Structural Similarity Metrics in Image Analysis and Compression
Electrical Engineering and Computer Science Department
Traditional image similarity metrics compare two images on a point-by-point basis. While such metrics can successfully account for perceptual effects, such as contrast and luminance masking, they are quite sensitive to spatial and intensity shifts, as well as contrast and scale changes. On the other hand, Structural SIMilarity (SSIM) metrics allow substantial point-by-point variations by relying on comparisons of region statistics. As the name suggests, the goal is to base image similarity on "structural" information. As such, they can potentially be more effective in quantifying suprathreshold compression artifacts as well as texture similarity for image analysis applications.
We compare specific SSIM implementations both in the image space and the wavelet domain, and evaluate their effectiveness in the context of realistic distortions that arise from compression and error concealment in video transmission applications. We also evaluate the performance of SSIM metrics in the context of measuring texture similarity, and propose new metrics that incorporate the best features of SSIM and eliminate the most serious drawbacks. We show that the proposed new texture similarity metrics outperform existing SSIM implementations, as well as traditional metrics.
Finally, we discuss ongoing research on a new "structurally lossless" approach for visual data compression that allows significant differences between the original and decoded images, which may be perceptible when they are viewed side-by-side, but do not affect the overall quality of the image.
Thrasyvoulos (Thrasos) Pappas received the S.B., S.M., and Ph.D. degrees in electrical engineering and computer science from MIT in 1979, 1982, and 1987, respectively. From 1987 until 1999, he was a Member of the Technical Staff at Bell Laboratories, Murray Hill, NJ. In 1999 he joined the Department of Electrical and Computer Engineering at Northwestern University. His research interests are in image and video quality and compression, perceptual models for image processing, model-based halftoning, image and video analysis, and multimedia signal processing.
Dr. Pappas has served as an elected member of the Board of Governors of the Signal Processing Society of IEEE (2004-2007), chair of the IEEE Image and Multidimensional Signal Processing Technical Committee, associate editor of the IEEE Transactions on Image Processing, and technical program co-chair of ICIP-01 and the Symposium on Information Processing in Sensor Networks (IPSN-04). Dr. Pappas is Fellow of IEEE and SPIE. Since 1997 he has been co-chair of the SPIE/IS&T Conference on Human Vision and Electronic Imaging. He has also served
as co-chair of the 2005 SPIE/IS&T Electronic Imaging Symposium.