Research on Surface Defect Detection of the Connecting Rod Based on Machine Vision

Abstract

In recent years, with the development of the automobile industry increasingly rapid, the engine production has also undergone tremendous changes, which not only requires mass production, but also pays attention to rapid production. As the power transmission link between engine crankshaft and piston, the importance of connecting rod quality in engine production is beyond doubt. In order to meet the requirements of engine batch and rapid production, machine vision inspection technology is essential in engine production. This paper takes the connecting rod image as the research object. First, according to the requirements of industrial production, the hardware device of the connecting rod surface defect system is designed. Then a set of automatic detection systems based on machine vision is designed to detect the surface defects of the connecting rod without damage. The system uses CCD camera to improve the speed of the detection system and reduce the performance requirement, making it easier to realize defect detection under existing conditions. The image segmentation threshold is automatically selected, and the connecting rod information is extracted from the image according to the actual threshold value, and the information in the connecting rod image is obtained to realize the defect detection. Based on the connecting rod image obtained by camera, the image noise is removed, and the defect on the connecting rod surface is segmented by binarization with the threshold value automatically selected to realize the automatic defect detection.