A Hybrid Hardware Technique for Image Compression Using Fractal Quadtree Decomposition

Abstract

For the current voluminous image data, handling of the image especially the efficient storage newlineand low bandwidth transmission is an important aspect in the digital image processing field. newlineThis requires great emphasize on low encoding time (in Nano seconds), high-value of Peak newlineSignal to Noise Ratio (PSNR), high Compression Ratio (CR) coupled with fast real time data newlinetransmission. newlineExisting digital image approaches for compression are inadequate to address the newlineabove requirement on standalone basis. Traditional digital image compression techniques newlinehave long time of encoding, degraded image output and use large available resources. Hence, newlinethere is a need to devise an effective hybrid compression technique which can combine newlineseveral traditional standalone image compression techniques to gain a high speed, great newlineefficiency and address the above mentioned drawbacks. newlineThe proposed hybrid hardware (DWT [Discrete Wavelet Transform] -- DCT [Discrete newlineCosine Transform] -- Fractal Quadtree with Huffman Encoding) architecture will address the newlineabove issues by providing high compression ratio, improved encoding time, high PSNR newlinevalue, and also provides a superior reconstructed image. newlineIn this research work, the proposed hybrid (DWT -- DCT -- Fractal Quadtree with newlineHuffman Encoding) algorithm for image compression will combine the benefits of the newlinestandalone DWT, DCT and Fractal Quadtree with Huffman coding compression methods. newlineThe algorithm will perform Fractal Quadtree compression on DCT and DWT frequency newlinecomponents. The experimentations are performed on a standard image. We are also newlinecomparing the results of our proposed method with standalone DWT, DCT, and Fractal newlineQuadtree with Huffman coding Compression Methods. newline