There are many problems in the main transportation system of coal mine,such as manual spot inspection,remote centralized control can not be implemented,monitoring data is incomplete,intelligent control method is lacked,there is no efficient means of fault detection.In view of the above problems,this paper puts forward the method of segment extraction key frame inspection video analysis,forms the inspection video anomaly detection technology of coal mine main transportation,and applies it to the inspection robot system,the fault detection rate is improved.The intelligent control system of main coal flow transportation is designed,which realizes the intelligent speed regulation based on coal flow tracking and the one-button coal-flow startup function.The visual detection technology of conveyor belt tear based on vision and lidar fusion and the state monitoring and fault diagnosis technology based on voiceprint analysis of belt conveyor improve the stability of the main transportation system.