工業4.0的發展導致自動化的需求增加。自動化的流程中需使用電動機械取代人力並搭配軟體算法取代人工操作，本篇論文使用了高明鐵企業股份有限公司開發的六軸堆疊式對位平台搭配自行開發視覺系統完成自動化工作。

近年來，科技的快速進步和網際網路的普及化，帶動了視訊通信需求。在生產鏡頭的流程中，鏡頭品質檢測是必要環節，以往以人工檢測方式需花費可觀時間、人力，故導入自動化為不可或缺之必要項目。

為了達到自動化，必須先將鏡頭對準品質檢測圖片、拍攝圖片、並使用人工檢測原有程式之功能輸出結果。其中在對位的環節中應用了影像追蹤、影像處理等相關技術，並搭配通用型六軸堆疊式對位平台做移動鏡頭動作，進而完成對位及整體自動化流程。

影像追蹤算法分別針對影像伺服、相似三角形投影及影像座標比例換算做實際測試及比較，最終選擇了適合此鏡頭品質檢測應用的影像座標比例換算方式做為追蹤手段。

本篇論文為探討何種影像追蹤方式適用於鏡頭對位流程，並導入自動化流程，整合人機介面、相機通訊、六軸堆疊式對位平台，設計成為鏡頭品質檢測自動化系統。主要貢獻於整合前述各細項，並解決產線自動化問題。

The development of Industry 4.0 has led to an increasing demand for automation. In the automation process, electric machinery is used to replace manual labor, and software algorithms are employed to replace manual operations. In this paper, the automation tasks are accomplished using the six-axis serially stacking platform developed by GMT, combined with a self-developed vision system.

In recent years, the rapid advancement of technology and the popularization of the internet have stimulated the demand for video communication. Within the lens manufacturing process, lens quality inspection is a necessary stage. In the past, manual inspection methods required considerable time and manpower. Therefore, the introduction of automation has become an essential requirement to streamline the process.

To achieve automation, the lens must be aligned with the quality inspection image, capture the image, and use the functionality of the existing manual inspection program to output the results. In this alignment process, relevant technologies such as image tracking and image processing are applied. Additionally, a six-axis serially stacking platform developed by GMT is used to move the lens and complete the alignment and overall automation process.

Various image tracking algorithms, including visual servoing, similar triangle projection, and image coordinate ratio conversion, were tested and compared. Ultimately, the image coordinate ratio conversion method suitable for this lens quality inspection application was chosen as the tracking method.

This paper focuses on which image tracking method is suitable for the lens alignment process and introduces the automation process. It integrates interface, camera communication, and six-axis is serially stacked platform to design an automated system for lens quality inspection. The primary contributions lie in integrating the aforementioned components and addressing the automation issues in the production line.

學位考試委員會審定書 I
論文摘要 III
Abstract V
誌謝 VII
目錄 IX
圖目錄 XI
表目錄 XV
第1 章緒論 1
1.1 動機與目的 1
1.2 文獻回顧 3
1.3 論文架構 4
第2 章影像處理與追蹤 5
2.1 影像特徵點擷取 5
2.2 影像追蹤 17
2.2.1 相機參數 17
2.2.2 基於影像平面座標系之影像伺服控制 22
2.2.3 像素座標映射 40
2.2.4 影像座標比例換算 43
第3 章硬體、軟體整合應用 49
3.1 硬體 49
3.2 通訊 52
3.2.1 相機通訊 52
3.2.2 六軸堆疊式對位平台通訊 53
3.3 軟體 55
第4 章測試結果與討論 59
4.1 對位結果 59
4.1.1 基於影像平面座標系之影像伺服控制對位結果 59
4.1.2 像素映射對位結果 70
4.1.3 影像座標比例換算對位結果 72
4.2 應用結果 74
第5 章結論 77
5.1 結論 77
5.2 未來展望 78
5.3 作者簡介 79
參考文獻 80

[1] C. S. Solutions, “Resolution of the captured image - measured,” (Accessed: 2023-07-27). [Online].Available: https://www.causewaysecuritysolutions.com/resolution-measured-tv-lines.html
[2] G. J. Agin, Real time control of a robot with a mobile camera, ser. Technical note. SRI International,1979. [Online]. Available: https://books.google.com.tw/books?id=XEAtGwAACAAJ
[3] A. C. Sanderson and L. E. Weiss, “Adaptive visual servo control of robots,” in Robot Vision, A. Pugh,Ed. Springer Berlin Heidelberg, 1983, pp. 107–116.
[4] Chaumette and Francois, “Potential problems of stability and convergence in image-based and position-based visual servoing,” in The confluence of vision and control, D. J. Kriegman, G. D. Hager, and A. S. Morse, Eds. Springer London, 1998, pp. 66–78.
[5] E. Malis, F. Chaumette, and S. Boudet, “2 1/2 d visual servoing,” IEEE Transactions on Robotics and Automation, vol. 15, no. 2, pp. 238–250, 1999.
[6] P. Corke and S. Hutchinson, “A new partitioned approach to image-based visual servo control,” IEEE Transactions on Robotics and Automation, vol. 17, no. 4, pp. 507–515, 2001.
[7] H. Kong, H. C. Akakin, and S. E. Sarma, “A generalized laplacian of gaussian filter for blob detection and its applications,” IEEE Transactions on Cybernetics, vol. 43, no. 6, pp. 1719–1733, 2013.
[8] OpenCV, “Image blurring,” (Accessed: 2023-07-10). [Online]. Available: https://docs.opencv.org/4.x/d4/d13/tutorial_py_filtering.html
[9] Z. Zhang, “Flexible camera calibration by viewing a plane from unknown orientations,” in Proceedings of the Seventh IEEE International Conference on Computer Vision, vol. 1, 1999, pp. 666–673 vol.1.
[10] J.-P. Su, “Ibvs-4 points,” Class Handout, National Dong-Hwa University, 2022.
[11] E. Marchand, F. Spindler, and F. Chaumette, “Visp for visual servoing: a generic software platform with a wide class of robot control skills,” IEEE Robotics and Automation Magazine, vol. 12, no. 4, pp.40–52, Dec 2005.
[12] VISP, “Image-based visual servo,” (Accessed: 2023-07-10). [Online]. Available: https://visp-doc.inria.fr/doxygen/visp-daily/tutorial-ibvs.html
[13] OpenCV, “Video capture,” (Accessed: 2023-07-10). [Online]. Available: https://docs.opencv.org/3.4/d8/dfe/classcv_1_1VideoCapture.html