摘要 针对员工对油封缺陷进行肉眼检测时存在检测效率低、易遗漏、成本高等不足,提出一种基于图像处理的油封缺陷在线检测方法。通过图像的采集系统采集油封图像,先对油封图像进行预处理,包括灰度处理、中值滤波,然后通过Otsu阈值分割和链码法提取油封边缘轮廓特征,采用最小二乘法拟合油封唇口圆形轮廓来确定样件是否存在缺陷,根据检测结果将有缺陷的样件筛选出来。实验结果表明,该检测系统能够在线完成产品缺陷检测,该方法具有检测成本低、可靠性高、实时性好、容易在线实施等优点。 In this paper,for the visual inspection of oil seal defects by employees,there are some inefficiencies such as low detection efficiency,easy omission and high cost.An online detection method for oil seal defects based on image processing is proposed.The oil seal image is acquired by the image acquisition system,and the oil seal image is preprocessed,including gray processing and median filtering.Then,the edge contour feature of the oil seal is extracted by Otsu threshold segmentation and chain code method,and the oil seal lip is fitted by least squares method.The circular contour is used to determine whether the sample is defective,and the defective sample is screened according to the detection result.The experimental results show that the detection system can complete product defect detection online.The method has the advantages of low detection cost,high reliability,good real-time performance and easy implementation on line.
机构地区 长春理工大学机电工程学院
出处 《长春理工大学学报:自然科学版》 2021年第2期48-54,共7页 Journal of Changchun University of Science and Technology(Natural Science Edition)
基金 吉林省科技厅重点项目(20190302069GX)。
关键词 油封 图像处理 阈值分割 最小二乘法 缺陷检测 oil seal image processing threshold segmentation least squares method defect detection
