摘要
本文围绕海上无人值守平台周界安防技术需求,以800m范围内的快艇、渔船、客轮和商船为航行目标构建运动目标跟踪监控系统,并集成开发了基于水下被动声学检测和水上云台控制的光学视频系统为一体的声光一体化监控系统,设计了一套基于深度学习的目标检测算法,实现了对水面航行目标的跟踪检测与分类识别。本文在目前基于单一方法进行水面目标跟踪的基础上进行了突破,系统综合了水听器远距离目标监控跟踪和近距离的水面视频监控目标识别,当两者同时确认目标,由声警器发出警报驱离可疑目标,系统将水下声学监控和水面视频监控有效结合,极大地降低了平台的误报概率,同时实验测试结果表明该系统可有效实现对4类水面航行目标的跟踪检测和特征提取,且改进后的算法可有效提升目标检测效果与识别精度,具有较好的鲁棒性和有效性。
In order to protect the safety of the platform and the instruments and equipment on board, this paper focuses on the technical requirements of the perimeter security of the offshore unmanned platform, and takes speedboats, fishing boats, passenger ships and merchant ships within 800 m as the navigation targets to construct the moving target tracking and monitoring system, an acoustic-optical integrated monitoring system based on underwater passive acoustic detection and underwater pan-tilt control is integrated, and a target detection algorithm based on deep learning is designed, it realizes the tracking detection and classification recognition of the navigation target on the water surface.In this paper, a breakthrough has been made on the basis of tracking water surface targets based on a single method at present. The system integrates the hydrophone long-distance target monitoring and tracking and the water surface video monitoring target recognition in close range, the system effectively combines underwater acoustic monitoring with surface video monitoring, which greatly reduces the chance of false alarm of the platform, at the same time,the experimental results show that the system can effectively realize the tracking detection and feature extraction of four kinds of surface navigation targets, and the improved algorithm can effectively improve the target detection effect and recognition accuracy, it has better robustness and effectiveness.
作者
洪波
刘雪芹
秦志亮
马本俊
王飞
刘映锋
HONG Bo;LIU Xueqing;QIN Zhiliang;MA Benjun;WANG Fei;LIU Yingfeng(Wenzhou Marine Environment Inspection Center Station,State Oceanic Administration,Wenzhou 325035,China;Key Laboratory of Underwater Acoustic Technology,Harbin 150001,China;Key Laboratory of Marine Information Acquisition and Security Industry and Information Technology,Harbin Engineering University,Harbin 150001,China;School of Hydroacoustic Engineering,Harbin Engineering University,Harbin 150001,China)
出处
《海洋技术学报》
2021年第4期62-73,共12页
Journal of Ocean Technology
基金
声学科学与技术实验室基金资助项目(GK2050260214,GK2050260217,GK2050260218,KY10500180084,KY10500190031)
国家自然科学基金资助项目(41304096,41876053)。
关键词
水面目标
检测
跟踪
特征提取
一体化监控
water surface target
monitoring
tracking
feature extraction
acoustic-optical integrated monitoring system
