Research On Lossless Compression Coding Technique Of VDR Radar Image

There are three problems of VDR radar image during its collecting, delivering and storing . (1)A digital image usually includes a great deal of noise because it is often disturbed by noise during its collecting, digitizing and delivering. Therefore, it is inconvenient for image to be analyzed, observed and compressed .(2) Because of limited storage and huge image data, it is hard to deliver and store images. (3) What we most care is just only a certain scope around own ship of radar image, named as ROI. And ROI coding is also concerned. For above problems denoising, lossless compression and ROI coding of VDR radar image are investigated.For denoising of VDR radar image, an improved wavelet packet algorithm is presented. But due to the subsection of low frequency part and high frequency part of the superstratum, wavelet packet has more accurate local analysis ability than that of wavelet analysis. However, because of down sampling of wavelet packet decomposition, the details dissatisfy the sampling theorem, and bring the problem of band overlapping . An improved wavelet packet algorithm is presented. By using the characteristic of signal frequency shift, the signal frequency is shifted and the highest frequency of the signal is cut down, or the decomposition coefficients of wavelet packet are reordered, to solve the problem of band overlapping, produced by downsampling. The image can be denoised with high quality.Aiming at the characteristic of high speed , efficiency and less memory an algorithm that combines the integer wavelet transform with improved SPECK is presented .The image is transformed by integer wavelet first, and then coded by SPECK . By using block structure, that is easy for parallel computing, SPECK not only speeds up coding, but also occupies minimum dynamic memory and good fault tolerance. Synthetic match and Hash table are introduced to improve SPECK. Improved SPECK has higher compression speed and compression ratio. The simulation result illustrates the validity of improved SPECK. What we most care is only a certain scope around own ship of radar image, ROI coding technique is investigated. By investigation of human vision, ROI coding algorithm based on integer wavelet and asymmetric shift is introduced. The image is transformed by integer wavelet first, and then the coefficients are shifted unevenly. MQ adaptive coder is employed. This method makes the ROI data recovered in advance in low bit-rate code stream. It makes the progressive reconstruction of the image data more convenient for human vision system. It is useful to improve image compression. It can be applied to lossless compression of several ROI. It has good future in application .