A SIMPLIFIED METHOD FOR FRACTAL IMAGE COMPRESSION USING QUAD TREE DECOMPOSITION WITH HUFFMAN CODING (KEY IJP************947)

• P Mohan

Abstract

Image compression is a process of reducing the size of a digital image file while preserving its essential visual information, typically by removing redundant or less important data to save storage space or transmission bandwidth. Many approaches in compression have been suggested but they do not satisfy the requirement of less encoding time and high quality reconstructed images. In this paper, Fractal image compression will be used. It uses fractal geometry of the image data stream to achieve lossy compression. Fractal compression of images will be highly effective form of compression. It is based on the parts of an image frequently similar to other parts of the same image. This will result in significant reduction in encoding time and improved quality of reconstructed images. Quad tree decomposition is a type of approach that splits an image into a set of homogenous regions and is the most commonly used partitioning mechanism. Each square image will be partitioned into four equal-sized square blocks, and each block will be examined for homogeneity. Quad tree Decomposition with Huffman Coding Technique will be considered for natural textured images. Huffman Coding, data compression, methodology will be lossless. In this approach, the performance metric of Peak Signal to Noise Ratio (PSNR) will be considered for evaluation.