摘要
提出了一种基于图像处理技术的拉链次品检测方案,采用直方图均衡和阈值分割技术处理图像,对处理后的图像进行识别和判断.实验证明该方案可以准确地检测拉链齿数,并能有效解决拉链生产过程中多齿、少齿缺陷的自动检测问题.
A new scheme about zipper automatic detection based on image processing is put forward in this paper. In this scheme, equalizing the histogram and thresholding are adopted. The detection result can be gotten through handling black-and-white image. Experiments indicate this scheme can both exactly detect the number of teeth in the zippers with low mistake check rate and can solve the automatic detection problem about tooth-missing drawback in the production process.
出处
《汕头大学学报（自然科学版）》
2006年第1期65-69,共5页
Journal of Shantou University：Natural Science Edition
关键词
图像处理
直方图均衡
阈值
image processing
histogram equalization
threshold value
