多方向Scharr的齿轮边缘检测与中心定位
2019-02-19
作者：李浩谊1，马春庭1，唐秀媛2
单位：1中国人民解放军陆军工程大学石家庄校区；232183部队
摘要：针对齿轮精度检测设备结构较复杂、使用和维护较麻烦等问题，提出基于多方向Scharr的齿轮边缘检测与中心定位方法。通过高斯滤波减小噪声的影响，采用分别代表0°、22.5°、45°、67.5°、90°、112.5°、135°、157.5°共8个方向的5×5算子模板进行边缘检测，结合Otsu方法得到边缘二值化图像，并结合最小包围矩形法、最小包围圆形法和重心法求出齿轮中心坐标。实验结果表明，该算法能够准确快速地提取齿轮边缘和中心信息，在一定程度上减少了噪声的影响。
关键词：边缘检测与中心定位；多方向Scharr；算子模板；中心坐标
中图分类号：TG86；TH162文献标志码：ADOI：10.3969/j.issn.1000-7008.2019.02.031
Scharrbased Edge Detection Algorithm for Multidirection Template
Li Haoyi，Ma Chunting，Tang Xiuyuan
Abstract:Since gear precision measuring equipment exists the complicated structure and difficulty in using and maintaining，an improved edge detection algorithm of multidirection Scharr operator template in gear edge detection and center location is proposed.Firstly，Gaussian filter is applied to reduce the effect of noise.The algorithm based on eight 5×5 size templates of 0°，22.5°，45°，67.5°，90°，112.5°，135°，157.5°different direction is used to detect edges with the Otsu method.And the minimum encircling rectangle method，the minimum enclosing circular method and the gravity method are used to acquire the gear center coordinate.The results show that the improved algorithm has advantages of accurate and fast edge detection and center location，and reduction of the noise effect to a certain extent.
Keywords:edge detection and center location；multidirection Scharr；templates；center coordinate