基于G0分布的高海况SAR船只目标检测方法

摘要

Ship target detection under rough sea conditions is a difficult one in SAR ship detections. The effectiveness of the ship target detections can be much influenced in accurately describing the sea clutter from the rough sea conditions by the selected distribution type. The G0 distribution commonly used in the modeling for the SAR images of urban area is here applied to modeling the sea clutter for rough sea conditions.The effectiveness of the G0 distribution in the modeling for various satellite-borne SAR images are analyzed to verify its suitability to the sea clutter from sea surface. It is learnt from the verification that the G0 distribution is better applicabe to the sea clutter from the extremely inhomongeneous sea area than the commonly used distributions. In combination with the CFAR target detection, the ship target detection algorithms based on the G0 distribution for satellite-borne SAR images is proposed, and its effectiveness is verified in the in-situ synchronized experiment.%高海况条件下船只目标检测一直是SAR船只检测的难点之一,选用何种分布来精确描述强海杂波对目标检测效果有重要影响.文章将常用于陆地城区SAR图像建模的G0分布应用于高海况条件下杂波建模,分析了G0分布对不同星载SAR数据的建模效果,验证了其用于海面杂波拟合的适应性,得出G0分布对极不均匀区域海杂波建模能力要优于常用分布;结合恒虚警目标检测算法,提出了基于G0分布的星载SAR船只目标检测算法;利用海上同步实验,验证了此方法的有效性.