Research On Radar Image Compression

The data gained from radar is enormous, which produces huge difficulties for high-speed transfers and real-time storage in modern radar systems. So it's a key matter for radar image compression to reduce the data to transfer and store them conveniently.In this paper, based on radar's purpose and character of radar image, it chooses suitable compression arithmetics to compress the radar images. The main research is compression in one frame, which is the base of image compression. For more frames, the compression can translate into the way to compression of one frame. Two transform arithmetics are used in this paper to wipe off relation part. Images after transformed are analyzed to choose suitable arithmetic. Compared to every arithmetic, compression effects are analyzed, which can offer references for applications.The main researches in this paper are as follows:1.Compared radar images with nature images from Statistic and correlation, analyzing the differences and similarities, the appropriate arithmetics can be chosen based on these characteristic of radar image;2. At first, DCT transform is chosen, because it is perfect from software to hardware. And arithmetic is simple. Through analyzing the radar image after transforming, the Huffman processing and Run Length Coding are chosen based on the characteristics;3. In this paper, the theory of wavelet transform and Mallat arithmetic used in wavelet transform Compression is discussed and the results are analyzed after transform. EZW (the embedded zerotree wavelet algorithm) is introduced. Compared radar image with nature image used EZW, the correlative results are educed;4. Analyzing character of every arithmetic and compression results in this paper, suitable compression arithmetic can be chosen in different appliance.