SIGNAL COMPRESSION BASED ON MODELS OF HUMAN PERCEPTION

The problem of signal compression is to achieve a low bit rate in the digital representation of an input signal with minimum perceived loss of signal quality. In compressing signals such as speech, audio, image, and video, the ultimate criterion of signal quality is usually that judged or measured by the human receiver. As we seek lower bit rates in the digital representations of these signals, it is imperative that we design the compression (or coding) algorithm to minimize perceptually meaningful measures of signal distortion, rather than more traditional and tractable criteria such as the mean squared difference between the waveforms at the input and output of the coding system. This paper develops the notion of perceptual coding based on the concept of distortion masking by the signal being compressed, and describes how the field has progressed as a result of advances in classical coding theory, modeling of human perception, and digital signal processing. We propose that fundamental limits in the science can be expressed by the semi-quantitative concepts of perceptual entropy and the perceptual distortion-rate function, and we examine current compression technology with respect to that framework. We conclude with a summary of future challenges and research directions.