基于机器视觉的轮齿缺陷检测研究

Research on Gear Tooth Defect Detection Based on Machine Vision

针对传统的人工检测齿轮缺陷方法效率低、精度不高的问题,提出采用机器视觉的方法对齿轮轮齿进行检测。提出了一种图像模板的制作方法,实现了对齿轮轮齿的分割。并针对单个轮齿的磨损检测问题,分别采用轮齿面积、轮齿周长作为判断单个轮齿磨损的依据。结果表明采用轮齿面积作为检测标准效果更好,能够满足轮齿缺陷检测的需要,此方法是可行的。

Aiming at the problems of low efficiency and low precision of the traditional manual detection method for gear defects,the method of machine vision was proposed to detect the gear teeth.A method of making image template was presented to realize the segmentation of gear teeth.And for the wear detection of single gear tooth,the tooth area and the tooth circumference were used as the basis to judge the wear of a single gear tooth.The results show that the tooth area as the detection standard is more effective and can meet the needs of tooth defect detection,and this method is feasible.