) index is a novel method for measuring the
similarity between two images. The SSIM index can be viewed as a quality measure
of one of the images being compared, provided the other image is regarded as of
perfect quality. It is an improved version of the universal
image quality index we proposed before. A description of the method can be
found here.
More details are given in the following paper:
Z. Wang, A. C. Bovik,
H. R. Sheikh, and
E. P. Simoncelli, "Image quality
assessment: From error measurement to structural similarity,"
IEEE Transactions on Image
Processing, accepted, May
2003.
A
Matlab implementation
of the SSIM index (ssim_index.m) is available here. You can download it for
free, change it as you like and use it anywhere, but please refer to its
original source (cite the above paper and this web page). Before using the code,
please go through the suggested
usage and demo
tests below to get an idea on how to use it and how it
works.
This is
the single scale version of the SSIM indexing measure, which is most effective
if used at the appropriate scale. The precisely “right” scale depends on both
the image resolution and the viewing distance and is usually difficult to be
obtained. In practice, we suggest
to use the following empirical formula to determine the scale for images viewed
from a typical distance (say 3~5 times of the image height): 1) Let F = max(1,
round(N/256)), where N is the number of pixels in image height; 2) Downsample
the image by a factor of F, and then apply the ssim_index.m program. For
example, for an 512 by 512 image, F = max(1, round(512/256)) = 2, so the image
should be downsampled by a factor of 2 before applying ssim_index.m (type “help
ssim_index “ to get more information about how to use it).
Multi-scale
method (with appropriate setup) can further improve the SSIM measurement. More
details are available in the paper below
Z. Wang, E. P. Simoncelli and A. C.
Bovik, "Multi-scale
structural similarity for image quality assessment," Invited Paper,
IEEE Asilomar Conference on
Signals, Systems and Computers, Nov. 2003.
Test
on JPEG/JPEG2000 Image Database
The SSIM
indexing algorithm has been tested on a database, which includes
344 JPEG and JPEG2000 compressed images. The database is created and
available for free download at the Lab for
Image and Video Engineering (LIVE) at the University of Texas at Austin. Two sample
images (cropped for display purpose) are shown below. Note that quantization in
JPEG and JPEG2000 algorithms often results in smooth representations of fine
detail regions (e.g., the tiles in the upper image and the trees in the lower
image). Compared with other types of regions, these regions may not be worse in
terms of pointwise difference measures (as shown in the root squared error map).
However, since the structural information of the image details are nearly
completely lost, they exhibit poorer visual quality. Close piece-by-piece
comparison of the SSIM index and the root squared error maps, we observe that
the SSIM index is more consistent with perceived quality measurement. Note: in
both distortion/quality maps, brighter means better
quality.
|
Original
Images |
JPEG/JPEG2000 Compressed
Images |
Root Squared Error
Map |
SSIM Index
Map |
|
|
|
|
|
|
|
|
|
|
The scatter plots of the subjective measurement (mean opinion score, MOS) versus the objective predictions (PSNR and MSSIM) are shown below, where each point represent one test image. Clearly, MSSIM is much better in predicting the perceived image quality.
|
|
|
|
PSNR vs. MOS |
MSSIM vs.
MOS |
Back
to Zhou Wang's
Research Work
Back
to Zhou Wang's
Homepage
Last
updated Jun. 2, 2003
