A Study on Perfect Reconstructive Subband Coding

With the evolution toward a broadband digital network, nationwide image-signal transmission systems will be constructed in order to offer high-quality and reliable services to customers. The largest users of Image transmission systems will be broadcasting stations. For contribution of Image signals inside a studio and between studios or broadcasting stations, the capability for multiple encoding and decoding of the signals without visible degradation is required. The Ideal system would be one where no quantization-noise Is detected. Subband coding is presented which has the possibility of distortion-free encoding. Input signals are decomposed Into four bands in the horizontal and vertical directions. Each band signal is assumed to be encoded using PCM and subsequent entropy coding. Quadrature mirror filters (QMF's) and short-kernel filters are compared in terms of entropy and SNR. It Is proved that the required entropy can be reduced by using short-kernel filters for subband splitting, while maintaining the original picture quality. New architectures of short-kernel subband filters which can furthermore reduce the entropy of subband signals are proposed.