摘要
随着我国社会的迅猛发展,地下矿山安全问题受到国家和监管部门的高度关注。以川西中咀铜矿为例,探讨基于AI视频智能辅助监管系统在高海拔地区地下矿山的应用,以增强中咀铜矿地下矿山的在线应急响应能力和预警能力。该系统由9个监测点构成,包括3个位于2600 m运矿隧道的监测点和6个井巷工程监测点。系统集成了9种算法,包括未佩戴安全帽识别、未佩戴自救器识别、烟火检测、巡检人员识别等,实现了对摄像头视频内容的在线实时分析,并对检测到的异常情况发出报警信号。在高海拔地区地下矿山的应用实践中,该系统对异常情况的监测准确率超过95%,并且随着系统运行时间的延长,通过算法的持续优化,监测准确率有望进一步提升。
With the rapid development of our society,underground mine safety issues have received great attention from the state and regulatory authorities.Taking the Zhongju Copper Mine in western Sichuan as an example,this paper explores the application of AI video intelligent auxiliary supervision system in underground mines in high-altitude areas to enhance the online emergency response and early warning capabilities of the Zhongju Copper Mine underground mine.The system consists of nine monitoring points,including three monitoring points located in the 2600 m ore transportation tunnel and six shaft and drift engineering monitoring points.The system integrates nine algorithms,including recognition of not wearing a helmet,recognition of not wearing a self-rescuer,smoke and fire detection,patrol personnel recognition,etc.,to achieve online real-time analysis of camera video content and send alarm signals for detected abnormal conditions.In the application practice of underground mines in high-altitude areas,the system’s monitoring accuracy for abnormal conditions exceeds 95%,and as the system runs longer,through continuous optimization of the algorithm,the monitoring accuracy is expected to be further improved.
作者
何建军
刘洋
武尚荣
陈帮洪
HE Jianjun;LIU Yang;WU Shangrong;CHEN Banghong(Sichuan Liwu Copper Co.,Ltd.-Zhongju Copper Mine,Ganzi Sichuan 626700,China;Kunming Engineering&Research Institute of Nonferrous Metallurgy Co.,Ltd.,Kunming Yunnan 650051,China)
出处
《有色金属设计》
2025年第3期47-51,92,共6页
Nonferrous Metals Design
关键词
AI
地下矿山
高海拔地区
灾害预警
监控监测
AI
Underground mine
High-altitude area
Disaster early warning
Surveillance and monitoring
