隨著工業科技技術的進步，檢測工具機的項目逐漸多樣化以及效率上的提升，精度要求也越來越嚴謹。在一台高水準的工具機鏟花是一項關鍵技術，其中，鏟花應用於工具機之主要兩大部分為滑動面與固定接合面，若滑動面不平整將導致滑塊在軌道行走時產生磨耗且將減少機台的使用壽命，固定接合面將影響工具機的剛性與精度，例如螺桿與螺帽接合面、主軸頭接合面、立柱接合面這幾項等等，都是鏟配的主要部分，並將檢測後的鏟花參數紀錄，建立完整的鏟花生產履歷。
由於目前大多鏟花面的品質判斷為師傅憑著經驗用肉眼檢視，這方式可能造成檢測上的不足導致工具機使用壽命縮短，針對這部分本研究利用PC進行圖像檢測，以C#撰寫搭配跨平台的Emgu CV影像處理庫，將拓印完後鏟花面掃描，經過使用者調整適合的影像參數再進行處理取得鏟花面的面積和參數，鏟花面檢測時間減少50%，大幅減少作業時間並且可檢測整體的加工面的狀況，然而連結本實驗室計畫的資料庫系統，以MariaDB引擎建立一套生產履歷資料庫，存取工件名稱、鏟花參數PPI、POP及整體加工面高點數，作為品管之依據。

With the technological advancement in industry, the project of inspection for machine tools is not only gradually diversified but also the efficiency gets a promotion, besides, the required precision is becoming more exact.
Therefore, scraping is one of the key technology that uses for manufacturing and assembly on a high-quality machine tool. The main two parts of the scraping are the sliding surface and the fixed joint surface. If the sliding surface is not flat, it will cause the slider and surface to wear when the track is walking then will reduce the life span of the machine. The fixed joint will affect the rigidity and precision of the machine, such as the screw and nut joint surface, the spindle joint surface, the column joint surface, etc., these surfaces will be inspected. The parameters are built a complete traceability of scraping.
Currently, quality of the scraping surface is determined by the master's eye because that is depended on experience. This method possibly shortens the life span of the machine tool due to the lack of inspection. For this part, the study is replaced human by personal computer to inspect. In the procedure that composed a cross-platform Emgu CV image processing library and Visual Studio c#. After using the scraping surface transfer by rubbing and scan, and then user obtains the area, PPI, POP these parameters of scraping surface through that adjustable system adjusts parameters of image to process it. Time of surface inspection was reduced 50% and realized the status of the overall surface. Finally, the inspection system connects to MariaDB database system of laboratory. Access to the name of workpiece, PPI, POP and the high points of overall processing surface as the basis of quality control.

摘要.......................................i
Abstract................................. ii
誌謝......................................iv
目錄.......................................v
表目錄..................................viii
圖目錄....................................ix
第一章 緒論...............................1
1.1 前言...............................1
1.2 研究目的...........................2
1.3 國內外相關研究......................2
1.3.1 視覺影像處理........................3
1.3.2 承斑視覺檢與自動鏟花技術.............8
1.4 論文架構 ..........................14
第二章 系統架構與流程介紹..................15
2.1 系統流程架構........................15
2.2 鏟花面承斑檢測........................17
2.2.1 鏟花高低點與接觸率........................19
2.3 影像軟體 ........................22
2.4 資料庫管理系統架構........................24
2.4.1 資料庫介紹........................25
2.4.2 MariaDB資料庫引擎介紹........................26
2.4.3 關聯式資料庫(Relation Database)介紹........................27
第三章 影像處理與原理分析........................31
3.1 數位影像類型........................31
3.1.1 影像解析度........................34
3.2 影像YUV種類........................35
3.2.1 YUV常用格式........................36
3.2.2 YUV與RGB轉換公式........................ 37
3.3 圖像分割法........................39
3.3.1 閥值二值化........................40
3.4 影像形態處理........................42
3.4.1 侵蝕(Erosion)........................42
3.4.2 膨脹(dilation)........................43
3.4.3 開運算(Opening)........................44
3.4.4 閉運算(Closing)........................45
3.5 像素連通區域........................46
3.5.1 連通區域標記(Connected Component Labeling).....47
3.5.2 篩選面積 ..............................50
3.5.3 區域的質心........................51
第四章 實驗檢測結果........................52
4.1 影像檢測流程........................52
4.1.1 檢測人機介面........................53
4.2 影像處理結果........................61
4.2.1 色彩灰階化........................61
4.2.2 閥值分割........................ 62
4.2.3 開運算(Opening)........................64
4.2.4 連通區域標記........................65
4.3 MATLAB Toolbox........................66
4.4 資料庫介面........................67
第五章 結論與未來展望........................70
5.1 結論........................70
5.2 未來展望 ........................71
參考文獻 ........................72
Extended Abstract........................76

[1]Sakai, N., S. YoneKawa and A. Matsuzaki, 1996, “Two-dimensional Image Analysis of the shape of Rice and its Application to Separating Varieties”, Journal of Food Engineering 27, P.397-P.407.
[2]黃膺任、李芳繁，1997，“使用影像處理進行棗子分級之研究”，農業機械學刊，第6卷，第2期，pp. 15-23.
[3]萬一怒，溫惠雯，1998，“糙米品質自動檢測分級系統(二)檢測系統之研製”，農業機械學刊，第7卷，第2 期，pp.13-27.
[4]梁有燈，1999，“機器視覺在BGA 銲球尺寸量測上之應用”，中華大學機械與航台工程研究所碩士論文。
[5]Miroslav Hrianka, Marek Slusarčík, 2003,“Quality Transmission of Additional Image Information by 2D Image Basis”, IFAC Proceedings Volumes, Volume 36, Issue 1, February 2003, pp.277-280.
[6]A.A.MalcolmaH.Y.LeongbA.C.SpowageaA.P.Shacklockc, 2007, “Image segmentation and analysis for porosity measurement”, Journal of Materials Processing Technology, Volumes 192–193, 1 October 2007, pp.391-396.
[7]Roberto A. Lotufo, Romaric Audigier, André V. Saúde, Rubens C. Machado, 2008, “Morphological Image Processing”, Microscope Image Processing 2008, pp. 113-157.
[8]Xiaoli Jin, Yutian Li, 2010, “Generation algorithm of direction-parallel tool path based on image processing”, 2010 International Conference on Mechanic Automation and Control Engineering, pp. 600-603.
[9]莊峻豪，2014，雨天車前擋風玻璃影像之還原技術，國立中山大學機械與機電工程學系，碩士論文。
[10]Md. Zainal Abedin , Atul Chandra Nath , Prashengit Dhar , Kaushik Deb , Mohammad Shahadat Hossain, 2017, “License plate recognition system based on contour properties and deep learning model”, 2017 IEEE Region 10 Humanitarian Technology Conference (R10-HTC) pp. 590-593.
[11]林貞仁，2009，“工具機精度與鏟花關係”，亞東學報，第5期，頁37-65。
[12]T. Yoshimi, S.Masafumi, Y. Tetsuya and C.Masahiro, 1968, “The Recognition of Bearings by Means of a CCD Line Sensor and the Automation of Scraping Works”, NII-Electronic Library Service, Vol. 71, pp.93-98.
[13]Yasuo YAMANE, Rong Sheng YIN, Hiroyuki OKAMOTO, Masanori YAMADA, You YONEZAWA and Norihiko NARUTAKI, 1989, “Application of Multi-Joint Type Industrial Robot to Scraping of Surface Plate”, Journal of the Japan Society for Precision Engineering ,Volume 55 Issue 10 pp. 1787-1792.
[14]H. Tsutsumi, R. Yamada, A. Kyus ojin and T. Nakamura, 1996, “Development of an Automatic Scraping Machine with Recognition for Bearing of Scraped Surfaces -Recognition of Black Bearing by CCD Camera (1rd report)”, Journal of the JapanSociety for Precision Engineering, Contributed Papers, Vol. 62, pp.219-223.
[15]H. Tsutsumi, A. Kyusojin, and T. Nakamura, 1996, “Development of an Automatic Scraping Machine with Recognition for Bearing of Scraped Surfaces-Development of a Scraping Motion Mechanism (2nd report),” Journal of the Japan Society for Precision Engineering, Contributed Papers, vol. 62, pp.554-558.
[16]Tadanori SUGINO, Yasuo YAMANE, Itaru SAKUMA, Norihiko NARUTAKI, 2003, “Digitalization and Intellectualization of Scraping Operation (1st Report) Investigation of a Series of Scraping Operation by Skiful Worker”, Journal of the Japan Society for Precision Engineering, Volume 69, Issue 7, pp. 949-954.
[17]堤博貴，山田隆太，久曽神煌，中村輝幸，2005，“当たり面認識装置を含む自動きさげ盤の開発(第3 報) ：自動きさげ盤の構成”， 精密工學會誌，vol.71，pp.358-362。
[18]周睿程，2009，“一種對於鏟花工件表面輪廓的光學量測法”，臺灣大學碩士論文。
[19]Tung-Hsien Hsieh, Wen-Yuh Jywe, Hsueh-Liang Huang and Shang-Liang Chen, 2011, “Development of a laser-based measurement system for evaluation of the scraping workpiece quality”, Optics and Lasers in Engineering, No.49, pp.1045-1053.
[20]陳嘉宏，2011，“開發鏟花加工數值控制流程與方法”，國立虎尾科技大學自動化工程學系碩士論文。
[21]陳正達，2011，“滑動面表面文理之摩擦特性探討”，國立中興大學機械工程學系碩士論文。
[22]莊朝盛，2012，“機器視覺於鏟花承斑檢測之應用”，國立勤益科技大學機械工程系在職班碩士論文。
[23]黃峻彥，2013，“搭配CMOS 相機與光罩之機器視覺系統於鏟花加工面檢測之研究”，國立勤益科技大學機械工程學系碩士論文。


[24]機械基礎工藝-說說刮研與鏟花，http://s-n-k.blog.163.com/blog/static/284450120161025111431644/
[25]鏟花訓練教材，https://wenku.baidu.com/view/a78e39170b4e767f5acfcec1
[26]鏟花技術交流平台，http://scraping.pmc.org.tw
[27]R. King，2008， “「手動及電動機械鏟花技術訓練」講義” ，基礎鏟花技術手冊。
[28]https://zh.wikipedia.org/wiki/%E8%89%B2%E5%BD%A9%E6%B7%B1%E5%BA%A6
[29]http://163.23.84.67/~teac/p/doc/p-r/p-r.htmI
[30]YUV，https://zh.wikipedia.org/wiki/YUV
[31]https://commons.wikimedia.org/wiki/File:Common_chroma_subsampling_ratios.svgAD7606 datasheet.
[32]圖像分割，https://blog.csdn.net/yangleo1987/article/details/53173753
[33]舒紅平，蔣建民，2003，“基于灰度最优阈值的图像分割方法及应用”，重庆工商大学学报(自然科学版)， 04期，pp. 80-83
[34]徐豐明，張家源，林銘哲，王守細，朱孝祥，“數位化筆記辨識管理系統”，朝陽科技大學資訊工程學系。
[35]https://ccw1986.blogspot.com/2012/11/opencverodedilate.html