针对杜儿坪煤矿井下带式输送机常发生的大块物料运输、堆煤和人员入侵导致的皮带撕裂、机架损坏以及人身安全等问题,基于视觉图像分析的安全防护检测系统进行研究。针对三种异常状态综合采用深度学习目标检测和图像特征分析等算法,以防爆摄像机、交换机、控制卡和工控机为硬件核心实现系统功能,通过实际应用效果检验,系统能够满足井下带式输送机安全防护需求。

In view of the problems often occurred in the downhole belt conveyor of Duerping Coal Mine, such as belt tearing, frame damage and personal safety, etc. caused by large material transportation, coal piling and human eye intrusion, research on security protection and detection system is conducted based on visual image analysis. Aiming at the three abnormal states, algorithms such as deep learning object detection and image feature analysis are comprehensively adopted, and taking explosion-proof cameras, switches, control cards and industrial control computers as the hardware cores to achieve system functions. Through the test of practical application effect, the system can meet the safety protection requirements of downhole belt conveyors.