WU Ye-lan1,2, QIN Yan-hong2, ZHANG Zhi-jing2
(1. College of Mechanical Engineering, Beijing Institute of Technology, Beijing 100081, China; 2. College of Computer &Information Engineering, Beijing Technology and Business University, Beijing 100048, China)
Abstract： In order to improve the edge detection precision of miniature parts in microscopic field of view, a sub-pixel edge detection algorithm combining non-orthogonal quadratic B-spline wavelet transform algorithm and Zernike moment algorithm is proposed. Non-orthogonal quadratic B-spline wavelet transform algorithm is adopted to get the pixel edge of miniature parts, and the moment invariant of Zernike moment algorithm is used for refining the pixel edge to get sub-pixel edges. A real-time detection system based on the proposed algorithm for miniature parts is established. The general system structure and operational principle are given, the real-time image acquisition and detection are completed, the results of edge detection are analyzed and the detection precision is evaluated. The results show that parts size can be 0.01-10 mm and the detection precision reaches 0.01%-0.1%. Therefore, the edge of the miniature parts can be accurately identified and the detection precision can be improved to sub-pixel level, which meets the requirements of miniature parts precision detection.
Key words： miniature parts; sub-pixel edge detection; wavelet transform; Zernike moment
CLD number： TN911.73Document code： A
Article ID： 1674-8042（2017）01-0054-06doi： 10.3969/j.issn.1674-8042-2017-01-009
References
［1］Burisch A, Raatz A, Hesselbach J. Challenges of precision assembly with a miniaturized robot. In: Proceedings of IFIP Advances in Information and Communication Technology, 2010, 315： 227-234.
［2］CHENG Peng-le. Micro-parts vision inspection technique for assembly. Journal of Electronic Measurementand Instrument, 2013, 27(2)： 151-155.
［3］ZHAO Lin-hui, WANG Qiang, ZHANG Jian-cheng, et al. Research on micro-assembly system for the micro-plate parts. Modern Manufacturing Engineering, 2012, 16(12)： 1701-1706.
［4］CHEN Xiang-wei, WANG Hai-yue. Dimensions measurement of shaft sleeve size based on machine vision. Manufacturing Technology & Machine Tool, 2014, (10)： 85-88.
［5］CHEN Xiang-wei, XIAO Bing, GAO Qiang. Inspection of cylindrical electronic components surface defect based on machine vision.Machine Tool & Hydraulics, 2014, (14)： 116-118,123.
［6］CHEN Lian-qing, YUAN Hong-bin, WANG Long-shan. Study on measurement of micro size based on computer vision technology. Transactions of the Chinese Society for Agricultural Machinery, 2006, 37(9)： 174-176.
［7］CHEN Lian-qing, GUO Jian-liang, YANG Xun, et al. Grinding roughness prediction model based on evolutionary artificial neural network.Computer Integrated Manufacturing Systems, 2013, 19(11)： 2854-2863.
［8］LIU Qing-min, ZHAO Ling, YANG Yao-hua. Precision integrated measurement on circular saw blade based on computer vision. Mechanical & Electrical Engineering Magazine, 2009, 26(10)： 19-22.
［9］ZHANG Zhi-ing, CHENG Peng-le, JIN Xin, et al. Edge detection technique for micro accessory edges having statistical characters. Transactions of Beijing Institute of Technology, 2007, 27(3)： 231-234.
［10］JIANG Li, WU Wei-ren, ZHANG Zhi-jing, et al. Automatic detection of micro image edges for accessories. Optics and Precision Engineering, 2013, 21(1)： 224-232.
［11］Zhang X. Edge feature extraction in digital images with the ant colony system. In： Proceedings of IEEE International Conference on Computational Intelligence for Measurement Systems and Applications, 2004： 133-136.
［12］TIAN Jing, YU Wei-yu, XIE Sheng-li. An ant colony optimization algorithm for image edge detection. In： Proceedings of IEEE Congress on Evolutionary Computation,2008： 751-756.
［13］YANG Fu-sheng. Project analysis and application of wavelet transform. Beijing： Science Publishing House, 2003.
［14］FU Li-si, LIU Peng-wei, LI Dan-dan. Improved moment invariant characteristics and object recongnition.Computer Engineering and Applications, 2012, 48(31)： 183-185.
［15］Papakostas G A, Boutalis Y S, Karras D A, et al. A new class of Zernike moments for computer vision applications. Information Science, 2007, 177(13)： 2802-2819.
显微视场下微型零件亚像素边缘检测方法
吴叶兰1,2， 秦艳红2， 张之敬2
（1. 北京理工大学 车辆与机械工程学院， 北京 100081；2. 北京工商大学 计算机与信息工程学院， 北京 100048）
摘要： 为了提高微型零件在显微视场下的边缘检测精度， 提出了一种非正交二次B样条小波变换结合Zernike矩的亚像素边缘检测算法. 采用非正交二次B样条小波变换算法得到图像的像素级边缘， 利用Zernike矩算法的矩不变性对像素级边缘进行亚像素边缘细化。 为了实现算法的原理， 建立了一套微型零件的实时检测系统。 给出了系统的总体结构和工作原理， 完成了实时图像采集与检测， 分析了检测结果， 并对检测精度进行了评估。 实验结果表明： 该系统检测零件尺寸可以达到0.01~10 mm， 检测精度可以达到0.01%~0.1%， 可准确识别出微型零件的边缘， 将检测精度提高到亚像素级， 满足了在显微视场下微型零件检测的需要。
关键词： 微型零件； 亚像素边缘检测； 小波变换； Zernike矩
引用格式：WU Ye-lan, QIN Yan-hong, ZHANG Zhi-jing. Sub-pixel edge detection method for miniature parts in microscopic field of view. Journal of Measurement Science and Instrumentation, 2017, 8(1)： 54-59. ［doi： 10.3969／j.issn.1674-8042.2017-01-009］
[full text view]
