Application of wavelets to image coding
Wayne Lawton · 2003
Summary form only given, as follows. Multilevel image coding methods for compression and related image processing applications are described. These methods first transform an image, using the specific orthonormal bases of compactly supported wavelets invented by Daubechies (1988), to obtain a multiresolution representation of the image in terms of its transform coefficients. Subsequent multilevel processing is performed directly on these coefficients using techniques such as vector quantization, correlation, and prediction to achieve typical compression, stereoscopic matching, enhancement, or pattern recognition goals. Computational complexity issues and numerical results are discussed.