Research Of Ship Detection And Refocusing Technology In SAR Image

As the development of Synthetic Aperture Radar(SAR)technology,the resolution of SAR image has been significantly improved and the amount of image data has increased in a massive manner.SAR image processing technology is widely applied in the field of remote sensing.Ship detection in SAR images is important part of SAR image interpretation,and has become an important research issue of ocean remote sensing.It has vital significance for military and civilian applications.Ship detection in SAR image is to detect ship target from clutter.It usually needs sea-land segmentation before ship detection.The quality of sea-land segmentation directly relates to subsequent ship results.It can reduce the false alarm rate after removing the land.There are many moving ships in the sea.The motion of ship causes the SAR image to be defocused.It is necessary to perform motion compensation for features extraction and ship classification.This paper mainly research on the ship detection in SAR image and the refocusing of moving ship target.The main research contents are as follows:Firstly,this paper proposes a sea-land segmentation method for ship detection in SAR image which combined Global Self-consistent,Hierarchical,High-resolution Geography Data-base(GSHHG)and improved CV model.The shoreline contour is generated with the GSHHG database as the initial contour of the CV model.The CV model is improved and simplified.ROEWA-based edge indication function replaces ?(?)term and the complex items are removed,hence the efficiency of curve evolution is improved.The experiments of sea-land segmentation based on real SAR images are conducted and the results show the proposed method is superior to other methods.Secondly,the ship detection method based on CFAR in SAR image is studied,and the sea clutter statistical modeling method and Constant False Alarm Rate(CFAR)detector are analyzed.The clutter fitting for SAR image experiments are carried out and the results are analyzed.According to the clutter fitting results,the CFAR method based on different distribution is used to performed ship detection on the image after removing the land and the detection results are analyzed.Finally,this paper studies the refocusing of moving ship in SAR image and proposes a refocusing method for defocused ship in SAR image with the ISAR processing technology.First,defocused moving ships are selected after ship detection.Then,the SAR image of ship is inverted back to raw data domain with inversion algorithm.A fast minimum entropy phase correction method for motion compensation is proposed in this paper,and the focused image is generated.The experiments of refocusing based on TerraSAR-X,GF-3 and airborne SAR image are conducted.Experimental results show that the proposed method of refocusing is effective.