煤矿“智慧安监”系统架构设计与实现

Design and implementation of“intelligent safety supervision”system in coal mine

煤矿安全监督检查是对煤矿安全生产的日常监督与管理,旨在及时发现事故隐患并督促排除。传统的煤矿监查模式是视频监控系统与人工检查相结合,对隐患的识别依赖人工肉眼检测,费时费力,且不能实现自动识别与预警,已不能够满足当前智能化背景下对煤矿安全监管的新要求。因此在智能化矿山的基础上探讨基于深度学习、云计算和物联网等先进技术的煤矿智能化安监系统架构。运用深度学习技术自动识别分析视频中的不安全行为、状态及因素,全天候24小时智能监管,大幅减轻监控人员的工作负担,达到主动、准确、实时的监控要求,实现矿山生产管理过程中可感知、可预警、可追溯、可管理的安全监管,降低人为因素造成的误报或漏报情况。

Coal mine safety supervision and inspection is the daily supervision and management of coal mine safety produc-tion,aiming at timely finding hidden dangers of accidents and urging elimination.The traditional coal mine inspection mode is the combination of video surveillance system and manual inspection,and the identification of hidden dangers relies on manual naked eye detection,which is time-consuming and laborious,and can not realize automatic identification and early warning,can not meet the new requirements of coal mine safety supervision under the current intelligent background.Therefore,on the basis of intelligent mine,the architecture of intelligent mine safety monitoring system based on advanced technologies such as deep learning,cloud com-puting and Internet of Things is discussed.The use of deep learning technology to automatically identify and analyze unsafe behav-iors,states and factors in videos,24-hour intelligent supervision,greatly reducing the work burden of monitoring personnel,to achieve active,accurate and real-time monitoring requirements,and to achieve perceptive,early-warning,traceable and manageable safety supervision in the process of mine production management.Reduce false positives or omissions caused by human factors.