臺灣博碩士論文加值系統

本論文提出了一個手勢識別系統。在手勢影像分割方面，我們利用HSL顏色模組中膚色範圍相關研究來抓取手勢區域，並對灰階影像中的像素以滑動視窗的方法統計成直方圖，從中找出影像的分割門檻值。在手勢特徵擷取方面，包括橢圓形環切到手勢中之手指數、手指之間的角度、手指到手掌中心的長度，以及手勢輪廓的曲率值等。而手勢辨識方面，對各個特徵值作距離運算，分別求取近似程度，再以統計的方法找出手勢種類。實驗過程包含探討以不同型態特徵的辨識率、左右手的辨識率、不同手勢數的辨識時間和辨識率及手勢資料庫的篩選等。

A gesture identification system is proposed in this thesis. For image segmentation, the HSL information of the image is used to find the gesture region, the gray level histogram is applied to determine the threshold for image partition. In feature extraction, an ellipse mask is constructed to determine the features such as the number of fingers, the angles between fingers, the distances between the fingers and the palm, and the curvature of the palm’s contour. In the recognition procedure, a similarity measure is achieved by calculating the distance between features, which is used to determine the most likely gesture. The experiments processes include the recognition rate using different features; the recognition rate without the restriction of left or right hand, the recognition speed, and the sieve of gesture database.

中文摘要……………………………………………………………………I
英文摘要……………………………………………………………………II
目錄…………………………………………………………………………III
圖目錄………………………………………………………………………V
表目錄………………………………………………………………………VII
第一章 序論…………………………………………………………………1
1.1 前言…………………………………………………………………1
1.2 方法概論……………………………………………………………2
1.3 論文組織架構………………………………………………………2
第二章 相關研究……………………………………………………………3
第三章 系統架構與方法……………………………………………………6
3.1 影像前處理…………………………………………………………7
3.1.1 取得影像…………………………………………………………8
3.1.2 影像分割…………………………………………………………9
3.1.3 影像修正…………………………………………………………11
3.2 特徵抽取……………………………………………………………13
3.2.1 橢圓環取特徵……………………………………………………14
3.2.2 輪廓曲率…………………………………………………………17
3.3 特徵辨識……………………………………………………………22
第四章 實驗結果……………………………………………………………25
4.1 實驗資料……………………………………………………………25
4.2 實驗結果……………………………………………………………26
4.2.1 不同特徵型態的辨識率比較……………………………………26
4.2.2 左右手勢的辨識率比較…………………………………………28
4.2.3 不同手勢數的辨識時間和辨識率比較…………………………32
4.3 方法的改良…………………………………………………………32
4.3.1 以特徵向量平均值之辨識………………………………………32
4.3.2 以相似度篩選特徵向量…………………………………………33
4.4 即時手勢辨識………………………………………………………34
第五章 結論與未來展望……………………………………………………37
5.1 結論…………………………………………………………………37
5.2 未來展望……………………………………………………………38
參考文獻 ……………………………………………………………………39

[1] Omar Al-Jarrah and Alaa Halawani, ”Recognition of gestures in Arabic sign language using neuro-fuzzy systems, ”Artificial Intelligence 133, pp. 117–138, 2001.
[2] Francisco Flórez Revuelta, Juan Manuel García Chamizo and Francisco Ibarra Picó, ”Hand Gesture Recognition based on Morphologic Features, ” Universidad de Alicante Departamento de Tecnología Informática y Computación Campus de S. Vicente del Raspeig Ap. de Correos 99, E-03080, Alicante. www.ua.es/i2rc/florez01(snrfai).pdf
[3] Chin-Chen Chang and Yea-Shuan Huang, ”Novel Feature Extraction and Matching Approach for Hand Posture Recognition, ” IPPR Conference on Computer Vision, Graphic and Image Processing, pp. 717-722, 2002.
[4] Chao-Hui Huang, ”Silhouette-Based Hand Gesture Recognition System, ”Department of computer Science and Information Engineering Chung Hua university, 2001.
[5] Lalit Gupta and Suwei Ma, ”Gesture-Based Interaction and Communication: Automated Classification of Hand Gesture Contours, ” IEEE Transactions On Systems Man And Cybernetics Part C: Applications And Reviews, Vol. 31, No. 1, pp. 114-120, 2001.
[6] Jui-Chen Wu, Yung-Sheng Chen and I-Cheng Chang, ”An automatic approach to extracting facial features, ” IPPR Conference on Computer Vision, Graphic and Image Processing, pp. 178-184, 2002.
[7] Rong-Qin Luo and Wei-Zhi Luo, ”Human Face Detection and Recognition by Improved Gray Matching Based on Geometric Invariant Features, ” IPPR Conference on Computer Vision, Graphic and Image Processing, pp. 56-65, 2002.
[8] Rafael C. Gonzalez and Richard E. Woods, ”Digital Image Processing, ”儒林圖書, 2001.
[9] Nils Karlsson, Bjorn Karlsson and Peter Wide, ”A Glove Equipped with Finger Flexion Sensors as a Command Generator used in a Fuzzy Control System, ” IEEE Transactions on Instrumentation and measurement, Vol. 47, No. 5,pp. 1330-1334, 1998.
[10] Cheng-Chang Lien and Chung-Lin Huang, ”The model-based dynamic hand posture identification using genetic algorithm, ” Machine Vision and Applications, pp. 107-121, 1999.
[11] Chih-Ming Fu, Chung-Lin Huang and Chih-How Jeng, ”A Real Time Hand Gesture Tracking For Human Computer Interface, ” IPPR Conference on Computer Vision, Graphic and Image Processing, pp. 579-585, 2002.
[12] Jong-Sung Kim, Won Jang and Zeungnam Bien, ”A Dynamic Gesture Recognition System For the Korean Sign Language(KSL), ”IEEE Transactions on Systems Man And Cybernetics Part B : Cybernetics, Vol. 26, No. 2, pp. 354-359, 1996.
[13] Mu-Chun Su, ”A Fuzzy Rule-Based Approach to Recognition 3-D Arm Movements, ” IEEE Transactions on Neural Systems And Rehabilitation Engineering, Vol. 9, No.2,2001.
[14] Mu-Chun Su, ”A Fuzzy Rule-Based Approach to Spatio-Temporal Hand Gesture Recognition, ” IEEE Transactions on Systems Man And Cybernetics Part C: Applications and Reviews, Vol. 30, No. 2, pp. 276-281, 2000.
[15] Rung-Huei Liang and Ming Ouhyoung, “A Real-time Continuous Gesture Recognition System for Sign Language, ”Proceeding of the IEEE International Conference on Automatic Face and Gesture Recognition, Vol. 3 pp. 558-567, 1998.
[16] 譚永恒, ”以數位影像處理技術做人臉自動追蹤系統之研究,”國立成功大學電機工程學系碩士論文, 2000.
[17] 柳美鈴, ”使用 K曲率法則於二維物體的分斷點偵測之研究,”朝陽科技大學工業工程與管理技術系碩士論文, 2000.
[18] 廖英翔, ”印刷電路版佈線弧度之自動化檢測與製程管制探討,”朝陽科技大學工業工程與管理系碩士論文, 2002.
[19] 王國榮, ”基於資料手套的智慧型手勢辨識之廣泛研究,”台灣科技大學電機所碩士論文,2001.
[20] 陳錦輝等,”C++ Builder 6完全功略,”金禾資訊, 2003.