Iterative optimized design method for selecting predetermined vector quantisation

Abstract

The present invention discloses an iteration optimum design method of selecting predictive vector quantification. The method comprises the steps of dividing training data by using conventional correlation between front and back frames, performing predictive parameter calculation and codebook design by using different training data, quantizing the training data by using designed predictive parameter and codebook and adjusting the training data division according to the quantization error value, and performing optimized selection of predictive vector quantification codebook through the training data division and codebook design iteration. According to the present invention, defect of fixed division of training data is avoided for effectively improving the performance of selecting predictive vector quantification design. The method also reduces the parameter quantizing distortion and improves the signal compression quality.

Description

technical field [0001] The invention relates to an iterative optimization design method for selecting predictive vector quantization, which belongs to the field of information source coding. Background technique [0002] In order to improve the degree of information source compression, vector quantization has been widely used in various information source compression coding algorithms. In various compression coding algorithms, people have adopted different vector quantization structures, including predictive vector quantization, multi-level vector quantization, split vector quantization, classification vector quantization, gain shape vector quantization and trellis vector quantization, etc. Among them, the predictive vector quantization multiplies the parameter of the previous frame by the corresponding predictive coefficient to obtain the predictive value of the current frame parameter, subtracts the predictive value from the current frame parameter to obtain the predictive...

AI Technical Summary

Problems solved by technology

[0005] The above design methods for selective predictive vector quantization all use fixed training data division, and do not fully consider the appropriateness of this division.

Images