Selectively tree-structured vector quantizer using Kohonen neural network

Author

Wang, Wei ; Li, Xung ; Lu, Dajin

Author_Institution

Dept. of Electron. Eng., Tsinghua Univ., Beijing, China

Volume

2

fYear

1996

fDate

14-18 Oct 1996

Firstpage

1504

Abstract

The computational complexity and image fidelity are two key issues of vector quantization (VQ), especially for real-time applications. This paper proposes a three-phase self-organizing feature map (TPSOFM) algorithm to design selectively three-level tree-structured codebooks. The computational complexity during the coding process is reduced by a factor of 20 over a full search. A ten times speed up in training time is also achieved. The degradation of the reconstructed image quality is less than 0.38 dB in PSNR while the bit rate is reduced

Keywords

computational complexity; image coding; image reconstruction; self-organising feature maps; tree data structures; vector quantisation; Kohonen neural network; PSNR; TPSOFM algorithm; VQ; bit rate reduction; computational complexity; image coding; image fidelity; real-time applications; reconstructed image quality; three-level tree-structured codebooks; three-phase self-organizing feature map; training time; tree-structured vector quantizer; vector quantization; Bit rate; Computational complexity; Computer networks; Degradation; Image quality; Image reconstruction; Neural networks; PSNR; Quantization; Tree data structures;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing, 1996., 3rd International Conference on

Conference_Location

Beijing

Print_ISBN

0-7803-2912-0

Type

conf

DOI

10.1109/ICSIGP.1996.571162

Filename

571162