Design Of Surface Inspection System For Car Carbon Brush Based On Machine Vision

This subject takes motor carbon brush used on the car as the object of the research. The performance of the pros and cons of the brush will directly affect the motor running stability and efficiency. Because carbon is a key component of most of the DC motor, the accuracy and efficiency of online detection during the production process must be guaranteed.In various industrial applications especially automotive micro motor carbon brush is an important and indispensable parts, particularly extensively used. Therefore, once there are defects on the surface, it will cause the car stop or even cause serious safety accidents. In order to reduce the occurrence of safety accidents, the quality of carbon brush in production process must be strictly controlled to make sure the detection will be done accurately and bad brush can be removed.Through contrasting and analyzing the advantages and disadvantages of detection methods on surface defect home and abroad. Finally, a real-time, non-contact, accurate detection of carbon brush, the surface quality of solution is given. The solution is based on the rapid development of intelligent machine vision, the utility model has the advantages of non-contact, rapidness and accuracy and has strong anti-interference ability in the electronic process. So it is applied more and more widely in a variety of industrial production.In order to achieve fast and accurate detection of machine vision defect brush effect, this paper has done the following work:First by comparing the human eye detection, traditional online detection with intelligent vision system for non contact detection of their respective characteristics, shortcomings and deficiencies of the human eye detection and traditional online detection are analyzed and using the intelligent vision system with improved hardware device to realize non-contact online 100% efficient detection brush automatic scheme is put forward.According to the requirement of the brush surface crack detection system, mainly including the accuracy requirements, requirements and efficiency requirements of fault detection rate and false negative rate, we make comprehensive design and selection for the hardware part of the whole detection system, based on the stability and reliability of the operation of the system to consider for each link parts of reasonable analysis and selection. For visual system photo process in the presence of camera fretting on camera mounting bracket erection method is improved; in vibration and material feeding device of card data and the feed problem of slow increase the linear feeder, so as to ensure automatic feed material in a stable and fast; carbon brush is the opposite of crack detection in mechanical turnover mechanism with.Finally, according to the characteristics of the carbon surface crack defect, the algorithm and processing software using carbon brush for rapid detection of surface crack defects are designed.The schemes of surface carbon brush real time detection have been implement and we design install, and commission the online automatic detection device. A leading enterprise has used the technique and has achieved good results----determine all the cracks accurately and at real time and can guarantee the quality with less time and effort.At present domestic carbon brush enterprises rarely use machine vision on-line detection of surface defects, so the device has wide application value, meanwhile, the device through change detection algorithm can also be applied in the on-line detection of other precision metal products.