Coal mine conveyor monitoring video enhancement algorithm based on DP-YOLOv7

With the advancement of intelligent construction in coal mines, the intelligent technology level of belt conveyor as a key equipment of coal mine main transportation is increasing, and the application of video AI technology in the monitoring and inspecting system of coal mine belt conveyor has become a hot spot. Aiming at the problems encountered by video surveillance during the operation of belt conveyor, such as insufficient light, moisture and large coal dust pollution, a new DP-YOLOv7 video detection algorithm for coal mine belt conveyor is proposed based on YOLOv7 target detection algorithm. By introducing detail processing module (DPM) and low frequency enhancement filter (LEF) into PENet network, the detection effect in low light environment of coal mine is enhanced, and the coal dust and water vapor are denoised by domain adaptive algorithm. The application test results show that the video surveillance enhancement technology of coal mine belt conveyor based on intelligent video improves the effectiveness and reliability of intelligent video surveillance, realizes the intelligent monitoring and control of the belt conveyor.