基于高时相探测的运动点目标检测方法  

Moving Point Target Detection Based on High Frame-rate Image Sequence

作　　者：牛文龙 吴勇 杨震 郑伟 刘波 NIU Wenlong;WU Yong;YANG Zhen;ZHENG Wei;LIU Bo(Key Laboratory of Electronics and Information Technology for Space Systems,National Space Science Center,Chinese Academy of Sciences,Beijing 100190;University of Chinese Academy of Sciences,Beijing 100049)

机构地区：[1]中国科学院国家空间科学中心复杂航天系统电子信息技术重点实验室,北京100190 [2]中国科学院大学,北京100049

出　　处：《空间科学学报》2019年第4期520-529,共10页Chinese Journal of Space Science

基　　金：中国科学院国家空间科学中心青年科技创新项目资助(Y9211BAE9S)

摘　　要：针对可见光探测中低信噪比运动点目标检测问题,提出一种基于高时相探测的运动点目标检测方法,构建了基于双谱分析的目标检测器来提取像元时域特征,对背景像元与目标像元进行区分.仿真分析与实验结果均表明,所提出的方法能够对低信噪比运动点目标进行有效检测,在一定帧频范围内,目标检测能力与采样帧频正相关.相比常用的运动点目标检测方法,本文方法具有更高的检测效能.A high frame-rate based framework is presented to detect moving point target in very low SNR.A novel target detector based on higher order statistics is proposed to analyze the time domain evolution of visual image sequence for distinguishing the background and target.Our method is formulated to detect a time-domain transient signal and the bispectrum is used to characterize the temporal behavior of pixels.The method is evaluated using both simulated and real-world high frame-rate data,and a comparison to other widely used point target detection approaches is provided.Experimental results demonstrate that the proposed framework can be used for robust moving point target detection in very low SNR.

关 键 词：高时相 目标检测 时间序列 运动点目标 高阶谱 

分 类 号：V528[航空宇航科学与技术—人机与环境工程]