The research on the detection method of belt deviation by video in coal mine

Belt deviation is the most common failure during the operation of belt conveyor, and may cause tremendous harmfulness. In order to improve the intelligence of traditional video monitoring system of belt in coal mine, according to the deficiency of traditional system, we deeply studied various types of edge detection algorithm, and proposed an improved algorithm to achieve the edge extraction based on canny operator and wavelet packet algorithm, and besides compared with the classical edge detection algorithms. We also adopted an improved Hough transform (HT) for line detection and realized the detection of belt deviation by the parameters of line. Finally, the experimental results showed that this system has advantages of high accuracy, good real-time performance and strong anti-jamming capability.