基于图像处理的小口径球面光学镜片划痕检测方法研究
Research on Scratch Detection Method for Small Aperture Spherical Optical Lenses Based on Image Processing
饶繁星 1李浩然 1向玉鹏 1曾美祖 2杨小品1
作者信息
- 1. 南昌工程学院电气工程学院,江西 南昌 330099
- 2. 江西天络光电科技有限公司,江西 上饶 334000
- 折叠
摘要
为了实现小口径球面玻璃光学镜片的缺陷检测,利用Ostu阈值分割+Hough变换的方法来自动检测镜片划痕.首先搭建了全自动视觉检测系统,通过暗场照明的方式获得镜片划痕和背景最大分离化的镜片表面图像,对该图像进行阈值分割,得出图像的灰度图,在灰度图的基础上通过Hough变换提取出划痕图像后进行划痕长度计算.实验结果表明所用的方法能够有效地检测出划痕并能准确计算出划痕长度,为镜片划痕的自动检测判断提供了理论和实践基础.
Abstract
In order to achieve defect detection of small aperture spherical glass optical lenses,this article uses the Ostu threshold segmentation+Hough transform method to automatically detect lens scratches.Firstly,a fully automatic visual in-spection system was constructed.The lens surface image with maximum separation of lens scratches and background was obtained by means of dark-field illumination.The image was thresholded and segmented to obtain the grey-scale map of the image,and the scratch image was extracted based on the grey-scale map by means of the Hough transform for the calculation of the length of the scratches.The experimental results show that the method used in this paper can effectively detect scratches and accurately calculate the length of scratches.
关键词
球面镜片/划痕检测/图像处理/Hough变换Key words
spherical lens/scratch detection/image processing/Hough transform引用本文复制引用
出版年
2024
NETL
NSTL
万方数据