Adaptive image processing a computational intelligence perspective - Perceptually Motivated Image Restoration

In Chapter 3, we mentioned the problems caused by considering the image to be an ensemble of stationary processes. Any restoration process based on this concept can only ever produce sub-optimal results. However, there is another consideration. When a restoration algorithm is designed with the aim of creating an image which will be more pleasing to the human eye, we must incorporate some kind of model of the human visual system. At the basis of most restoration algorithms is some form of image error measure which is being minimized. The most common method to compare the similarity of two images is to compute their mean square error (MSE). However, the MSE relates to the power of the error signal and has little relationship to human visual perception. An important drawback to the MSE and any cost function which attempts to use the MSE to restore a degraded image is that the MSE treats the image as a stationary