Chapter 2 discussed entropy coding algorithms possessing the desirable feature that the data obtained from decompression are identical to the original data. That is, the compression algorithms described in that chapter are lossless. As mentioned in Chapter 1, some applications (such as certain medical imaging systems) require lossless compression, while other applications may tolerate some amount of distortion in the decompressed data in return for a smaller compressed representation. Quantization is the element of lossy compression systems responsible for reducing the precision of data in order to make them more compressible. In most lossy compression systems, it is the only source of distortion.
KeywordsVector Quantization Viterbi Algorithm Probability Density Function Scalar Quantization Voronoi Region
Unable to display preview. Download preview PDF.