视觉与AIS融合的桥区水域船舶自动监测方法  

Automatic ship monitoring method in bridge area by fusion of vision and AIS

作　　者：杜子俊 贺益雄[1,2] 于德清 赵兴亚 张锐 黄立文[1,2] DU Zijun;HE Yixiong;YU Deqing;ZHAO Xingya;ZHANG Rui;HUANG Liwen(School of Navigation,Wuhan University of Technology,Wuhan 430063,China;Hubei Key Laboratory of Inland Shipping Technology,Wuhan University of Technology,Wuhan 430063,China)

机构地区：[1]武汉理工大学航运学院,湖北武汉430063 [2]武汉理工大学内河航运技术湖北省重点实验室,湖北武汉430063

出　　处：《中国航海》2025年第1期34-42,共9页Navigation of China

基　　金：国家自然科学基金面上项目(52071249)。

摘　　要：为保障桥区通航安全,提出一种视觉与船舶自动识别系统(Automatic Identification System,AIS)融合的船舶自动监测方法。基于YOLOv5(You Only Look Once version 5)目标检测算法和Canny算法提取船舶图像轮廓信息,构建桥区水域目标距离、方位和高度视觉测量模型与方法,实现船舶三维定位。利用融合视觉与AIS的船舶航行态势数据建立异常行为检测模型,自动识别、监测桥区水域危险船舶。试验结果表明:在单、多船的情况下视觉与AIS数据关联准确率分别达到98.45%、91.29%;能有效监测桥区船舶的运动状态。本研究可为保障船舶和桥梁的安全提供有效方法。To ensure the safety of navigation in the bridge area,this paper proposes a ship automatic monitoring method based on the fusion of vision and AIS(Automatic Identification System).The ship contour information in the image is extracted by the YOLOv5(You Only Look Once version 5)target detection algorithm and the Canny algorithm.A distance,azimuth,and height measurement model of the visual target in the bridge area is constructed to achieve the three-dimensional positioning of the ship.An abnormal behavior detection model is established using the ship navigation situation data from the fusion of vision and AIS to automatically identify and monitor monitoring of dangerous ships in the bridge area.The experimental results show that:In cases of single and multiple ships,the accuracy of visual and AIS data association is 98.45%and 91.29%,respectively;The method can effectively monitor the motion state of ships in the bridge area.This paper provides an effective method for ensuring the safety of ships and bridges.

关 键 词：船舶自动监测方法 目标检测 数据融合 异常行为检测 

分 类 号：U675.96[交通运输工程—船舶及航道工程]