Compression Of Remote Sensing Image Based On Wavelet Transform

With the development of remote sensing technology and the user's requirements of high resolution,the resolution of remote sensing images are more and more high.For storage and transmission,huge information must to be compressed.Remote sensing image has weak of space correlation , abundant image details and plenty of information.The wavelet transform have better space-frequency characteristics,multi-resolution and decorrelation.It has the advantage of describing steady and non-steady signal.Wavelet transform becomes the main subject of remote sensing image compression coding.Firstly,this paper elaborate the significance, the principle ,the development and the research of image compression coding.Then some algorithm and standard about compression are introduce.Basic theory and the superiority of wavelet transform, multi-resolution analysis,Mallat algorithm and lifting wavelet have been illustrate.Finally,the improved algorithm of SPIHT is presented for remote sensing image.The effect of waveletbase and decomposition level to the quality of remote sensing image recomstruction are analyzed by both theory and experiment.And then get the best waveletbase and decomposition level for remote sensing image.Because the large calculation and the low coding speed of SPIHT algorithm,the original algorithm is improved for remote sensing image.The low frequency and the high frequency of the remote image are seperated.The low frequency is lossless compressed.The logical redundancy between node of the same tree are reduced.The list of max threshold reduce zero-coding for LIP and LIS.It reduce the calculation and the scanning time. Run-length algorithm and WNC arithmetic coding algorithm are used in the improved algorithm.It increase the compression ratio.The experimental data indicate that the PSNR,comprssion ratio and run time of the reconstruction image by the improved algorithm are better than the original algorithm.