Research On Automatic Optical System Of The End Face Detection Of Spark Plug

The spark plug,commonly known as “the fire mouth”,Its role is to ignite the mixed gas in the cylinder.This dissertation describes is : gasoline leaks caused by the defect of the end face of the spark plug.If the defects are too large,the spark is weak,and prone to leakage of the phenomenon.Moreover,when the defects are too large,the bigger pressure required to boost the air,and the engine is hard to start,on the highway prone to “leakage of fire” phenomenon.At present,the Automatic Optical Inspection technology is not applied to the spark plug end detection whether at home or abroad.In this dissertation,based on the actual needs of spark plug detection,a fully automatic optical detection system based on machine vision is designed.And the related theories and key technologies of AOI system design and defect recognition and classification of spark plug end face defects are deeply studied.The main achievements and relative contents are listed below:(1)According to the characteristics of the end face of the spark plug and the scene of the scientific design of a set of automatic optical detection system,the use of surface array camera,high-light super-depth lens with coaxial light source composed of image acquisition system,with high precision hardware movement system,Improve the quality of drawing.The experimental results show that the proposed system can capture the clear image of the end face of the spark plug at the speed of 1 m/s,and the distortion is small enough to meet the demand of industrial production.(2)According to the characteristics of the flushing plug end face defect,a parallel algorithm for the end face of the spark plug has been designed to deal with the acquired image.For the first time,the algorithm of morphological feature extraction,morphological filtering and local threshold segmentation are applied to detect the defects of spark plugs.The experimental results show that the detection rate of the algorithm is as high as 97%,which can not only meet the requirements of industrial production,but also the detection period is less than 1 s to meet the demand of industrial online production.(3)Realize the design of the system software part,design a set of Qt4-based graphical interface and image processing algorithm software part,to achieve the online image acquisition real-time display and defect statistics,including defective imagepreservation and location mark,The total number of defects and the type of statistics and the export of statistical results and other functions;(4)To achieve the whole system production line real-time online testing and testing.According to the results of our main database,the image processing technology is first reported to apply to the detection of defects at the end of the spark plug.