An analytical model of the perceptual threshold function for multichannel image compression

The human observer is often the final judge of the quality of compressed images. One way to design a compression system that attempts to reduce or eliminate subjective distortions in the coded images is to incorporate a perceptual threshold function model into the compression system. The perceptual threshold function describes the amount of quantization error that can be introduced into a particular component of the image without introducing any visual distortions. This paper describes an analytical approach for the determination of the perceptual threshold values for use in an arbitrary multichannel image compression system. Experimental results obtained from a compression system that incorporates the perceptual threshold function are also included in the paper.