K-Means Clustering and Wavelet Based Image Compression

Main Article Content

T. Kathirvalavakumar
E. Ponmalar

Abstract

This work proposes a novel method of performing image compression using a hybrid of K-means and wavelet techniques. K-means clustering technique is used as a vector quantizer to generate centroids when it is applied over the vectors representing an image. The proposed work compresses the centroids (code vectors) by applying the Discrete Wavelet Transform over the centroids. The approximation coefficients of the centroids along with the index values of the centroids corresponding to each input vector representing the image blocks form the compressed image. The proposed technique is implemented successfully and the experimental results show that the hybrid of K-means clustering and wavelet transform techniques produces better image compression with improved performance measures.

 

Keywords- Image Compression, K Means Clustering, Wavelet transform, DWT coefficients, decompression

Downloads

Download data is not yet available.

Article Details

Section
Articles