Research On Surface Defect Detection System Of Bearings Inner Ring Based On Machine Vision

With the development of science and technology,more and more automatic equipments are needed in the industrial field.Bearings are indispensable parts in mechanical equipment.In the production process,if the bearings are defective,the quality of products will be affected.At present,most of the methods of bearing inspection are mainly based on manual visual inspection.With the continuous improvement of product quality requirements,the traditional detection technology has been unable to meet the requirements of online testing.As a new nondestructive testing method,machine vision technology is more and more widely used in the field of bearing appearance inspection.In this thesis,a set of vision inspection device for bearing inner ring's defect is designed and realized in combination with the quality inspection standard set by bearing manufacturer.In the hardware scheme,the lens,camera,light source and other related hardware are firstly selected.Then,the lighting system is designed according to the characteristics of the bearings inner ring's defects.Finally,a complete image acquisition platform of bearing inner ring is built.In the software scheme,it mainly includes two aspects.One is the design and implementation of software.The other is the design and implementation of image processing algorithm for bearing inner ring.In the design of software,the parallel control of multi station camera is designed,and the data fusion is carried out with multi station data,and the result is fed back to the user.The whole set of software is based on Visual Studio,and the multithreading technology is used to solve the problem of camera parallel control,and an effective and feasible scheme is proposed for data synchronization and inter thread communication.In the image processing section,the acquired image is firstly preprocessed.The preprocessing process includes the extraction of the region of interest,the localization of the bearing inner ring,and the de-noising.Then,the local threshold method is used to extract the defects and filter them according to the characteristics of the defects.Finally,after the processing result is obtained,the result fusion is carried out through inter thread communication.The results are displayed on the human-computer interface,and communicate with the lower computer to drive the mechanical devices to screen.Through the field test,the result shows that this system can effectively detect surface defects of bearing inner ring.It has good real-time performance.This system can effectively detect the defects of bearing inner ring instead of traditional detection methods.