Chapter 8.3 in Handbook of Image and Video Processing, second edition, Al Bovik, ed., Academic Press, May 2005
Structural Approaches to Image Quality Assessment
1Laboratory for Computational Vision,
2Laboratory for Image and Video Engineering
Abstract: This chapter introduces the basic ideas and algorithms of structural approaches for image quality assessment. We describe the concepts, the structural similarity index algorithm, as well as the image synthesis-based performance evaluation algorithm in the image space. We demonstrate that image distortions along different directions in the image space have different perceptual meanings. The structural approaches attempt to separate the directions associated with structural distortions from those with non-structured distortions. This separation gives a new coordinate system in the image space, which is not fixed as in traditional image decomposition frameworks (e.g., Fourier and wavelet types of transforms), but adapted to the underlying image structures.