Realtime human vision system behavioral modeling

摘要

A method of realtime human vision system modeling to produce a measure of impairment of a test image signal derived from a reference image signal processes the two signals in respective channels. The signals are converted to luminance image signals and low-pass filtered in two dimensions. The processed image signals are then segmented and block means values are obtained which are subtracted from the pixels in the corresponding processed image signals. Noise is injected into the segmented processed image signals and a variance is calculated for the reference segmented processed image signal and also for the difference between the segmented processed image signals. The variance of difference segmented processed image signal is normalized by the variance for the reference segmented processed image signal, and the Nth root of the result is taken as the measure of visible impairment of the test image signal. The measure of visible impairment may be converted into appropriate units, such as JND, MOS, etc.