New clustering algorithm for Vector Quantization using Haar sequence

Codebook generation plays an important role in Vector Quantization (VQ) such that the distortion between the original image and the reconstructed image need to be minimum. The paper presents an effective clustering algorithm to generate codebook for vector quantization. In Kekre's Error Vector Rotation (KEVR) while splitting the cluster every time new orientation is introduced using error vector sequence. This error vector sequence is binary representation of numbers., so cluster orientation change slowly in every iteration. The Kekre's Error Vector Rotation using Walsh (KEVRW) uses Walsh sequence to rotate the error vector. Because of this cluster orientation change rapidly in every iteration. The proposed codebook generation technique Thepade's Haar error vector rotation (THEVR) is based on KEVR algorithm. Here the error vector used for splitting the clusters in Vector Quantization is proposed to be prepared using discrete HAAR transform matrix. The proposed methodology is tested on different training images for various codebook sizes. The obtained results show that THEVR gives less MSE as well as less distortion as compared to KEVR., KEVRW indicating better image compression.