Research On Fault Detection Algorithm Of Steel Cord Conveyor Belt Joints Based On Image Matchin

Belt conveyors have been widely used in coal,mining,electric power and other fields.Conveyor belt is the key component of belt conveyor traction and loading.The whole steel cord conveyor belt actually used in the belt conveyor is formed by overlapping several steel cord conveyor belts.The joint is the weakest part of the whole steel cord conveyor belt,which is prone to faults such as joint twitching and steel cord breakage,which may lead to belt breakage accidents.In order to improve the accuracy and automation of joint fault detection,this thesis uses the X-ray image of the steel cord conveyor belt joint to study the steel cord conveyor belt joint fault detection algorithm based on image matching.The main work is as follows:(1)A steel cord end point extraction algorithm based on iterative adaptive threshold is proposed.The local binary image of the non-joint area of the image is used to calculate the number of steel cords,then the convolution operation is performed with the image by the Scharr operator.The end point and its coordinates of the steel cord are determined by contour detection and extracting the center of the minimum circumscribed circle.The experimental results show that the algorithm can quickly and effectively extract the end points of the steel cord.(2)A steel cord end matching algorithm based on the improved ORB algorithm is proposed.The FAST feature point extraction algorithm of ORB algorithm is replaced by the steel cord end point extraction algorithm to form the pattern of "steel cord end point extraction algorithm + BRIEF description + Hamming distance matching",and the length and slope constraints are used to remove the false matching to obtain a reliable matching point pair,and the matching optimization method is used to match the unmatched end points.The experimental results show that the algorithm can achieve one-to-one correspondence between the end points of the steel cord.(3)An automatic detection algorithm for joint faults of steel cord conveyor belts based on image matching is proposed.Aiming at joint twitching faults,the reliable matching point set of the steel cord ends is used as a reference point to calculate the twitching amount of the upper and lower ends of each steel cord,the average twitching amount of the upper and lower ends of the steel cord and the overall twitching amount of the joint;for the steel cord fracture failure,the difference in the number of end points between two adjacent pairs of reliable matching points is used to realize the detection of the steel cord fracture failure.Experimental results show that the algorithm realizes the accurate,automatic and comprehensive detection of joint failures,as well as the identification of abnormal steel cord end points and steel cord fracture areas.The automatic detection algorithm for joint failures of steel cord conveyor belts based on image matching proposed in this thesis is a further improvement of the existing functions of the X-ray-based non-destructive testing system for steel cord conveyor belts.It can realize automatic detection of joint twitching and steel cord fracture failures,prevent belt breakage accident,achieve the goal of "reducing staff through automation ",improve production efficiency,and have significant economic and social benefits.