Design And Research Of Belt Conveyor Foreign Object Detection System Based On Computational Vision

As a large coal producer,coal is one of the main energy sources in our country.The main transportation method of raw coal is belt conveyor transportation.In the transportation process,foreign matter is often mixed into the coal and input into the next process to cause equipment damage.In order to solve the problem that the existing coal belt conveyor foreign matter detection method cannot meet the current actual production needs,we used a highly reliable and easy-to-maintain computer vision method for detection.At the same time,I optimized the algorithm according to the workshop environment and the characteristics of the coal conveyor to enhance the efficiency and stability of the system.According to the characteristics of non-coal foreign objects,we first perform threshold processing on the input video frames to distinguish coal from non-coal foreign objects,and eliminate the interference of moving coal blocks on foreign object detection;Then the improved three-frame difference method is used to enhance the foreign body movement information to increase the accuracy of detection,and the background difference method background image is dynamically updated.The frame difference method and the background difference method are combined to fill the hole to obtain a complete foreign body contour,and perform morphological opening operations to eliminate White dot interference;Finally,edge detection is performed to obtain the complete contour of the foreign body,and then the position and information of the non-coal foreign body is obtained through the contour detection.Experimental results show that the improved algorithm of fusion frame difference method and background difference method combines the advantages of the two algorithms,which is more accurate in detecting non-coal foreign objects than the traditional three frame difference method and background difference method.At the same time,according to the engineering requirements of non-coal foreign body detection on the belt machine,we developed a coal belt conveyor foreign body detection system based on the improved algorithm of fusion frame difference method and background difference method.The system can detect the belt conveyor in real time.After the foreign body is detected,the system can perform real-time detection of belt conveyors.After detecting foreign objects,the system will save relevant information to the database and the warning sound is played at the same time,prompting the staff to deal with it.