Comparison of K-Means, K-Means++, X-Means and Single Value Decomposition for Image Compression

Uncompressed digital images need huge storage space and large transmission bandwidth for the transmission over the network. Image compression reduces the image size and enables the efficient storage and transmission of image files. The main goal of image compression is the best image quality at the given compression rate. In this paper, the performance of Singular Value Decomposition (SVD), K-means, K-means++, and X-means for image compression is discussed. K-means, K-means++, and X-means clustering reduce the number of colors of images. Singular Value Decomposition performs a low-rank approximation of the image matrix. Image compression parameters mean-square error (MSE), peak signal-to-noise ratio (PSNR). and compression ratio (CR) are evaluated for the considered compression methods.