臺灣博碩士論文加值系統

物件(object)之形狀識別(shape recognition)在電腦視覺(computer vision)上是非常重要的,特別是在自動化製造(automatic manufacturing)上.而形狀失真(shape distortion)主要有四種形式：位移(Translation),放大縮小(scaling),旋轉(rotation)及偏斜(skew)；本文提出一個形狀識別的方法去處理這四種形狀失真.首先以影像緊密法(image compacting)將物件正規化(normalizing)至其分散矩陣(dispersion matrix)最緊密的狀態去處理失真,然後用樣板比對(template matching)與建立好的緊密影像物件樣板資料庫做比對.

Shape recognition is very important in computer vision, especially in automatic manufacturing. There are four basic shape distortion forms: translation, rotation, scaling and skew. To recognize the object shap under these forms, we first compact the object such that its dispersion matrix is in the most compact form, then use the template matching algorithm to match the compact shape image in two opposite directions, and thus recognize the object.

中文摘要 i
ABSTRACT ii
誌謝 iii
目 錄 iv
表目錄 vi
圖目錄 viii
第一章 緒論 1
1.1研究動機與目的 1
1.2文獻探討 1
1.3論文架構 3
第二章 發展系統架構 4
2.1發展環境 4
2.1.1開發工具 4
2.1.2硬體配備 4
2.2研究範圍與限定 5
2.3系統流程 5
第三章 物件形狀辨識 7
3.1影像前處理 7
3.1.1影像灰階化處理 8
3.1.2二值化與反向判定 9
3.2影像緊密法 12
3.2.1位移 13
3.2.2旋轉 14
3.2.3放大縮小與正規化 16
3.3資料庫比對 18
第四章 實驗結果與討論 23
4.1物件影像與樣板資料庫比對 23
4.2加雜訊之物件影像與樣板資料庫比對 49
第五章 結論與未來研究方向 52
5.1結論 52
5.2未來研究方向 52
參考文獻 54
附錄A 57

[1]L. J. Latecki, R. Lakamper, “Application of planar shape comparison to object retrieval in image databases,” Pattern Recognition, Vol. 35, pp 15-29, 2001.
[2]C. T. Zahn and R. S. Roskies, “Fourier descriptors for plane closed curves,” IEEE Trans. Comput., Vol. 21, pp 269-281, 1972.
[3]E. Persoon and K. S. Fu, “Shape discrimination using Fourier descriptors,” IEEE Trans. Syst., Man. Cybern., Vol. 7, pp 170-179, March 1977.
[4]曾怜玉, “使用小波描述子和傅立葉描述子為基礎之二維物件辨識系統,” 中興大學應用數學研究資訊組, 2000.
[5]劉蓉, “使用傅立葉描述子在支持向量機與自我組織映射圖於形狀影像檢索之研究,” 中華大學資訊管理學系, 2006.
[6]M. K. Hu, “Visual pattern recognition by moment invariants,” IRE Trans. Information Theory, pp 179-187, Feb. 1962.
[7]S. O. Belkasim, M. Shridhar and M. Ahmadi “Pattern recognition with moment invariants: a comparative study and new results,” Pattern Recognition, Vol. 24, pp 1117-1138, Dec. 1991.
[8]D. H. Ballard, “Generalizing the Hough transform to detect arbitrary shape,” Pattern Recognition, Vol. 13, pp 111-122, Feb. 1981.
[9]S. C. Jeng and W. H. Tsai, “Scale and orientation invariant generalized Hough transform --- a new approach,” Pattern Recognition, Vol. 24, pp 1037-1051, Nov. 1991.
[10]徐偉邦, ”運用影像分割技術於精密零件幾何形狀辨識,” 元智大學機械工程研究所, 2005.
[11]彭國軒, ”快速物件辨認與定位–環狀樣板比對,” 清華大學動力機械工程學系, 2003.
[12]A. Delopoulos, A. Tirakis and S. Kollias, “Invariant image classification using triple-correlation-based neural networks,” IEEE Trans. Neural Networks, Vol. 5, pp 392-407, March 1994.
[13]L. Spirkovska and M. B. Reid, “Robust position, scale, and rotation invariant object recognition using higher-order neural networks,” Pattern Recognition, Vol. 25, pp 975-985, Sep. 1992.
[14]李宗憲, ”以VQ編碼方法為基礎的物件偵測方法,” 中山大學機械與機電工程學系, 2002.
[15]曾智宏, “利用局部特徵不變量與Hopfield網路辨識重疊物件,” 中山大學機械與機電工程學系, 2003.
[16]李國清, “以輪廓特徵及模糊比對為基礎的平面圖形辨識系統,” 中興大學應用數學系, 2000.
[17]J. G. Leu, “Shape normalization through compacting,” Patt. Recogn. Lett., Vol. 10, pp 243-250, 1989.
[18]K. K. Yu, C. T. Chuang and R. H. Hwang, “Shape Recognition Using Compacting and Radial Ray Intersections,” in Proc. of 13th IPPR Conference on
Computer Vision, Graphics and Image Processing, Taipei, Taiwan, Aug. 20-22, Vol. 1, pp. 273-278, 2003.
[19]S. C. Pei and C. N. Lin, “Image normalization for pattern recognition,” Image and vision Computing, Vol. 13, pp. 711-723, 1995.
[20]Otsu, N., “A threshold selection method from gray-level histograms,” IEEE Trans. Systems Man Cybernet, vol. 9, no.1, January 1979, pp.62–69.
[21] R. C. Gonzalez and R. E. Woods., “Digital Image Processing,” 2nd Edition, 2002.