Genetic algorithm with deterministic crossover for vector quantization
Pattern Recognition Letters
Data mining: concepts and techniques
Data mining: concepts and techniques
Genetic Algorithms
Density-Based Multiscale Data Condensation
IEEE Transactions on Pattern Analysis and Machine Intelligence
Diagonal axes method (DAM): a fast search algorithm for vector quantization
IEEE Transactions on Circuits and Systems for Video Technology
Hi-index | 0.00 |
Vector quantization (VQ) is a commonly used method in the compression of images and signals. The quality of VQ-encoded images heavily depends on the quality of the codebook. Conventional codebook training techniques are all based on the LBG (Linde-Buzo-Gray) method. However, LBG-based methods are noise sensitive and are not able to handle clusters of different shapes, sizes, and densities. In this paper, we propose a density-based clustering method that can identify arbitrary data shapes and exclude noises for codebook training. In order to rapidly approach an optimal solution, an improved version of a genetic algorithm is designed that demonstrates efficient initialization of codewords selection, crossover, and mutation. The experiments show that the proposed method is more robust in generating a common codebook than other LBG-based methods.