Research On The Image Compression Algorithm Of The Set Partitioning In Hierarchical Trees Based On Wavelet Transform

As the important information source that the people use to understand the world, digitalimage have a very large amount of data. However, if people directly to transmit, process andstorage digital image, not only the cost is very high, but also the storage equipment and linebroadband can not bear such a large load, so the exploration of efficient image compressioncoding method has become a hot topic in the field of image processing. The wavelet transformhas been widely used in the field of image compression for its good time domain andfrequency domain localized features and multi-resolution decomposition characteristics. Theimage compression coding method based on wavelet transform have high compressionperformance, its research and improvement is an important research topic in the field of imagecompression.The thesis first elaborates the research background and significance of the subject, andthen describes the research status of the subject and the research focus and trends of waveletimage compression. After that, this thesis analysis the feasibility of image compression andbasic method of image compression in-depth, discuss the evaluation parameters and evaluationmethods of image compression quality and the international standard of image compressioncoding.Secondly, by in-depth understanding of the basic theory of the wavelet transform andanalysing the wavelet coefficients, the advantages of wavelet transform coding is summed up.Then the basic ideas and coding model of image compression coding method based on wavelettransform are mainly studied. After that, wavelet image compression algorithm are analysed,the SPIHT algorithm because of its good performance is choosed as the research object. TheSPIHT coding algorithm based on wavelet transform is studied in-depth and the codingprocess of SPIHT algorithm is presented. By analyzing the SPIHT algorithm in-depth, itsdisadvantages are pointed out.Finally, aiming at the problems of complicated convolution process of traditional wavelettransform and the unsatisfied effect of SPIHT algorithm for image compression, an improvedalgorithm is proposed. At first, the texture image is decomposed into N order with the help ofthe lifting wavelet and the first-order high frequency sub-bands are decomposed into N-1order by the lifting wavelet, and then the wavelet coefficients are encoded by the improved SPIHTcoding. The improved SPIHT coding improved the process method of the wavelet coefficientsin the low-frequency sub-bands and the detection method of some important coefficient in theL collection of the original SPIHT coding. Experiments show that the improved algorithm canretain the texture information of texture image more effectively and the quality ofreconstructed image and peak signal to noise ratio are better than the original algorithm at thesame rate. The effect is better especially at low rate, so the improved algorithm is an efficientcompression method for image compression.