臺灣博碩士論文加值系統

本論文目的在於開發相關之電腦視覺技術並應用於自動化錢幣檢測。本研究所提技術可以有效處理三個核心問題，包括「複雜背景」、「錢幣旋轉」、以及「部分遮蔽」。
此論文主要技術包含「錢幣偵測」及「辨識」。首先，偵測模組以樣板匹配合併後置分析來穩定地偵測出每張影像上的多個錢幣。為了能抵抗錢幣旋轉，我們定義了一種新特徵並配合似字串比對方式可以將旋轉後的錢幣與資料庫中的參考模型進行對齊(Alignment)以找出旋轉角度。同時，辨識模組亦可將錢幣被遮蔽的區域予以標示，並用未被遮蔽部份之特徵來對被遮蔽錢幣進行更可靠的辨識。經由使用由田科技提供之測試資料，結果顯示本研究提出之技術確實可令人滿意地解決上述之三大問題。

The study in this thesis aims at developing machine vision techniques for an automatic coin recognition system. Three major problems, including complex background, coin rotation and partial occlusion, are effectively handled in the proposed techniques.
Two main modules in the proposed system are the coin detection and the coin recognition. The detection module employs the template matching combined with some post-analyses to robustly detect multiple coins on each input image. To tolerate the rotation of coins during recognition, we define a novel feature and a string matching method which can align the rotated coins with the reference models stored in database. Meanwhile, the recognition module can mark the occluded regions of coins and use only the features of non-occluded regions for reliable recognition of occluded coins. According to the experiments conducted on a public dataset from UTechzone Co., Ltd., the results show that the proposed techniques do tackle the above-mentioned three problems very well.

摘　要 2
ABSTRACT 3
致　謝 4
目 錄 5
圖目錄 7
表目錄 9
第 1 章 緒論 11
1.1 研究動機 11
1.2 研究目的 12
1.3 系統架構 14
1.4 章節組識 15
第 2 章 相關研究 17
2.1 圓偵測 17
2.2 特徵抽取及辨識 19
2.2.1 基於影像校正方法 19
2.2.2 旋轉不變性特徵（Rotation Invariance） 20
第 3 章 錢幣自動檢測 23
3.1 錢幣偵測（Coins Detection） 23
3.1.1 Sobel邊緣檢測 24
3.1.2 樣板匹配（Template Matching） 25
3.1.3 第二次sobel測邊 26
3.1.4 雜訊濾波器 (threshold 及 labeling) 27
3.1.5 圓遮蔽排除 27
3.2 特徵擷取（Feature Extraction） 28
3.2.1 影像範圍選取 29
3.2.2 極座標轉換 30
3.2.3 影像正規化 30
3.2.4 PCA色彩校正 30
3.2.5 遮蔽抵抗：標記（Label） 32
3.2.6 旋轉抵抗：似字串比對（String Matching） 34
3.3 錢幣辨識（Coin Recognition） 36
3.3.1 特徵函式（Feature Function） 36
3.3.2 二階段辨識 36
第 4 章 實驗結果 39
4.1 環境介紹 39
4.2 實驗結果與討論 39
4.2.1 錢幣偵測實驗分析 39
4.2.2 抗旋轉實驗分析 53
4.2.3 抗遮蔽實驗 55
4.2.4 抗光線變化實驗分析 59
第 5 章 結論與未來研究方向 63
5.1 結論 63
5.2 未來研究方向 63
參考文獻 65

[1]R.O.Duda, R.E.Hart., Use of the Hough Transform to Detect Lines and Curves in Pictures, CACM(15). No. 1, January 1972, pp. 11-15.
[2]Minoru Fukumi, Sigeru Omatu, Fumiaki Takeda, Toshihisa Kosaka, “Rotation-Invariant Neural Pattern Recognition System with Application to Coin Recognition”, IEEE Transactions, Neural Networks, Vol 3, No.2, March 1992.
[3]Michael Nolle, Harald Penz, Michael Rubik, Konrad Mayer, Igor Hollander, Reinhard Granec, “Dagobert-A New Coin Recognition and Sorting System”, in Proceedings of the 7th International Conference on Digital Image Computing—Techniques and Applications (DICTA '03), vol. 1, pp. 329–338, Sydney, Australia, December 2003.
[4]R.Bremananth, B.Balaji, M.Sankari and A.Chitra, “A New Approach to Coin Recognition using Neural Pattern Analysis”, IEEE Conference, Chennai, India Dec 2005, pp.11-13.
[5]Reinhold Huber, Herbert Ramoser, Konrad Mayer, Harald Penz, Michael Rubik, “Classification of coins using an eigenspace approach”, Pattern Recognition, 2005, pp.61-75.
[6]Marco Reisert, Olaf Ronneberger, Hans Burkhardt, “An Efficient Gradient Based Registration Technique for Coin Recognition”, Proceedings of the Muscle CIS Coin Competition Workshop, Sep.11, 2006, Berlin, Germany, pp. 19-31.
[7]P.Thumwarin, S.Malila, P.Janthawong, W.Pibulwej, T.Matsuura, “A Robust Coin Recognition Method with Rotation Invariance”, IEEE, 2006.
[8]Ali Ajdari Rad, Karim Faez, Navid Qaragozlou, “Fast Circle Detection Using Gradient Pair Vectors”, Proc. VIIth Digital Image Computing: Techniques and Applications, Sun C., Talbot H., Ourselin S. and Adriaansen T. (Eds.), 10-12 Dec. 2003, Sydney.
[9]Teh-Chuan Chen, Kuo-Liang Chung, “An Efficient Randomized Algorithm for Detecting Circles”, Computer Vision and Image Understanding 83, 172–191 , 2001.
[10]Victor Ayala-Ramirez , Carlos H. Garcia-Capulin, Arturo Perez-Garcia, Raul E. Sanchez-Yanez, “Circle detection on images using genetic algorithms”, Elsevier Pattern Recognition Letters 27, 2006, pp.652–657.
[11]黃仲裕, “不同光線下彩色影像色彩校準之研究(The Study of Color Corrections for Color Images Under Varying Illuminants)”, 國立東華大學資訊工程學系論文, 2002.