Implementation of Efficient Lossless Image Coding Using Enhanced SPIHT Algorithm

• Published in: I-Manager's Journal on Electronics Engineering Vol. 1; no. 1; pp. 39 - 46
• Main Authors: Deepa, P, Vasanthanayaki, C
• Format: Journal Article
• Language: English
• Published: Nagercoil iManager Publications 01.09.2010
• Subjects: Algorithms; Compression ratio; Image coding; Image compression; Lossless; Matlab; Wavelet
• ISSN: 2229-7286; 2249-0760

Lossless image compression is a subject of great importance today in the fields like medical imaging, satellite communication, testing of materials etc., as it represents an efficient way of preserving the exact data content of original image. Among the various wavelet based image coding algorithms, SPIHT (Set Partitioning In Hierarchical Tree) algorithm is a very effective and computationally simple for both lossy and lossless image coding. In the proposed work, enhanced SPIHT algorithm for image compression has been taken to achieve better lossless coding performance without harming its simplicity. The conventional SPIHT uses more bits, as it encodes four direct descendents of each significant set found to be insignificant in the initial iteration. In the enhanced SPIHT, the significance of the direct children is tested as a group instead of sorting them separately. If the number of significant sets is much higher than insignificant sets, every set is directly partitioned without testing. The proposed enhanced SPIHT algorithm has been implemented using image processing tool box and wavelet tool box in MATLAB 7.8. Performance of the proposed algorithm has been evaluated in terms of PSNR, compression ratio, bit rate, complexity and finally it has been analyzed for noise. Also the performance is compared with the conventional SPIHT algorithm for many samples. Simulation result shows that there is a significant improvement in compression ratio and PSNR over the conventional SPIHT algorithm.