Research On Defect Detection Technology Of Soft Magnetic Ring Based On Machine Vision

Soft magnetic ring is a kind of magnetic material widely used in modern communication,automatic control,automotive electronics,aerospace and other fields.Magnetic ring is easy to be damaged in production.At present,the defect detection of magnetic ring mainly relies on manual labor,which has low efficiency and high cost.Therefore,this paper studied the algorithm about size measurement and defect detection based on machine vision technology and developed a set of on-line defect detection equipment optical detector for soft magnetic ring,called optical inspection machine.This paper mainly studied from the following aspects.Based on the principle of visual imaging,the hardware of the imaging system was selected.The lighting scheme of each detection station was worked out and the lighting experiment platform was built.Based on Opencv which is an open source library and QT platform,an image processing interface program for optical detector was developed.In the end,the automatic process of image acquisition,image analysis,camera focus and the blanking was realized.In the pre-processing part,contrast enhancement,filtering,segmentation and morphological processing were performed on the collected images,which reduced the difficulty of subsequent defect feature extraction.In the dimension measurement part,an improved fitting circle algorithm is proposed to measure the inner and outer diameter of the magnetic ring,and the upper and lower edge lines are extracted by setting a ROI region to complete the height measurement of the magnetic ring.In the defect detection part,a multi-ring region classification detection algorithm was designed for the defects on the upper and lower surfaces of magnetic rings.A defect detection algorithm based on mask technology and contour detection technology was designed for the defects on the inner and outer surfaces of magnetic rings.The optical inspection machine 's defect detection system has a accuracy of 97% for the10 mm magnetic ring parts.For others with different sizes,the accuracy rate is also more than96% by adjusting the parameters,which effectively reduced the enterprise cost,improved the detection efficiency,and also made the whole soft magnetic ring industry chain more perfect.Finally,a large number of experiments have proved that the defect detection algorithm developed in this paper has fast detection speed,high accuracy,stability,robustness,and high accuracy,which meets the needs of industrial inspection well.