AN OPTIMAL-DESIGN PROCEDURE FOR INTRABAND VECTOR QUANTIZED SUBBAND CODING

Subband coding with vector quantization is addressed in this paper. Fo rming the data vectors from both between and within the subbands is co nsidered. The former of these two schemes is referred to as interband coding and the latter as intraband coding. Interband coder design is r elatively straightforward since the design of the single codebook invo lved follows readily from a representative set of interband data vecto rs. Intraband coder design is more complicated since it entails the se lection of a vector dimension and a bit-rate for each subband. The mai n contribution of this work is an optimal methodology for intraband su bband vector quantizer design. The problem formulation includes constr aints on the bit-rate and the encoding complexity and is solved with n onlinear programming methods. Subband vector quantization image coding in conjunction with a human visual system model is thoroughly investi gated. Results of a large number of experiments indicate that the opti mal intraband coder yields superior results from quantitative as well as' subjective points of view than the interband coder for comparable bit-rates. This improvement becomes more pronounced as the computation al complexity of the intraband encoder is allowed to increase.