Development Of The Visual Inspection Device For The Surface Defects Of The Bearing Inner Sleeve

With the development of factory automation,the requirement for completeness of automation equipment in industrial field is higher and higher.As a necessary element in mechanical device,bearing may produce various defects in production process,which will cause serious impact on product quality.At present,most of the bearing manufacturers mostly use manual detection.A few enterprises use electromagnetic detection methods,such as environmental pollution and energy waste.As a new nondestructive testing method,machine vision technology has developed rapidly in the field of bearing surface detection.In this paper,the structure optimization and redesign of the surface detecting device for the inner sleeve of the four station bearing are carried out.Combined with the quality inspection standards of the bearing production enterprises,a single station testing system is designed and realized.This paper introduces the production process and common defects of the inner sleeve of the bearing,and puts forward the key and difficult points of the visual inspection system of the bearing inner sleeve.The integrated complex detection system of integration of light,machine and electricity is designed and realized.In the hardware design,according to the bearing manufacturers to draft detection requirements,select the appropriate industrial camera,lens,light source imaging equipment,design a new structure of lighting for side and inner surface,build a complete inner sleeve of the bearing defect detection experiment platform.In the software design,there are mainly five parts: system parallel software architecture,high-performance data buffer queue,MODBUS communication protocol based on VC environment,human-machine interaction interface based on MFC,and image processing algorithm of bearing sleeve.The work of writing,debugging and testing of the detecting system software and the design and implementation of the system motion control scheme are completed.In the software architecture,multilevel multithread structure is designed and line pool technology is applied.In the high performance data buffer queue,the non blocking synchronization of the data is realized by the chain combination queue.In the process of bearing inner sleeve image algorithm,the smoothing field algorithm is used to solve the problem of uneven image intensity.The location of detection targets is realized by the method of the largest class variance and the morphological method.The segmentation of the defects is realized by the improved frame difference method.Finally,the detection results are compared with the detection standard.Through many tests on the experimental platform,the test results show that the visual inspection system of the surface defects of the bearing sleeve can run steadily,efficiently and reliably.Both the detection efficiency and the detection precision are higher than the artificial visual method,which has a good application prospect.