Wavelet-based signal processing and optics
Yao Li · Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE · 1994
Recent advances in mathematical theory related to the representation of signals and images by some limited-extent-wave functions have generated substantial interest in applications of waveforms that are localized in both time (space) and frequency (spatial frequency) domains to signal and image processing, pattern recognition, data compression, and neural networks. As a particularly useful model, the so-called wavelet transforms provide a variety of multiresolution signal (image) time-frequency or space-frequency decomposition tools. Digital parallel implementations of the wavelet transforms are computationally intensive both because of the nature of the coordinate dilation and erosion of these transforms and because of the large quantity of convolution/correlation operations accompanied. This paper is intended to outline some of the most useful properties of the wavelet transforms, their similarities, and differences to other known joint representations. The links between the wavelet transforms and optics also are discussed.