中文摘要：

针对传统齿轮检测仪器的检测速度慢、准确度低等问题,提出了齿轮缺齿缺陷的机器视觉检测方法。首先对齿轮图像进行灰度化、滤波、二值化、形态学运算等预处理操作,剔除背景图像的影响,得到清晰的齿轮目标;然后对齿轮轮廓进行检测,进而对齿轮特征点进行识别,提取位于轮齿上的点,实现齿数的自动检测以及缺齿缺陷的准确识别。

英文摘要：

Aiming at the problems such as slow detection and low accuracy of traditional gear detection equipment,a gear tooth defect detection method based on machine vision is put forward in this paper.Firstly,a series of image preprocessing operations such as grayscale processing,filtering,binarization,and morphological processing are implemented,excluding the background image,and the clear gear target is obtained.Then,the gear contour is detected and the gear feature points are identified.Extract the points on the teeth,and the automatic detection of the number of teeth and the accurate identification of missing teeth are achieved.