Abstract:Vector Quantization (VQ) is a technique with a wide range of applications. For example, it can be used for image compression. The codebook design for VQ has great significance in the quality of the quantized signals and can benefit from the use of swarm intelligence. Initialization of the Linde–Buzo–Gray (LBG) algorithm, which is the most popular VQ codebook design algorithm, is a step that directly influences VQ performance, as the convergence speed and codebook quality depend on the initial codebook. A widely used initialization alternative is random initialization, in which the initial set of codevectors is drawn randomly from the training set. Other initialization methods can lead to a better quality of the designed codebooks. The present work evaluates the impacts of initialization strategies on swarm intelligence algorithms for codebook design in terms of the quality of the designed codebooks, assessed by the quality of the reconstructed images, and in terms of the convergence speed, evaluated by the number of iterations. Initialization strategies consist of a combination of codebooks obtained by initialization algorithms from the literature with codebooks composed of vectors randomly selected from the training set. The possibility of combining different initialization techniques provides new perspectives in the search for the quality of the VQ codebooks. Nine initialization strategies are presented, which are compared with random initialization. Initialization strategies are evaluated on the following algorithms for codebook design based on swarm clustering: modified firefly algorithm—Linde–Buzo–Gray (M-FA-LBG), modified particle swarm optimization—Linde–Buzo–Gray (M-PSO-LBG), modified fish school search—Linde–Buzo–Gray (M-FSS-LBG) and their accelerated versions (M-FA-LBGa, M-PSO-LBGa and M-FSS-LBGa) which are obtained by replacing the LBG with the accelerated LBG algorithm. The simulation results point out to the benefits of the proposed initialization strategies. The results show gains up to 4.43 dB in terms of PSNR for image Clock with M-PSO-LBG codebooks of size 512 and codebook design time savings up to 67.05% for image Clock, with M-FF-LBGa codebooks with size N=512, by using initialization strategies in substitution to Random initialization.

Codebook Design for VQ of LSP

Vector Quantization Based On Genetic Simulated Annealing

Vector Quantization Based On Genetic Algorithm

A tabu search based maximum descent algorithm for VQ codebook design

Edge Classification Based Initialization Technique for Image Vector Quantizer Design.

LG-VQ: Language-Guided Codebook Learning

Codebook Sharing in Multi-Stage Vector Quantization

Multi-Stage Codebook Optimization for LSF Quantization with Inter-Stage Prediction

VQ Codebook Design Using Modified K-means Algorithm with Feature Classification and Grouping Based Initialization

Fast Codebook Design Method for Image Vector Quantisation

Structural Codebook Design And Address-Dependent Vector Quantization

Resizing codebook of vector quantization without retraining

Edge and Contrast Classified K-means Algorithm for Image Vector Quantizer Design.

On the Initialization of Swarm Intelligence Algorithms for Vector Quantization Codebook Design

Research on Low Delay Low Bit Rate Speech Coding Algorithm

VQ Image Coding Using Sub-Vector Techniques.

Equal-Average Equal-Variance Equal-Norm Nearest Neighbor Search Algorithm for Vector Quantization

An efficient encoding algorithm for vector quantization based on subvector technique.

Advanced Nonlinear SCMA Codebook Design Based on Lattice Constellations

SKVQ: Sliding-window Key and Value Cache Quantization for Large Language Models

Parallelization of the LBG Vector Quantization Algorithm for Shared Memory Systems