Set partitioning in hierarchical trees(SPIHT) is an efficient algorithm which is used for the image compressionwidely. SPIHT operates sequentially so, its parallel implementation isdifficult. In this study, the SPIHT algorithm is improved for providing that itis suitable for the parallel processing applications, and the correspondingpipelined datapath is designed for the proposed enhanced SPIHT algorithm. Thedatapath is designed to have three stages as preprocessing, list generation andoutput stream. In the preprocessing stage, the flags which are supports thelist generation stage are constituted. List of insignificant sets (LIS), listof insignificant pixels (LIP) and list of significant pixels (LSP) are formedin list generation stage. These lists contain the bit values which generate theoutput bit stream. The performance of the improved datapath design has beentested by compressing different images, and the obtained results are given.
Cite this article as: Çekli S, Akman A.Enhanced SPIHT Algorithm with Pipelined Datapath Architecture Design.Electrica, 2019; 19(1): 29-36.