Low Complexity Image Compression Architecture Based on Lifting Wavelet Transform and Embedded Hierarchical Structures

Several primary concern points should be deliberated in the wireless sensor network WSN design. When the networks are included with cameras, limitation in the image data sizes pose as a new problem. Hence there is a necessity to find new ways for data processing and communication. If the size of data could be minimized, image compression would reduce the memory requirement and thus communication costs. Recently, transform-based image compression methods are still very attractive and popular. These methods are mainly based either on Discrete Cosine Transform DCT such as the Joint Photographic Experts Group JPEG or Discrete Wavelet Transform DWT such as JPEG2000. DCT based algorithms are fast with low-complexity and low-memory. However, they often cause annoying blocking artifacts in the low bit rate transmission. The low complexity embedded DWT-based coders generate a bitstream that can be decoded at multiple transmission bit rates with an acceptable quality of the reconstructed image at the reception. Set Partitioning in Hierarchical Trees SPIHT is among the most popular quality-scalable wavelet based image coders. In this paper, the lifting scheme LS implementation of wavelets is also investigated before the set-partitioning coding is applied to compress the images. However, with fewer bits to transmit using the SPIHT coder results, this technique will be suitable to restricted property with limited resources platforms.