Non-uniform Illumination X-ray Image Correction Of Wire Rope Core Conveyer Belt And Study On The Method Of Joint Tap Detection

Wire rope core conveyer belt is the main transmission equipment of material transportation in modern production.The joints in the long distance conveyor belt often cause twitch due to the excessive external force,which can cause the fracture of the conveyor belt seriously,resulting in huge economic losses and even casualties.Therefore,in the field of conveyor belt,eliminating the influence of non-uniform illumination caused by signal acquisition system,real-time and accurate detection of joint twitch is a necessary measure to prevent accidents.In view of the existing engineering methods of signal acquisition and adaptive correction is imprecise and joint position detection is inaccuracy.In this paper,we improved from three aspects,including adaptive preprocessing of non-uniform illumination X light image,accurate location of joints and accurate detection of joint twitch.A new X light image processing and joint twitch detection method for steel cord conveyor belt has been designed.In view of many unfavorable factors exist in the process of image acquisition caused by the non-uniform illumination X light image,this paper analyzes the special texture characteristics of steel wire rope core conveyer belt,proposed the model based on the maximum uneven background theory,combined with the Retinex theory to calculate X light images after grayscale correction.Both theoretical analysis and experimental results verify that the above method can effectively remove the uneven illumination effect of X light image of steel cord conveyor belt,thereby indirectly improving the accuracy of joint recognition based on X image.Aiming at the inaccuracy of the calculation of the joint edge,a method of locating the joint edge based on contour and centroid calculation is proposed in this paper.The edge of the wire rope is detected by the Sobel edge detection operator,and the centroid of the joint edge is extracted as the joint point.In view of the traditional joint detection instability problem,this paper proposes a joint detection method based on the algorithm of BRIEF: firstly,linear fitting on the joint point after stratification,then matching the joint point of current image and the reference image,the joint point is regarded as the feature point,and the feature point descriptor is obtained by using the BRIEF algorithm to get the matching result.finally,comparison the joint points to their joint fitting a straight line distance,if the distance difference is greater than a certain threshold indicates that joint tap.The test results show that this method can effectively eliminate the uneven illumination effect,effectively improve the accuracy of wire rope joint recognition,and to detect the exact length of joint occurrence of tap,provides a powerful guarantee for the conveyor belt fault on-line monitoring system for real-time monitoring of the tic joint.