Medical Image Compression Based On Improved Spiht

We recognize the world and perceive the world through image information, these information are colorful and full of amount of details, which contains important and unimportant points. With the development of modern communication technology, computer network technology and information processing technology, image information processing, storage and transmission play more and more important role in social life. The compression coding solution for image storage and transmission gradually become the research hotspot around the globe, The traditional image coding method is better to remove statistical redundancy of image information, but it has obvious block effection in high compression rate, The wavelet transform is a time-frequency information analysis method, It has the mutil-analysis characteristic, and overcome the traditional shortcomings of image compression algorithm. it also has a flexible application in the field of image compression. Encoding is one of core technology of wavelet transformation, which directly decides the characteristic of wavelet coefficients, and influent other operations in subsequent compression process, and eventually have great effect on image reconstruction quality and compression rate. Hence, the method how to improve encoding efficiency and to optimize wavelet coefficient so that the wavelet encoding algorithm can become faster and more efficient, has become one of hot spots of current study.As one kind of the Grey images with which medical image has the same characters, however, Medical images has it particularity, such as it should meet the requirement of clinic diagnosis, According to the characters above, this paper aims to improve the quality of reconstructed image by lifting the efficient of coding algorithm, with the precondition of guaranteeing the image compression ratio.Firstly, this paper makes deeply analyze about some mainly issues that produced in the process of wavelet transforming on the basis of Wavelet transform theory. such as, Considering validity and real-time of image compression, this paper discusses the selection of wavelet basis, characteristic of wavelet transform and wavelet constructed by lifting wavelet to replace the traditional method of research, and ultimately select a better wavelet basis, then use this selected Wavelet bases to compare data from different layer of the medical image, Lastly, select the better method according to the character of selected Wavelet bases and some kinds of Border extension methods. Secondly, We make a detailed analysis of the most classic coding algorithm based on grasping the principle of wavelet transform, that is Set partitioning in hierarchical wavelet tree algorithm(SPIHT algorithm), analyzing SPIHT algorithms in detail, Through analysis of SPIHT algorithm, it needs to repeat to scan the wavelet coefficients, comparing with its threshold, we can find important coefficient, thereby increase the number of memory access and memory load; when classify to scan the LSP and the LIP, at the same time, encode the D type and L type, there exists situation that some important wavelet coefficients are encoded redundantly, which greatly reduces the efficiency of the codec. Against the above shortage, on the basis of the SPIHT algorithm, medical data compression algorithm based on improved SPIHT is put forward, it achieves the coding of medical images. The algorithm not only can reduce the coding time, but also can get higher compression ratio and higher PSNR of the reconstructed image. Improved SPIHT algorithm mainly sorts the wavelet coefficients, and then eliminates some of the important coefficient coding redundancy, and finally adjusts the scan set partitioning strategy to improve SPIHT algorithm via these. Comparing with the traditional SPIHT algorithm, encoding scheme based on wavelet decomposition and improved SPIHT algorithm can significantly improve the PSNR of the reconstructed images, and also guarantee the quality of the reconstruction of medical images, simultaneously can meet storage of the medical data and transmission of the need.At last, the paper introduces the existing compression coding technology and wavelet transform coding theory, And then make some simulation experiments base on Visual studio2008software platform. Reveals the wavelet transform image compression in real terms-with a small amount of the most important wavelet coefficients to reconstruct the whole image and achieve the purpose of image compression. Through using10pieces of medical images which its size is512×512×8to make a simulation test, and use the standard512×512×8Lena image to compare, we use int9/7wavelet, compare the SPIHT algorithm and improved SPIHT algorithm,5level wavelet decomposition, the whole process is fully implemented in the software platform.Experiment show that the algorithm can better retain the objective image quality, at the same time, it can improve the efficiency of the codec and we can get better compression. the PSNR of improved algorithm is better than that of original SPIHT at the same ratio.