微小型结构件显微图像边缘的自动识别

姜黎; 吴伟仁; 张之敬; 金鑫; 节德刚

doi:10.3788/OPE.20132101.0224

您当前的位置：

首页 >

文章列表页 >

微小型结构件显微图像边缘的自动识别

信息科学 | 更新时间：2020-08-12

- 微小型结构件显微图像边缘的自动识别
- Automatic detection of micro image edges for micro accessories
- 光学精密工程 2013年21卷第1期页码：224-232
- 作者机构：
  
  北京理工大学机械与车辆工程学院2. 国防科工委月球探测工程中心
- 作者简介：
- 基金信息：
- DOI：10.3788/OPE.20132101.0224
  中图分类号： TP391.4
- 收稿日期：2012-06-26，
  
  修回日期：2012-09-17，
  
  网络出版日期：2013-01-24，
  
  纸质出版日期：2013-01-15
- 稿件说明：
移动端阅览
姜黎吴伟仁张之敬金鑫节德刚. 微小型结构件显微图像边缘的自动识别[J]. 光学精密工程, 2013,21(1): 224-232

LI JIANG WU Wei-ren ZHANG Zhi-jing JIN Xin JIE De-gang. Automatic detection of micro image edges for micro accessories[J]. Editorial Office of Optics and Precision Engineering, 2013,21(1): 224-232
姜黎吴伟仁张之敬金鑫节德刚. 微小型结构件显微图像边缘的自动识别[J]. 光学精密工程, 2013,21(1): 224-232 DOI： 10.3788/OPE.20132101.0224.

LI JIANG WU Wei-ren ZHANG Zhi-jing JIN Xin JIE De-gang. Automatic detection of micro image edges for micro accessories[J]. Editorial Office of Optics and Precision Engineering, 2013,21(1): 224-232 DOI： 10.3788/OPE.20132101.0224.

摘要

在前一阶段的研究工作中，针对微小型结构件的显微图像边缘提取，作者提出了一种基于工艺匹配的显微图像边缘提取算法（Micro Processing Technology Matching based Micro Image Edge Detection

MPTM-MIED）。使用MPTM-MIED提取图像边缘之前，需要手工从零件显微图像中选取边缘过渡区域，这一操作将极大影响微小型零件的检测速度，从而使MPTM-MIED不能应用在实时自动检测中。针对MPTM-MIED的这一不足，本文利用BP神经网络技术重新设计并实现了MPTM-MIED，提出了一种新的自动提取显微图像边缘的方法（An Automated Micro Image Edge Detection Method，AMIED）。为了验证该方法的有效性，本文还利用AMIED对4种工艺微小型结构件显微图像的边缘进行了提取，并对线切割工艺零件的尺寸进行了测量。边缘提取的分析结果表明：AMIED提取出的显微图像边缘与MPTM-MIED提取出的基本一致；与常用的边缘检测算法相比，AMIED提取出的显微图像的边缘线形连接程度较好。测量尺寸的分析结果表明：MPTM-MIED和AMIED测量的尺寸基本相同，与Canny测量的相比，更接近万能工具显微镜测得的尺寸。

Abstract

In the previous research

a Micro Processing Technology Matching based Micro Image Edge Detection method (MPTM-MIED) was developed for detecting micro image edges of micro accessories. Before employing MPTM-MIED an edge transition zone has to be picked up manually from the micro image of micro accessories. This exerts a considerable influence on inspection speed and thus prevents MPTM-MIED from measuring sizes of micro accessories online and automatically. To overcome this shortcoming BP Neural Networks has been employed to design and implement a novel Automated Micro Image Edge Detection method (AMIED). From analysis results for micro accessories edge detections the following two conclusions are arrived at: the detected edges by AMIED and MPTM-MIED are almost the same; compared with some common used detection algorithms AMIED has the smallest edge-connectivity. From analysis results for measured sizes the following conclusions are drawn: the measured sizes by AMIED are almost equal to these measured ones by MPTM-MIED; compared with Canny the measured sizes by AMIED are closer to these measured ones by MPTM-MIED.

关键词

Keywords

references

张之敬，金鑫，周敏. 精密微小型制造理论、技术及其应用[J]. 机械工程学报，2007,43（1）：49-61.ZHANG Z J, JIN X, ZHOU M. Precise and microminiature manufacturing theory, technology and its appliance [J]. Chinese Journal of Mechanical Engineering, 2007, 43(1):49-61.(in Chinese)[2]王向军，王峰．显微精密成像与微型机械尺寸检测技术[J] ．光学精密工程，2011，19(6)：511-513．WANG X J, WANG F. Study of micro mechanical size inspection technology by microscope precision digital image [J]. Opt. Precision Eng., 2011, 19(6):511-513. (in Chinese)[3]NIXON M S, AGUADO A S. Feature Extraction and Image Processing [M]. 2nd ed. United Kingdom:Academic Press, 2008.[4]张之敬, 程朋乐, 金鑫, 等. 具有统计特征的微小型零件边缘检测技术[J]. 北京理工大学学报, 2007, 27(3): 231-234.ZHANG Z J, CHENG P L, JIN X, et al.. Edge detection technique for micro accessory edges having statistical characters [J]. Transactions of Beijing Institute of Technology, 2007, 27(3): 231-234. (in Chinese)[5]XU G S. Approach of image edge detection based on wavelet scale multiplication [J]. Applied Mechanics and Materials, 2012, 130-134:4282-4285.[6]田永刚，董毅．小波多尺度分割算法在细胞图像上的应用[J]．计算机仿真，2011，28（11）：255-301．TIAN Y G, DONG Y. Application of image division algorithm based on wavelet transform in cellular image [J]. Computer Simulation, 2011, 28(11):255-301. (in Chinese)[7]陈颖，蒋远大，孙志斌．基于Log-Gabor滤波的小波显微图像融合[J]．计算机工程与设计，2010，31(6)：1316-1319．CHEN Y, JIANG Y D, SUN Z B. Wavelet transform microscopy images fusion based on Log-Gabor filter [J]. Computer Engineering and Design, 2010, 31(6):1316-1319. (in Chinese)[8]陶雪容，孙兴波，汤秀华．基于融合的显微图像Gabor 边缘提取算法[J]．计算机工程与设计，2010，31(4)：798-804．TAO X R, SUN X B, TANG X H. Gabor filtering algorithm for micrograph edge detection based on fusion method [J]. Computer Engineering and Design, 2010, 31(4):798-804. (in Chinese)[9]SU T C, YANG M D, WU T C. Morphological segmentation based on edge detection for sewer pipe defects on CCTV images [J]. Expert Systems with Applications, 2011, 38(10):13094-13114.[10]ZHANG X Q, YANG K, HAO B Q. Cell-edge detection method based on canny algorithm and mathematical morphology [C]. The 3rd International Congress on Image and Signal Processing, Yantai, China: CISP, 2010: 894-897.[11]ZHU S H. Edge detection based on mathematical morphology and image fusion[C]. 2011 Cross Strait Quad-Regional Radio Science and Wireless Technology Conference, Harbin, China: CSQRWC, 2011: 1290-1293.[12]杜芳．基于显微视觉的小模数渐开线直齿轮检测方法研究[D]．北京：北京理工大学，2009: 37-38．DU F. Research on Detection Technology of Fine-pitch Involute Spur Gears Based on Micro-vision [D]. Beijing: Beijing Institute of Technology, 2009: 37-38. (in Chinese)[13]张之敬，杜芳，金鑫，等．微小尺寸零件复杂边缘识别算法[J]．光学精密工程，2009，17(2)：355-361．ZHANG Z J, DU F, JIN X, et al.. Complex edge recognition algorithm of micro-accessory [J]. Opt. Precision Eng., 2009, 17(2):355-361. (in Chinese)[14]张林，张之敬，杜芳，等．基于工艺匹配的微小型结构件边缘检测方法[J]．北京理工大学学报，2009，29(12)：1063-1066．ZHANG L, ZHANG Z J, DU F, et al.. Processing matching principle for edge detection method of micro parts [J]. Transactions of Beijing Institute of Technology, 2009, 29(12):1063-1066. (in Chinese)[15]SONKA M, HLAVAC V, BOYLE R. Image Processing, Analysis and Machine Vision [M]. Toronto: Thomson Learning, 2008.[16]胡伍生．神经网络理论及其工程应用[M]．北京：测绘出版社，2006．HU W S. The Theory of Neural Network and its Applications in Engineering [M]. Beijing: Surveying and Mapping Press, 2006. (in Chinese)[17]KIRCHEN L, ROSENFELD A. Edge evaluation using local edge coherence [J]. IEEE Transactions on System Man and Cybernetics, 1981, 11(9):597-605.

浏览量

233

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

暂无数据