The Design Of System Of Defect Detection Of Spark Plug Gasket Based On Machine Vision

Spark plug gasket defects will cause the cylinder leak, misfire, and result in the engine shake, lack of power. Leakage in cylinder is of great danger to engine and people, so the defect detection of sealing gasket is very important.The present study studies the systematical design of defect detection of gasket based on machine vision, which is situated in the defect detection of Torch’s gasket. This system has the characteristics of non-contact, high accuracy, high speed and so on which mainly includes the hardware of vision and the software of image processing algorithm. Through their combination, the functions of automatic collection, transmission, processing and result feedback of the front, back and surface of gasket can be realized.Firstly, based on the analysis of the origin of the design and the current situation of related technology at home and abroad, two solutions of location detection are decided. Evaluating the parameters of the camera and lens according to the specific needs, selecting the photosource according to the workpiece properties to be detected to decide the lighting mode, and then selecting the industrial computer, image acquisition card and photoelectric switch devices so as to complete the design of the hardware system. Secondly, putting forward the image-processing ideas according to the overall image and the defect characteristics, and designing the procedures according to the idea, introducing and contrasting the other different algorithms that have the same function, then ultimately choosing the median filtering, gray-scale stretch, matching positioning based on gray, the least squares fitting circle, mask and Sobel operator splitting spots to complete the entire image processing algorithm. Finally, this paper introduces the realization of the function of software, including software interface and the implementation of the above algorithms based on OpenCV.The study presents a specific discussion of the designing processes of the software and hardware, debugs all parts of the software process and shows the final image to extract image defects. Large quantities of workpiece tests proved the feasibility of the system.