The Realization And Research Of SAR Image Target Detection Algorithm In DSP

The detection and identifying technique of radar targets holds the quite important position in the contemporary military application domain, because the synthetic aperture radar (SAR) is one kind of high resolution remote sensing imagery radar. It has the data's ability in all-weather, simultaneously has the penetration certain vegetation and the cover ability, compares with optics image easier to distinguish the ground the camouflage target and the hideaway feels the interest target in mountain, therefore obtains the widespread application in the military target reconnaissance and detection, therefore its image analysis and the target detection technology changes more and more importantly.With the unceasing progress of the technical and the image data gathering ability, how carried on accurate, the understanding and the detection to these images quickly have more and more arouse people's interest. But receives more and more many values to the target detection because of it is a sensation vital role of modern battlefield.Extended Fractal and the DP-CFAR form to carry on fusion processing, unifies the Dempder-Shafer evidence theory, carries on the detection classification in the decision-making strata to in the SAR target, was allowed to reduce the false alarm probability, enhanced the system detection capability. This topic mainly studied and researched the Extended Fractal algorithm, DP-CFAR algorithm and two algorithms confluction to the SAR image target detection, and confluction algorithm put into practice at TigerSHARC TS-201S EZ-KIT simulator of ADI Corporation by way of enhancing the utilization efficiency and the usability of the algorithm.Chapter1 mainly introduces the Extended Fractal and the DP-CFAR algorithm,as well as application and development status of fusion theory in SAR image target detection.It introduces chnological development of DSP, the image processing state and the present situation on the hardware, as well as the radar signal processing application on DSP.Chapter2 studies the basic principle of the Extended Fractal and the DP-CFAR ,and two algorithms fusion in the SAR image target detection application, and through carrying on the analysis and the comparison to real data processing.Chapter3 introduces the software and hardware resources in the DSP chip of TigerSHARC TS-201S EZ-KIT simulator in Transtech Corporation , and studies the development of the environment establishment and the application procedure, as well as the debugging step.Chapter4 designes three kind of realizations plans that bases on algorithm characteristic of Extended Fractal algorithm and the DP-CFAR algorithm,and has realized the Extended Fractal algorithm,the DP-CFAR algorithm and fusion algorithm on the simulator board, and evaluates the best realization efficiency according to different project.Chapter5 is a summary of the full text work, and has pointed out next step needs to continue to study the question.In this dissertation,the SAR image target-sized objects algorithm accomplish,based on the characteristic fusion of the Extended Fractal algorithm and the DP-CFAR algorithm,is reseached preliminarily on TigerSHARC TS-201S EZ-KIT simulator, and accumulates the certain experience of the next step of the realization of the SAR image target detection in the hardware environment.