中文 | English

J. Hefei Univ. Tech. (Nat. Sci.)

合肥工业大学学报(自然科学版)

JOURNAL OF HEFEI UNIVERSITY OF TECHNOLOGY (NATURAL SCIENCE)

Competent Authorities: Ministry of Education of the People's Republic of China

Sponsor by: Hefei University of Technology

ISSN 1000-1646, CN 34-1001/N

Gear center positioning algorithm based on machine vision

Download Full Text

基于机器视觉的齿轮中心定位算法

Journal Information

J. Hefei Univ. Tech. (Nat. Sci.), Apr.2023, Vol.46 No.4: 433-437, 446

DOI: 10.3969/j.issn.1003-5060.2023.04.001

Authors

REN Yongqiang, LI Zhangzhu, LI Run

(School of Mechanical Engineering, Hefei University of Technology, Hefei 230009, China)

Abstract and Keywords

Abstract: The gear machining accuracy is affected by complex factors, which necessitates higher requirements for the accuracy measurement. The positioning accuracy of gear center (base circle center) is very important in the detection process. In order to reduce the impact of defects and noise on the positioning accuracy of gear center in industrial visual detection, this paper presents a method of tooth tip arc segmentation combined with improved random sample consensus (RANSAC) algorithm to locate the gear center. Canny operator is used to detect the edge of the collected gear image, Freeman code is used to track the edge, and the rough datum of the center of the circle is fitted by the least squares method. The tooth top arc is quickly segmented by comparing the distance between the contour point and the rough datum. The sub-pixel edge and more accurate threshold are used to accurately segment the arc, and the arc is fitted in combination with the RANSAC algorithm, so as to get the gear center. Experimental results show that the presented algorithm is accurate and applicable.

Keywords: edge tracking; arc segmentation; sub-pixel edge; random sample consensus (RANSAC); center positioning

作者信息

任永强，李掌珠，李润

（合肥工业大学机械工程学院，安徽合肥230009）

摘要和关键词

摘要: 齿轮加工精度影响因素复杂,精度测量要求高,齿轮中心(基圆圆心)的定位精度在检测过程中至关重要。为降低工业视觉检测中缺陷与噪声对齿轮中心定位精度的影响,文章提出一种齿顶圆弧分割结合改进随机抽样一致性(random sample consensus, RANSAC)算法定位齿轮中心。对采集的齿轮图像采用Canny算子检测边缘,使用Freeman码进行边缘跟踪;以最小二乘法拟合得到圆心粗基准,通过比较轮廓点到粗基准的距离快速分割出齿顶圆弧;采用亚像素边缘与更精确的阈值对圆弧进行精分割,并结合随机抽样一致性算法对圆弧进行拟合,得到齿轮中心。实验结果表明该算法具有较高的定位精度和较强的实用性。

关键词: 边缘跟踪；圆弧分割；亚像素边缘；随机抽样一致性（RANSAC）；中心定位

Funding

国家产业技术基础公共服务平台资助项目（2019-00899-2-1）

Next

Back to Contents