摘要
由于河道采砂管理难度大,技术手段不足,监管人力缺乏等问题,致使非法采砂难以控制。针对河道非法采砂流动性大、隐蔽性强、取证难等特点,研究基于深度学习的图像识别技术,以提高采砂船识别及监管的能力,突破采砂监管中的重点、难点,结合智能追踪技术,实现对非法采砂行为的及时发现和预警,并形成足够的证据链来证实非法采砂的事实,对提高河道采砂监管效能具有重要意义。
The management of sand mining in rivers is very difficult due to lack of technical means and supervision manpower,which makes illegal sand mining uncontrollable.Against mobility,concealment and difficulty in obtaining evidence of illegal sand mining,image recognition technology based on deep learning is researched to improve the identification and supervision ability of sand dredges and break through the key points and difficulties in supervision.In combination with intelligent tracking technology,timely detection and early warning of illegal action is realized,and enough evidence chain is formed to confirm the fact.Therefore,the application of these technology has great significance for improving the efficiency of sand mining supervision.
作者
汤文华
陈灿斌
向舒华
申寒冬
TANG Wen-hua;CHEN Can-bin;XIANG Shu-hua;SHEN Han-dong(Hubei Water Resources Research Institute,Wuhan 430070,Hubei Province,China;Xiamen Xingkangxin Polytron Technologies Inc,Xiamen 361016,Fujian Province,China;Hubei Water Resources and Hydropower Science and Technology Promotion Center,Wuhan 430070,Hubei Province,China)
出处
《中国农村水利水电》
北大核心
2021年第5期108-112,共5页
China Rural Water and Hydropower
基金
湖北省省直部门预算项目(2019-218-006-002)。
关键词
非法采砂
深度学习
图像识别
船舶分类
智能跟踪
illegal sand mining
deep learning
image recognition
ship classification
intelligent tracking
