›› 2016, Vol. 29 ›› Issue (11): 78-.

• 论文 • 上一篇    下一篇

基于HALCON的刹车片尺寸和表面缺陷检测系统

左东祥,陈晓荣   

  1. (上海理工大学 光电信息与计算机工程学院,上海 200093)
  • 出版日期:2016-11-15 发布日期:2016-11-24
  • 作者简介:左东祥(1992-),男,硕士研究生。研究方向:图像处理。陈晓荣(1974-),女,博士,副教授,硕士生导师。研究方向:图像处理等。

Size and Defect Detection System of Brake Based on HALCON

ZUO Dongxiang,CHEN Xiaorong   

  1. (School of Optical-Electrical and Computer Engineering,University of Shanghai for Science and Technology, Shanghai 200093, China)
  • Online:2016-11-15 Published:2016-11-24

摘要:

针对刹车片尺寸和表面缺陷人工检测效率低下的问题,文中提出了一种基于机器视觉软件HALCON的检测系统。该系统充分利用了刹车片的形态学和灰度值特征,运用边缘提取和局部阈值分割算法,定位出刹车片外边缘,最终计算出刹车片尺寸,并标记出缺陷位置。通过对200张图像样本进行测试,检测系统将尺寸误差控制在0.47%,刹车片缺陷检查的误判率为1%。该系统运行高效稳定,满足了实际需求。

关键词: HALCON, 边缘提取, 阈值分割, 尺寸检测, 缺陷检测

Abstract:

A detection system based on machine vision software named HALCON is proposed in the paper in face of the low efficiency of the size and imperfection detection of the brake. The system makes full use of the characteristics of the morphology and gray value of the brake, and uses edge detection and local threshold segmentation algorithm. The system locates position of margin, calculates the size of brake and marks the location of surface defect finally. According to the running result of the system with the input of 200 figure samples, it has a very high accuracy within 0.47% error, and the misdiagnosis rate of the defect is 1%.The system can run fast and stably, which meets practical requirements.

Key words: HALCON, edge extraction, threshold, size detection, defect detection

中图分类号: 

  • TP242.6+2