FPGA Implementation Technology Of Image Processing For Spaceborne Large FPA Camera

With the continuous development of infrared detection technology,infrared image processing systems are widely used in civil and military fields.Spaceborne infrared detectors have a high data rate when imaging with a large FPA,and the bandwidth of the ground transmission channel is limited.Therefore,image compression is the key to ensuring the effective download of on-board data.For the task of detecting and recognizing dim and small targets on space,the loss of target information is unacceptable.Therefore,it is necessary to study high-performance image compression methods to preserve target information.This thesis focuses on the research of suspected target detection and image frame motion matching in the image compression task of preserving target information.An infrared image processing system is designed and implemented to complete infrared small target detection and frame motion matching in real time.The existing image compression system is closely integrated to complete the task of on-board image processing.Aiming at the problems of low signal-to-noise ratio and high false alarm rate in the detection of infrared dim and small targets under the complex background of deep space,based on the DLCM detection algorithm,the image is preprocessed by means of filtering,and the gray gradient characteristics of the target are fully utilized Perform post-processing to reduce the detection false alarm rate;use the strategy of lossless compression on the target area and slightly lossy compression on the background area to improve single frame compression performance;use the full search block matching algorithm based on the absolute error sum as the matching criterion,and use 16 ×64 image blocks perform motion matching and cooperate with the LOCO-3D compression algorithm to improve the performance of multi-frame compression.Finally,under the premise of ensuring system performance,the algorithm flow is pipelined,and the threshold calculation is optimized,which improves the running speed of the algorithm on the FPGA.This thesis adopts the idea of "global pipelining and local parallelization",and flexibly uses the methods of "area for speed" and "speed for area" to realize framing,image preprocessing,infrared target detection and inter-frame movement matching modules on the FPGA platform.These module,together with the image compression module and other interface modules,forms a complete infrared image processing system.In order to effectively reduce the interference of the single event effect on the system,this thesis has carried out a three-mode redundancy design for the key modules and signals of the system to make the system more reliable.Finally,this article conducts functional testing and performance analysis of the infrared image processing system based on FPGA platform.The experimental results show that the infrared imaging weak target detection and dynamic image compression processing system implemented in this article has high target detection rate,low false alarm rate,and reliability.The characteristic of high performance meets the needs of engineering applications in the space environment.