Discrete Wavelet Transform based Fractal Image Compression using Parallel Approach

Abstract

Fractal based technique for compression is one of the popular methods for compression of videos and images. It has generated much interest due to its promise of high compression ratios at good decompression quality and it enjoys the advantage of very fast decompression and resolution independent decoding . But it suffers from highly computationally intensive encoding process which makes it unsuitable for real time applications. Many approaches have been suggested but they do not satisfy the requirement of low encoding time and high quality reconstructed images. In this paper parallel algorithm for fractal image compression using NVIDIs GPGPU is proposed. Also novel discrete wavelet transform based feature detection is used to reduce the number of block comparisons. Experimental results show significant reduction in encoding time and quality of reconstructed images is also good compared to other approaches making this technique suitable for real time applications such as image retrieval, image denoising, Image authentication and encryption, satellite and medical imaging.