臺灣博碩士論文加值系統

在傳統產業中量測工件時因對工件尺寸要求精度高，而所使用的設備價格多為昂貴且精密的儀器來對產品進行檢驗，但隨著科技不斷進步，本研究期望以低成本的非接觸式量測設備對工件進行檢測，並導入自動光學系統技術來達到高效能經濟效益。
本研究透過Visual Basic搭配MIL影像函式庫建立一套環形工件輪廓量測系統。為了在量測工件時能獲得準確之位置資料，因此必須先取得已知影像座標及相對應之世界座標之校正點資料，進而取得影像座標與世界座標之轉換關係，再由架設於工作平台上方雙攝影機分段擷取環型工件影像，接著運用二值化、邊緣偵測、尋找輪廓等方法來得到環形工件影像座標，最後將轉換完後的世界座標匯入至繪圖軟體做尺寸上量測。
藉由驗證圖紙實驗之結果顯示，在校正點誤差分析方面X軸座標總誤差為0.198mm，Y軸座標總誤差為0.251mm。而本系統所提出邏輯校正方法在X軸上的解析度為0.139mm/pixel，Y軸上解析度為0.143mm/pixel，因其值非常接近，故本校正系統具有其可信度，但在量測方面總誤差約為0.6mm，代表系統精密度與準確度仍有改善空間。

Most of the measurement equipment are expensive and sophisticated to inspect the mechanical product due to the high precision requirement when measuring the workpiece in traditional mechanical industry. However, due to new technology development, this study expects to measure the workpiece with cost effective by means of non-contact measuring technology.
Measurement system for contours of ring-shaped workpiece was developed by Visual Basic with MIL image library in this study. In order to obtain accurate positioning information in the measurement of the workpiece, the first priority is to obtain relationship between image coordinate system and the world coordinate system by calibration process. Then, the image of the ring-shape workpiece is captured by double camera above the working platform segmentally. In order to obtain the contour of ring-shaped in image coordinate system, binary thresholding, edge detection and finding contour methods are used. Finally, after converting to world coordinate system, these contour data points are exported to CAD software to quantify the size of important place.
According to experimental results of this study, the total errors of X and Y axes are 0.198 mm and 0.251mm, respectively, in error analysis of calibration points. The above values are close to the hardware resolution which are 0.139 mm/pixel and 0.143 mm/pixel in X and Y axes, respectively. Because the above values are very close, the calibration method is reliable. However, the total error for contour measurement is about 0.6 mm by comparing to ATOS measurement, which means the system precision and accuracy is needed to improve for further study.

中文摘要 I
Abstract II
誌謝 IV
目錄 V
圖目錄 VII
表目錄 X
第一章 緒論 1
1-1 前言 1
1-2 研究動機與目的 2
1-3 文獻回顧 3
1-4 論文架構 12
第二章 理論方法 14
2-1 影像處理與機械視覺概述 14
2-2 打光技術方式 15
2-2-1 正向打光 17
2-2-2 背向打光 17
2-2-3 結構打光 17
2-3 影像處理概論 18
2-3-1 二值化(Binary) 18
2-3-2 Blob分析(Blob Analysis) 19
2-3-3 坎尼邊緣偵測(Canny Edge Detector) 20
2-3-4 尋找輪廓(Find Contour) 22
2-3-5 邊緣量測(Edge Measurement) 22
2-3-6 霍夫轉換(Hough Transform) 24
第三章 實驗設備 26
3-1 硬體系統介紹 26
3-1-1 背光模組介紹 28
3-1-2 ATOS光學掃描設備 30
3-1-3 校正版校正點分配與佈置 32
3-1-4 三次元量測儀 34
3-2 軟體介紹 36
3-2-1 GOM Inspect 36
3-2-2 Visual Basic 6.0 38
3-2-3 Matrox Image Library 5.12 38
3-2-4 VisualC# 39
3-2-5 Emgu CV 40
3-3 環形工件製程介紹 41
第四章 研究方法 43
4-1 逆向工程 43
4-2 校正程式邏輯解說 45
4-3 人機介面功能設計 50
第五章 結果與討論 51
5-1 GOM Inspect誤差分析 51
5-2 驗證校正點誤差分析及系統解析度 54
5-3 產生輪廓 56
5-4 提高解析度後校正點誤差分析及量測 60
第六章 結論與建議 63
6-1 結論 63
6-2 建議與未來展望 64
參考文獻 65

[1]張承宗， “ATOS光學量測系統說明”，中工高雄會刊，第一期，第22卷。
[2]Md. H.Ali,S. Kurokawa,K. Uesugi,2014, “Camera based precision measurement accuracy in improving measurement accuracy,” Measurement Vol.49,pp.138-147.
[3]陳建男，2006，以機械視覺建立工件輪廓軌跡之研究，碩士論文，南台科技大學機械工程研究所。
[4]Md. H.Ali ,S. Kurokawa, K. Uesugi,2014, “Application of machine vision in improving safety and reliability for gear profile measurement,” Machine Vision and Applications ,Vol.25,pp.1549-1559.
[5]J. Loizou,W. Tian,J. Robertson,J.Camelio,2015, “Automated wear characterization for broaching tools based on machine vision systems,” Journal of Manufacturing Systems,Vol.37,pp.558-563.
[6]C. S. Lin, T. C. Wei, A. T. Lu, S. S. Hung, W. L. Chen and C. C. Chang, 2011“A rehabilitation training system with double-CCD camera and automatic spatial positioning technique”, Optics and Lasers in Engineering , Vol. 49, pp. 457-464.
[7]F. Samopa and A. Asano,2008, “Hybrid Image Thresholding Method Using Edge Detection”, IJCSNS International Journal of Computer Science and Network Security, Vol. 8, No. 11, pp. 292-299
[8]F. Adamo, F. Attivissimo, A. D. Nisio and M. Savino,2009, “A low-cost Inspection System for Online Defects Assessment in Satin Glass”, Measurement, Vol. 42, pp. 1304-1311.
[9]J. Canny,1986, “A Computational Approach to Edge Detection,” IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 8, No. 6, pp. 679-698.
[10]X. Yu, X. Lin,Y. Dai,K. Zhu,2017, “Image edge detection based tool condition monitoring with morphological component analysis,” ISA Transactions.
[11]F. Röhrbein, P. Goddard, M. Schneider, G. James, K. Guo,2015, “How
does image noise affect actual and predicted human gaze allocation in assessing
image quality?,” Vision Research, Vol. 112, pp. 11-25.
[12]陳建軍，2012，引入低對比度環境下視覺感知機制的輪廓檢測模型，模式識別與人工智能，第25期，第22卷。
[13]F. Remondino,C. Fraser,2006, “Digital camera calibration methods：
considerations and comparisons," International Archives of Photogrammetry,
Remote Sensing and Spatial Information Sciences, 36(5),pp.266-272.
[14]J. Heikkila, O. Silven,1997, “A four-step camera calibration procedure
with implicit image correction, ” Computer Vision and Pattern Recognition,
Proceedings, IEEE Computer Society Conference on.pp.1106-1112.
[15]T.A. Clarke,J. G. Fryer,1998, “The development of camera Calibration methods and models," The Photogrammetric Record, 16(91),pp.51-66
[16]M. S. Joun, S. W. Lee and J. H. Chung, “Finite element analysis of a multi-stage axisymmetric forging process,” International Journal of Machine Tools & Manufacture, Vol. 38, No. 7, pp. 843-854.
[17]Hui-Fuang Ng, 2006, “Automatic thresholding for defect detection” , Pattern
Recognition Letters, Vol. 27, pp. 1644-1649.
[18]Xin Li and Micnael T. Orchard, 2011, “New Edge-Directed Interpolation”,
IEEE Transactions on Image Processing, Vol. 10, No. 10.
[19]邱振洋，2000，“應用電腦視覺於IC檢測之研究”，碩士論文，國立雲林
科技大學工業工程與管理研究所。
[20]陳慶瀚，電腦視覺(Computer Vision)講義第八單元Hough 轉換，中央大學資訊工程學系。
[21]昌達 慶仁，2011， “圖解數位影像處理程式範例教本使用C語言”，旗標出版股份有限公司。
[22]毛星雲、冷雪飛等人編著，2015，OpenCV 3 編程入門，電子工業出版社
[23]Sony -XC77規格手冊
[24]ATOS光柵投影設備、GOM Inspect簡介，http：//www.rat.com.tw/home.html，
馬路科技公司。
[25]Mitutoyo BH504使用手冊，台灣三豐儀器股份有限公司。
[26]張祐慈，2013，非接觸式表面瑕疵檢測精密度提升之研究，碩士論文，國立高雄應用科技大學機械與精密工程研究所。
[27]Matrox Electronic Systems Ltd., Matrox Imaging Library User Guide, 1998。
[28]黃聰明，2013，Visual C#物件導向程式設計，松崗資產管理股份有限公司。
[29]蔡文龍、林義証等人編著，2015，Visual C#學習經典，松崗資產管理股份有限公司。
[30]Emgu CV官網，http：//www.emgu.com/wiki/index.php/Main_Page
[31]蘇春熺譯著，2014，機械製造第七版，歐亞書局有限公司。
[32]陳昆明，2015，機械視覺應用於不同類型汽車晴雨窗加工定位系統開發，
碩士論文，國立高雄應用科技大學機械與精密工程研究所。