Research On The Wheel Model Automatic Identification System Base On Template Matching

In recent years,with the rapid development of China’s economy,the automobile industry has been rapidly developed,and also led to the rapid expansion of wheel production enterprises.The traditional manual identification of the wheel type is not able to meet the requirements of modern industrial development due to its low degree of automation and high labor costs.In order to improve production efficiency and reduce labor cost,this paper has carried out in-depth research on some key technologies in the automatic identification system of wheels,and designed a wheel model automatic identification system based on template matching.Firstly,this paper studies the overall architecture of the wheel model identification system,and discusses the hardware selection of the machine vision system and the control system.The machine vision system mainly includes the selection of hardware such as light source,industrial camera,lens and industrial computer.The control system mainly includes PLC model selection.The PLC control system flow is designed to realize the automatic sorting function of the wheels on the assembly line.Secondly,this paper deeply studies the template matching algorithm,and uses ring projection as the template matching image feature to achieve fast template matching of arbitrary rotation angle.The algorithm can quickly locate the position of the target in the image to be identified.But because of the algorithm’s accuracy is poor,this paper proposes a radial projection algorithm.Using radial projection as the template image feature can improve the matching accuracy and obtain the rotation angle of the target.In addition,in order to apply template matching to various complex situations,this paper proposes a multi-scale template matching algorithm to achieve template matching in template scaling.Based on the research of region partitioning algorithm,this paper implements template matching for arbitrary polygons.Finally,based on the research of machine vision system and template matching algorithm,this paper designs a software system for automatic identification of wheel models.This paper proposes two algorithms for quickly locating the center and radius of the hub in the image,which greatly increases the speed of template matching.The two features of ring projection and radial projection are used as image features of the template wheel to determine the final wheel model,which guarantees the accuracy of the wheel model identification results.