基于时空关系的井下安全生产AI图像识别系统的应用  

Application of AI Image Recognition System for Underground Safety Production Based on Space-Time Relationship

作　　者：权淑娟 QUAN Shujuan(Shanghai Information Technology Co.,Ltd.,Shanghai 200030,China)

机构地区：[1]上海煤科信息科技有限公司,上海200030

出　　处：《煤矿机电》2023年第2期17-21,共5页Colliery Mechanical & Electrical Technology

摘　　要：许多煤矿事故是由井下人员的违规违章行为或关键场所巡检不到位造成的,因此,井下人员行为识别和关键位置的有效巡查是保障煤矿安全生产的重要措施。系统依托工业物联网、大数据、人工智能图像识别等技术,以煤矿井下安全为核心,将“互联网+”理念与安全生产过程中的设备、人员等管控要素相结合,集成基于图像识别的井下安全生产系统。该系统可实现实时数据的采集与分析、智能联动报警与分级推送等功能,满足井下全方位安全监管的需要,大幅降低事故发生概率,为煤矿安全高效运行保驾护航。Many coal mine accidents were caused by violations of regulations by underground personnel or inadequate inspections of key locations.Therefore,identifying the behavior of underground personnel and conducting effective inspections of key locations were important measures to ensure coal mine safety production.The system relies on industrial internet of things,big data,artificial intelligence image recognition and other technologies,takes coal mine safety as the core,combines the“Internet plus”concept with equipment,personnel and other management and control elements in the safety production process,and integrates the underground safety production system based on image recognition.Real-time data collection and analysis,intelligent linkage alarm and hierarchical push functions were realized,the needs of comprehensive underground safety supervision were met,the probability of accidents was significantly reduced,and the safe and efficient operation of coal mines was ensured.

关 键 词：人工智能 图像识别 算法模型 煤矿井下安全 

分 类 号：TD67[矿业工程—矿山机电] TP18[自动化与计算机技术—控制理论与控制工程]