研究生: |
曹凱翔 Kai-Siang Cao |
---|---|
論文名稱: |
使用機器視覺進行電線束 之瑕疵檢測 Application of Machine Vision to Detect Defects of a Wire Harness |
指導教授: |
李維楨
Wei-Chen Lee |
口試委員: |
修芳仲
Fang-Jung Shiou 徐繼聖 Gee-Sern Hsu |
學位類別: |
碩士 Master |
系所名稱: |
工程學院 - 機械工程系 Department of Mechanical Engineering |
論文出版年: | 2018 |
畢業學年度: | 106 |
語文別: | 中文 |
論文頁數: | 150 |
中文關鍵詞: | 機器視覺 、電線束 、瑕疵檢測 |
外文關鍵詞: | machine vision, wire harness, defect detection |
相關次數: | 點閱:494 下載:5 |
分享至: |
查詢本校圖書館目錄 查詢臺灣博碩士論文知識加值系統 勘誤回報 |
為了解決電線束組裝之瑕疵檢驗問題,本論文開發出針對電線束之零組件進行裝配檢測、定位檢測與尺寸量測等之自動化光學檢測系統。本研究使用單一攝影機進行實驗。利用影像之平均灰階值對電線束之標籤進行裝配檢測。使用模板匹配法對電線束裝配件之束帶、端子與車輪零件等進行裝配檢測與定位檢測。再使用邊緣偵測法對電線束之支線及全長與標籤進行尺寸量測與定位檢測。最後對十條電線束產品進行機器視覺瑕疵檢測,同時以人工目視檢測的結果作為參考,與本研究所開發的影像檢測系統之檢測結果進行比較,以此得知兩個檢測方法的差異。本研究所開發之影像系統對其中八條電線束進行瑕疵檢測結果與人工目視檢測結果相符,而有少數幾個長度量測值差異為1 mm。其中對第四條電線束進行瑕疵檢測時,發生標籤定位檢測誤判問題,其原因是邊緣偵測法偵測到電線束上之反光點導致誤判發生。對第七條電線束進行瑕疵檢測時,發生長度量測異常問題,其視覺量測與人工量測相差4 mm,誤判原因是人工安裝電線束時,安裝錯位導致長度量測起始點錯誤而引發誤判。最後將誤判起因修正後,重新對十條電線數進行檢測,其檢測結果皆與人工目視檢測結果相符合。因此本研究所開發的方法,可應用於電線束產品瑕疵檢測。
In order to solve the problem of defect inspection of a wire harness assembly, the objective of the research was to develop an automated optical inspection system for defect detection, component location detection and length measurement of the components of the wire harness.
A CCD camera was used for this research. The label on the wire harness was inspected using the average gray scale value of the image. The template matching method was used for assembly detection and positioning detection on the strap, connector and wheel components on the wire harness. In addition, the edge detection method was used to measure the length and position of the wire branch, the length and the label of the wire harness.
At last, the inspection system was tested using ten wire harness products. The results of visual inspection were used as references and compared with the results of the inspection system developed in this study to learn the difference between the two methods. The detection results of the eight wire harness were the same as the results of visual inspection. The differences in length were about 1 mm.
When the fourth wire harness was tested, a misjudgment of the label positioning detection occurred. The reason was that the edge detection method detected the light reflection point on the wire harness and caused misjudgment. When the seventh wire harness was tested, the incorrect length measurement occurred, and the visual measurement is different from the optical measurement by 4 mm. The reason for the misjudgment was that when the wire harness was manually installed, the misalignment caused the starting point of the length measurement incorrect.
Finally, in the case of the correct installation of the wiring harness, the method developed in this study can be applied to inspect the wire harness product without any problem.
[1] G. M. Shi and W. Jian, "Wiring harness assembly detection system based on image processing technology," in 2011 International Conference on Electronics, Communications and Control (ICECC), 2011, pp. 2397-2400.
[2] A. Vallan and F. Molinari, A Vision-Based Technique for Lay Length Measurement of Metallic Wire Ropes vol. 58, 2009.
[3] S. Liu, J. Liu, P. Jin, and X. Wang, "Tube measurement based on stereo-vision: a review," The International Journal of Advanced Manufacturing Technology, vol. 92, pp. 2017-2032, 2017/09/01 2017.
[4] C. C. Hsu, M.-C. Lu, W. Y. Wang, and Y. Y. Lu, Three-dimensional measurement of distant objects based on laser-projected CCD images vol. 3, 2009.
[5] Q. Al-Jubouri, W. Al-Nuaimy, M. Al-Taee, and I. Young, "An automated vision system for measurement of zebrafish length using low-cost orthogonal web cameras," Aquacultural Engineering, vol. 78, pp. 155-162, 2017/08/01/ 2017.
[6] D. Hu, W. Huang, J. Yang, L. Shang, and Z. Zhu, "Shape matching and object recognition using common base triangle area," IET Computer Vision, vol. 9, pp. 769-778, 2015.
[7] S. Mattoccia, F. Tombari, and L. D. Stefano, "Fast Full-Search Equivalent Template Matching by Enhanced Bounded Correlation," IEEE Transactions on Image Processing, vol. 17, pp. 528-538, 2008.
[8] A. Sibiryakov, "Fast and high-performance template matching method," in CVPR 2011, 2011, pp. 1417-1424.
[9] Y. R. Charles and R. Ramraj, "A novel local mesh color texture pattern for image retrieval system," AEU - International Journal of Electronics and Communications, vol. 70, pp. 225-233, 2016/03/01/ 2016.
[10] M. Gouiffès, A. Romero Mier y Terán, and L. Lacassagne, "Color enhanced local binary patterns in covariance matrices descriptors (ELBCM)," Journal of Visual Communication and Image Representation, vol. 49, pp. 447-458, 2017/11/01/ 2017.
[11] Z. Zhang, M. Zhao, B. Li, P. Tang, and F.-Z. Li, "Simple yet effective color principal and discriminant feature extraction for representing and recognizing color images," Neurocomputing, vol. 149, pp. 1058-1073, 2015/02/03/ 2015.
[12] S. Aligholi, G. R. Lashkaripour, R. Khajavi, and M. Razmara, "Automatic mineral identification using color tracking," Pattern Recognition, vol. 65, pp. 164-174, 2017/05/01/ 2017.
[13] S. Ghidoni, M. Finotto, and E. Menegatti, "Automatic Color Inspection for Colored Wires in Electric Cables," IEEE Transactions on Automation Science and Engineering, vol. 12, pp. 596-607, 2015.
[14] V. Chauhan and B. Surgenor, "A Comparative Study of Machine Vision Based Methods for Fault Detection in an Automated Assembly Machine," Procedia Manufacturing, vol. 1, pp. 416-428, 2015/01/01/ 2015.
[15] National Instruments, IMAQ Vision Concepts Manual, 2000.
[16] 鍾國亮,影像處理與電腦視覺: 台灣東華書局股份有限公司,2017年10月.
[17] 鐘正宜,"機器視覺之加工補正研究" 國立交通大學 工學院精密與自動化工程學程 碩士論文,2012年.