Development Of Remote Sensing Image Compression Unit For Ocean Surveillance Satellite

Ocean surveillance satellite is a kind of reconnaissance satellite which is used to detect and monitor the activities of marine ships,submarines and other targets.Among them,ocean surveillance satellite based on optical remote sensing imaging has been widely used in military reconnaissance,strategic early warning and other fields.With the rapid development of space remote sensing technology,the width and frame rate of remote sensing imaging continue to improve,resulting in a sharp increase in the amount of image data,which brings huge pressure to the data transmission bandwidth and onboard storage.Therefore,remote sensing image compression has become a necessary means to reduce the amount of data and relieve the transmission bandwidth and storage pressure.At present,lossy compression method is widely used to compress remote sensing image at high magnification.Although this method reduces the amount of image data to a large extent,it also causes the loss of key information in the region of interest of the image,which further hinders the development of ground observer's recognition.Therefore,in the ocean surveillance task,it is of great significance to achieve high-speed image data compression while ensuring the integrity of information in the region of interest.The main research work of this paper is as follows.Firstly,for the ocean surveillance application scenario of the micro nanosatellite platform,this paper analyzes the specific functions and technical indicators of the image compression unit.On this basis,the flow analysis of the image compression method and the overall design of the hardware platform is completed.Based on the high-performance heterogeneous processor,the satellite remote sensing image compression unit is designed,and the hardware platform of the satellite image compression unit is realized.Secondly,aiming at the loss of key information in the region of interest in lossy compression,the remote sensing image in the ocean scene is analyzed,and it is concluded that it has the characteristics of a single image scene and a big difference between target and background features.Based on this feature,this paper proposes an interest extraction method based on the spatial attention model and OTSU,which can effectively achieve the high-precision extraction of ship,port,and other regions of interest in the ocean background.Thirdly,H.265 is selected to be used in video compression coding according to the actual requirements of compatible video and still image compression functions in the reconnaissance mission,and the compatibility of still image compression functions is realized by cutting the inter-frame compression part of its compression algorithm.In addition,considering that the size of wide remote sensing image is too large,which is not conducive to coding processing,the original image is sliced and recombined in this paper,which not only meets the input format requirements of the compression coding unit,but also improves the compression efficiency of the image.Finally,based on the hardware platform of the satellite image compression unit,the proposed remote sensing image compression method is implemented and verified by the software and hardware co-design method.The experimental results show that the compression unit designed in this paper can effectively improve the overall compression ratio of the image and the effective data transmission efficiency on the premise of ensuring the image quality of the region of interest,which has a high practical value.