基于机器学习的刀具表面缺陷检测及分类方法

刀具在生产的过程中,由于人员、机器、环境等多方面原因,刀具的表面会出现各种缺陷,如划痕、碰撞凹坑、涂层剥落和边缘豁口;这些缺陷会严重影响刀具的质量和外观,对于刀具的缺陷检测,目前主要采用人工目检的方式,人工检测方法效率和准确率都比较低;为解决上述问题,提出一种刀具缺陷的自动化检测及分类算法;针对刀具图像的预处理,提出了一种基于双边滤波的降噪方法和基于差分的对比度增强算法;对于刀具的缺陷检测任务,提出了基于图像差分的缺陷检测算法;对于缺陷的分类任务,提出了一种基于SVM的分类算法,即通过提取缺陷区域的形状、纹理等特征来训练SVM分类器;最后对提出的缺陷检测及分类算法进行实验,结果表明算法的缺陷检出率达97.2%,分类准确率可达94.3%;算法能够很好地满足工业需求,可以替代人工实现刀具缺陷的自动化和高效率检测。

Tool surface defect detection and classification based on machine learning

In the process of tool production, due to personnel, machinery, environment and other reasons, the surface of the tool will appear a variety of defects, such as scratches, impact pits, shedding and edge break. These defects will seriously affect the quality and appearance of the tool, for the tool defect detection, the current main method is manual visual inspection, manual detection method efficiency and accuracy are low. In order to solve the above problems, an automatic tool defect detection and classification algorithm is proposed. For tool image preprocessing, a noise reduction method based on bilateral filtering and contrast enhancement algorithm based on difference are proposed. For tool defect detection task, a defect detection algorithm based on image difference is proposed. A classification algorithm based on SVM is proposed for defect classification task. The SVM classifier is trained by extracting the shape, texture and other features of the defect area. Finally, the experiment of the proposed defect detection and classification algorithm is carried out, and the results show that the defect detection rate of the algorithm is 99.7%, and the classification accuracy is 95%. The algorithm can meet the needs of industry and replace the manual to realize the automation and high efficiency detection of tool defects.