应用光学, 2017, 38 (6): 931, 网络出版: 2018-01-04  

色度亮度空间中的线束端子显微图像分割

Wire harness terminals microscopic image segmentation in chromaticity brightness space
作者单位
1 河南理工大学 计算机科学与技术学院, 河南 焦作 454000
2 广东省数据科学工程技术研究中心, 广东 广州 510631
摘要
为提取检测线束端子质量的相关数据, 提出一种在色度亮度空间中分割线束端子显微图像的算法。该方法利用线束端子显微图像的颜色分布特点, 用阈值法分割线束端子。通过分析线束端子显微图像的颜色特点, 把彩色图像转换至色度亮度空间。为了保证图像内部的颜色均匀, 对其中的色度向量进行滤波。借助获取色度向量的红色通道, 考虑红色通道直方图双峰夹一谷和双峰夹多谷情况下阈值并分割图像, 使用数学形态学方法处理二值图像获得内径轮廓。实验结果表明, 分割得到的端子内径轮廓周长偏差小于0.1%, 表明此方法可以准确地分割线束端子显微图像。
Abstract
In order to extract relative data for quality control of wire harness terminals, an microscope image segmentation for wire harness terminals in chromaticity brightness space was proposed. This method divided the harness terminal by threshold method using the color distribution characteristic of the harness terminal micrograph.After analyzing the color distribution character, the microscopic image was first converted to brightness chromaticity space, and to insure the color homogenization within the image ,the chromaticity vector was filtered. Then for the color channel, an optimal thresholding was determined for the two cases: the histogram with two peaks and one valley and the histogram with two peaks and two or more valleys. Finally, the mathematical morphology was performed on the binary image to produce the inner contour. The experimental results show that the circumference deviation of the inner diameter of the terminal is less than 0.1%, which indicates that the method can segment the harness terminal image completely and accurately.

芦碧波, 刘甜甜, 郭海儒. 色度亮度空间中的线束端子显微图像分割[J]. 应用光学, 2017, 38(6): 931. Lu Bibo, Liu Tiantian, Guo Hairu. Wire harness terminals microscopic image segmentation in chromaticity brightness space[J]. Journal of Applied Optics, 2017, 38(6): 931.

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!