Improved fractal-SPIHT hybrid image compression algorithm

Abstract

Image compression play an important role in efficient data storage and faster transmission in multimedia applications. In this paper, a hybrid compression algorithm is proposed combining Quadtree Fractal Image Compression (FIC) and Set Partitioning in Hierarchical Tress (SPIHT) techniques for better compression. First the image is decomposed using wavelet transform. Then Quadtree FIC is used on the approximation sub-band of the decomposed image and SPIHT is applied on all the detailed sub-bands. The performance of the proposed algorithm is assessed on standard images. Compared with the existing methods, it shows the superior results providing better PSNR of 37.07, low encoding time of 1.610 seconds and SSIM of 0.9805 for standard Lena image.