摘要
针对当前普通摄像头对夜间航行船舶检测研究方面的不足,提出一种基于光斑检测和跟踪的方法,实现夜间航行船舶检测。采用边缘检测算子(Laplacian of Gaussian,LOG)对视频图像光斑进行检测,采用阈值化方法对光斑进行过滤;Kalman滤波算法对多光斑目标进行跟踪,通过对光斑进行标记,判断光斑的属性,从而实现光斑的新增和删除。以实际夜间船舶航行监控视频为例进行方法验证,结果表明:该方法能对夜间船舶进行有效检测和跟踪,从而验证采用光斑跟踪的方法能提高夜间船舶的检测精度。
Light spot detection and tracking that enhances ordinary monitoring system is proposed for normal cameras achieving ship monitoring in nighttime.The light spots in the video images are detected through LOG(Laplacian of Gaussian)operator processing and threshold-filtering.They are subsequently tracked by a Kalman filter and each light spot is marked according to its attributes which determines whether the light spot should be accepted or discarded.The accuracy of the method is verified with actual marine surveillance video in nighttime.Improvement of the detection and tracking of ships is demonstrated.
作者
王群朋
张金水
赵银祥
张明阳
WANG Qunpeng;ZHANG Jinshui;ZHAO Yingxiang;ZHANG Mingyang(School of Navigation,Guangzhou Maritime University,Guangzhou 510725,China;Intelligent Transport Systems Research Center,Wuhan University of Technology,Wuhan 430063,China;National Engineering Research Center for Water Transport Safety,Wuhan University of Technology,Wuhan 430063,China;School of Engineering,Aalto University,Espoo 20110,Finland)
出处
《中国航海》
CSCD
北大核心
2020年第4期52-58,共7页
Navigation of China
基金
国家自然科学基金(52071247,52071091)。
关键词
船舶跟踪
光斑检测
卡尔曼滤波
夜间监控视频
ship tracking
light spot detection
Kalman filter
nighttime surveillance video
