COMBINATION OF LOSSY AND LOSSLESS FOR IMAGE COMPRESSION

Abstract

Image compression is a fast paced and dynamically changing field with many different varieties of compression methods available. Images contain large amount of data hidden in them, which is highly correlated. A common characteristic of most images is that the neighboring pixels are correlated and therefore contain redundant information. This research presents an algorithm for image compression consist of a combination of lossy and lossless methods which based on a discrete wavelet transform and apply entropy coding as a lossless compression with using a quantization and thresholding techniques to produce a high compression ratio and high quality of image. Two images (.bmp format) of (256×256) pixels were used as samples to test the algorithm. Experimental results were estimated using the peak signal-to-noise ratio (PSNR) which considered as a measurement to test the quality of reconstructed images samples.