Piston Surface Defect Recognition Method Based on Image Processing

Abstract

With the rapid development of the automobile industry, great changes have occurred in the engine manufacturing industry, which not only needs mass production, but also needs to pay attention to rapid production. As the main working part of the engine, the importance of piston surface quality in engine production is beyond doubt. Machine vision detection technology collects the image of the target object through the industrial camera, processes it with the image processing software to obtain the detection result, and the controller takes the corresponding operation. This technology has the characteristics of non-contact and high precision. The paper takes the piston as the research object, and the piston surface defect detection system based on machine vision is designed according to the needs of industrial production. Then, a set of piston surface defect nondestructive detection system based on machine vision is established. It mainly includes hardware and software module design. The hardware of the detection system consists of a light source, image acquisition card, industrial camera, optical lens, and testing platform. The piston image obtained by the industrial camera is preprocessed by gray conversion and median filtering, and then the gray histogram is used for threshold segmentation, edge detection, and morphological processing, and finally the area feature is used for defect detection. At the same time, the research results prepare for the classification of piston quality grade in the later stage. The research conclusion of this paper can provide a reference for solving the detection problem of piston surface defects.