A Fast VQ Codebook Design Algorithm
Jiwen Dong
Abstract
Jiwen Dong
Abstract
Vector quantization is a very important signal compression approach with wide applications in such areas as image and speech signal compression.A key issue of VQ is the design of a codebook.Many algorithms have been presented and applied to the design of a codebook.However,most of them were only applied to a codebook which included less codes.With the numbers of vectors increasing,computation complexity increases significantly.The presented approach can well decrease computation complexity,and the FCM algorithm in this approach can avoid the local minimization.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Vector quantization is a very important signal compression approach with wide applications in such areas as image and speech signal compression.A key issue of VQ is the design of a codebook.Many algorithms have been presented and applied to the design of a codebook.However,most of them were only applied to a codebook which included less codes.With the numbers of vectors increasing,computation complexity increases significantly.The presented approach can well decrease computation complexity,and the FCM algorithm in this approach can avoid the local minimization.
Key concepts: Codebook, Linde–Buzo–Gray algorithm, Vector quantization, Algorithm, Computation, Computer science, Computational complexity theory, Quantization (signal processing)