Filtering and Searching Vector Quantization

Authors:
Shih-Yu Huang
Affiliations:
Department of Computer Science, Ming Chuan University, 5 Teh-Ming Rd., Gwei Shan District, Taoyuan Country 333, Taiwan
Venue:
Journal of VLSI Signal Processing Systems
Year:
2003

Citing 3
Cited 0

An algorithm for finding nearest neighbours in (approximately) constant average time

Pattern Recognition Letters
Vector quantization and signal compression

Vector quantization and signal compression
Fast VQ encoding by an efficient kick-out condition

IEEE Transactions on Circuits and Systems for Video Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

Under the consideration of computational complexity and design regularity, in this paper, a FASVQ (filtering and searching vector quantization) is presented to compress images. FASVQ utilizes a heuristic to filter codevectors with small costs and then employs full-search VQ within the surviving codevectors. We have proven that the proposed heuristic can easily be implemented by a table lookup technique and over 95% codevectors can be filtered. Although, the quantized codevector of FASVQ wouldn't be optimal, the experimental results show that the PSNR degradation between full-search VQ and FASVQ is only 0.24 dB on the average.