On-line Detection And Sorting Of Appearance Defects Of Magnetic Materials Based On Machine Vision

“Rare Earth” is known as the “gold” of industry and is an indispensable strategic resource for modern industry.Magnetic materials,which are one of the "rare earth" products,also play a decisive role and are widely used in various industries.However,due to the relatively backward production process of magnetic materials,it has been unable to meet the needs of modern industrial production,and the upgrading of the magnetic material industry is imminent.Combining with the practical problems faced by enterprises,this thesis proposes an on-line detection and sorting system for appearance defects of magnetic materials based on machine vision.It uses machine vision detection technology and servo robot sorting technology to solve traditional manual quality inspection and manual handling problem.In this thesis,the difficulties of visual inspection of magnetic materials and the problems faced by enterprises are expounded in detail.The detection scheme based on machine vision is proposed and the feasibility of the scheme is demonstrated.Secondly,based on the proposed scheme,this thesis introduces the selection analysis of machine vision application equipment from industrial camera,industrial lens and industrial light source.According to the sorting requirements,the scheme of using the servo controller to control the six-axis robot is proposed,and the design details are introduced in detail.Then,based on the relevant hardware platform,this thesis focuses on the magnetic material appearance detection algorithm,which is divided into magnetic material appearance size measurement algorithm and magnetic material appearance defect detection algorithm.In addition,based on the hardware platform and algorithm,the upper computer software system and the lower computer software system are designed respectively.Finally,the experimental test platform was built using the existing resources of the laboratory,and the indicators of the system were tested and analyzed.After the system test,the detection accuracy can reach 0.1mm,the sorting robot can also perform the sorting action stably,and the system realizes the basic functions.However,some problems have been found in the actual test.On the one hand,the accuracy and stability of the detection algorithm need to be improved.On the other hand,the false positive rate of the system needs to be further reduced.