Employing hybrid methods for compression color images

Abstract

The purpose of compressing images to represent the image less to provide cost data storage and transmission time, however, the effectiveness of pressure accomplished by bringing the original image (instead of the exact loss). Therefore clustering problem defined as a collection of objects which are “similar” between them and are “dissimilar” to the objects belonging to other clusters; so, as every other problem of this kind, it deals with finding a structure in a collection of unlabeled data that providing a novel solution to the color image compression by exploiting the ability to generating groups of data by using k-mean algorithm. In this study true color images converted to YIQ color space then k-mean algorithm applied on the Y component to determine the number of cluster which used to constructing the clustered images. Run length encoding (RLE) algorithm applied on the resulting YIQ clustering images. In decompression stage RLE decompression algorithm used to reconstruct the RGB color images. Number of quality measurement computed like (peak signal to noise ratio (PSNR), (mean square error) MSE and signal to noise ratio (SNR) to measure the amount of distortion in this processes also the compression ratio calculated.