![]() |
|
以作者查詢圖書館館藏 、以作者查詢臺灣博碩士 、以作者查詢全國書目 、勘誤回報 、線上人數:36 、訪客IP:18.224.52.110
姓名 凌宙綸(Chou-Lun Ling) 查詢紙本館藏 畢業系所 資訊工程學系在職專班 論文名稱 利用鍊碼及符號描述做人臉五官精細定位
(The precise location for facial components based on chain code and symbol descriptor)相關論文 檔案 [Endnote RIS 格式]
[Bibtex 格式]
[相關文章]
[文章引用]
[完整記錄]
[館藏目錄]
[檢視]
[下載]
- 本電子論文使用權限為同意立即開放。
- 已達開放權限電子全文僅授權使用者為學術研究之目的,進行個人非營利性質之檢索、閱讀、列印。
- 請遵守中華民國著作權法之相關規定,切勿任意重製、散佈、改作、轉貼、播送,以免觸法。
摘要(中) 人臉五官定位是人臉識別的先置作業,也是追踪人臉及表情辨識的重要步驟。會遇到的問題包括環境光線的變異、人臉方向的改變、複雜的背景,及相機本身的特性。過去有許多方法被提出來,依需求來解決上述的困難。我們的人臉及五官定位將應用在固定相機的門禁系統上;因此要解決的問題大致只有光源的變異而已。
本論文包含三個主要議題: (i) 人臉定位,(ii) 鍊碼追踪與符號描述, (iii) 眼鼻口定位。第一個主題將探討擷取人臉的方法。第二主題是將人臉區域的輪廓轉為鍊碼和符號描述,提供匹配眼睛所需的符號描述資料。第三個主題探討如何利用符號描述來找出雙眼,之後再利用五官模板來找到鼻口。
用膚色偵測,是最直接取得人臉定位資訊的方法,不受複雜背景的影響,也不需要經過學習與分類。所以我們採用膚色定位的方法,將偵測出來的膚色區塊交給下一步來判定人臉。判定人臉的方法,使用邊偵測後,鍊碼追踪再轉成符號描述群組並檢測群組是否存在人臉五官比例,這樣的做法能獲得較可靠的判定結果。
經過判定的人臉,就可以使用符號描述群組的索引找雙眼最後再以人臉模板匹配出鼻和口。我們定義偵測區塊要包含臉部器官的70% 範圍,且偵測區域面積不大於該器官面積的2 倍,才算偵測到該器官。總共使用56張照片,共有63個人臉做實驗,而五官定位方法經實驗測試在戶外偵測左右眼、鼻、口及取得各五官範圍可達71.4% 的偵測率,在室內的大頭照則可達到77.1% 的偵測率;採用MUCT人臉資料庫取用150樣本偵測眼、鼻和口符號群組達74.7% 的偵測率。未來將對頭髪和眼鏡結構來進一步探討,就可以再提高定位的機率。
摘要(英) The face-feature location is a pre-task of face recognition; it’s also an important procedure for facial expression and face tracking. The encounted problems in the face-feature location include the change of ambient light, the direction the face, the complicated background, and the sensation of camera. Many methods have been proposed to resolve these problems.
This paper consists of three main topics: (i) face detection, (ii) chain code tracking for faces, (iii) and face-feature location. The first topic discusses an approach of face detection. The second topic describes how to transfer the face contour and face-feature shapes into chain code representation and symbol description to match and verify eyes. The third topic utilizes the symbol description to find the locations of nose and mouth by face-feature template.
The skin color is the most useful characteristic for detecting faces without affecting by the complicated background. We here adopt the skin color to extract face candidate following face verification. The verification procedure uses edge detection and chain code representation to describe the face. We feel that based on such an approach, we can get more reliable results.
After the face extraction, the eyes are found by utilizing the symbol description and the nose and mouth are found by matching the face template. Experiments show that eye, nose, and mouth were correctly located at a rate of 71.4% in outdoors, and at a rate of 77.1% for indoors. In the future, we will explore the structure of hair and glasses to enhance the detection rate.
關鍵字(中) ★ 人臉定位
★ 五官定位
★ 膚色偵測
★ 鍊碼
★ 符號描述
★ 邊偵測
★ 網格關鍵字(英) ★ face feature
★ location of face
★ skin color
★ symbol descriptor
★ lattice
★ sobel
★ chaincode論文目次 第 1 章 緒論 ........................................................................................................ 1
1.1研究動機 ................................................................................................ 1
1.2研究方法 ................................................................................................ 1
1.2.1膚色偵測 ...................................................................................... 2
1.2.2邊偵測 .......................................................................................... 2
1.2.3輪廓與鍊碼 .................................................................................. 3
1.2.4五官定位 ...................................................................................... 3
1.3論文架構 ................................................................................................ 5
第2 章 相關研究 ................................................................................................ 6
2.1邊偵測方法比較 .................................................................................... 6
2.1.1 Sobel 邊偵測 ............................................................................. 6
2.1.2 Canny 邊偵測 ............................................................................ 7
2.2人臉與五官定位方法 ............................................................................ 9
2.2.1膚色定位 ...................................................................................... 9
2.2.2人臉特徵 ...................................................................................... 9
2.2.3人臉模板匹配 ............................................................................ 10
2.2.4 AdaBoost 測檢方法 .................................................................. 10
2.3模式識別 .............................................................................................. 12
第3 章 膚色和人臉偵測 .................................................................................. 15
3.1膚色偵測方法 ...................................................................................... 15
3.2人臉判定方法 ...................................................................................... 16
第4 章 邊偵測的方法 ...................................................................................... 19
4.1視覺作用 .............................................................................................. 19
4.1.1觀察視覺作用 ............................................................................ 19
4.1.2對比產生的輪廓線條 ................................................................ 20
4.2網格擷取線條架構 .............................................................................. 20
4.3網格擷取與Sobel 邊偵測的比較 ....................................................... 29
第5 章 鍊碼追踪到符號描述 .......................................................................... 31
- vii -
5.1鍊碼的編碼 .......................................................................................... 31
5.1.1分割區塊 .................................................................................... 31
5.1.2輪廓的鍊碼追踪 ........................................................................ 32
5.1.3鍊碼去雜訊及修正鋸齒 ............................................................ 34
5.2鍊碼到符號描述 .................................................................................. 36
5.2.1抽取鍊碼線條的特性 ................................................................ 36
5.2.2加入節點 .................................................................................... 37
5.2.3符號描述編成 ............................................................................ 39
5.2.4輪廓橡皮擦 ................................................................................ 41
5.3人臉模板的人臉判定 .......................................................................... 43
第6 章 眼鼻口的定位 ...................................................................................... 46
6.1利用符號描述找雙眼 .......................................................................... 46
6.1.1將符號描述依座標組成群組 .................................................... 46
6.1.2符號描述的線段轉成4種線性符號 ........................................ 48
6.1.3符號描述群組與9宮格相對位置配置 .................................... 49
6.1.4定義眼部符號描述 .................................................................... 50
6.1.5搜尋規則 .................................................................................... 52
6.1.6符號描述群組匹配眼睛符號描述的方法 ................................ 55
6.2 人臉模板找鼻子和嘴巴 ..................................................................... 56
6.2.1人臉五官模板的取得 ................................................................ 56
6.2.2找鼻子和嘴巴 ............................................................................ 58
第7 章 實驗 ...................................................................................................... 59
7.1使用工具介紹 ...................................................................................... 59
7.2綜合環境人臉五官定位測試 .............................................................. 59
7.3 MUCT人臉資料庫測試 ....................................................................... 69
第8 章 結論 ...................................................................................................... 72
參考文獻 .............................................................................................................. 74
參考文獻 pp. 579-583, 1978.
[2] Belhumeur, P. N., J. P. Hespanha and D. J. Kriegman, "Eigenfaces vs. fisherfaces: recognition using class specific linear projection," IEEE Trans. on Pattern Analysis and Machine Intelligence, vol.19, no.7, pp.711-720, 1997.
[3] Beus, H. L. and S. S. H. Tiu, "An improved corner detection algorithm based on chain-coded plane curves," Pattern Recognition, vol.20, no.3, pp.291-296, 1987.
[4] Bradski, G. and A. Kaehler, Learning OpenCV: Computer Vision with the OpenCV, O’’Reilly Media, Sebastopol, CA, 2008.
[5] Canny, J. F., "A computational approachto edge detection," IEEE Trans. on Pattern Analysis and Machine Intelligence, vol.8, no.6, pp.679-698, 1986.
[6] Chen, L.-F., H.-Y. M. Liao, M.-T. Ko, J.-C. Lina, and G.-J. Yu, "A new LDA-based face recognition system which can solve the small sample size problem," Pattern Recognition, vol.33, no.10, pp.1713-1726, 2000.
[7] Chow, C. K. , "An optimum character recognition system using decision functions," IRE Trans. Electronic Computers, vol.6, no.4, pp.247-254, 1957.
[8] De Leon, R. D. and L. E. Sucar, "Human silhouette recognition with fourier descriptors," in Proc. 15th Int. Conf. on Pattern Recognition, Barcelona , Spain, Sep.3-7, 2000, pp.709-712.
[9] Freeman, H. and L. S. Davis, "A corner-finding algorithm for chain-coded curves," IEEE Trans. on Computers, vol.26, no.3, pp.297-303, 1977.
- 75 -
[10] Fu, A. M. N. and H. Yan, "A curve bend function based method to characterize contour shapes," Pattern Recognition, vol.30, no.10, pp.1661-1671, 1997.
[11] Fu, A. M. N. and H. Yan, "Effective classification of planar shapes based on curve segment properties," Pattern Recognition Letters, vol.18, no.1, pp.55-61, 1997.
[12]Gonzalez, R. C. and R. E. Woods, Digital Image Processing, 2nd Edition, Prentice Hall, Upper Saddle River, New Jersey, 2002, Ch.9.
[13] Han, C.-C., H.-Y. M. Liao, G.-J. Yu, and L.-H. Chen, "Fast face detection via morphology-based pre-processing," Pattern Recognition, Vol.33, pp.1701-1712, 2000.
[14] Hayashi, S. and O. Hasegawa, "A detection technique for degraded face images," IEEE Conf. on Computer Vision and Pattern Recognition, New York, Jun.17-Jun.22, 2006, pp.1506-1512.
[15] Hsu, R.-L., M. A. Mottaleb, and A. K. Jain, "Face detection in color images," IEEE Trans. Pattern Analysis and Machine Intelligence, vol.24, no.5, pp.696-706, 2002.
[16] Jarvis, R. A., "Focus optimization criteria for computer image processing," Microscope, vol.24, no.2, pp.163-180, 1976.
[17] Kaneko, T. and M. Okudaira, "Encoding of arbitrary curves based on the chain code representation," IEEE Trans. Communications, vol.33, no.7, pp.697-707, 1985.
[18] Kim, H., D. Kim, and S. Y. Bang, "Face recognition using LDA mixture model," Pattern Recognition Letters, vol.24, no.15, pp.2815-2821, 2003.
[19] Koplowitz, J. and S. Plante, "Corner detection for chain coded curves," Pattern Recognition, vol.28, no.6, pp.843-852, 1995.
[20] Krotkov, E. P, Active Computer Vision by Cooperative Focus and Stereo, Springer, New York, 1989.
[21] Rosenfeld, A. and J. S. Weszka, "An improved method of angle detection on digital curves," IEEE Trans. on Computers, vol.24, no.10, pp.940-941, 1975.
- 76 -
[22] Rupp, R., Committed to Memory: How We Remember and Why We Forget, Diane Pub Co, Darby, Pennsylvania, 1998.
[23] Siana, L., A Study of Human Tracking and Face Detection on A Pantilt-zoom Camera, Inst. of Electrical and Control Engineering National Chiao-Tung Univ., Master Thesis, 2005.
[24] Soriano, M., B. Martinkauppi, S. Huovinen, and M. Laaksonen, "Using the skin locus to cope with changing illumination conditions in color-based face tracking," in Proc. IEEE Nordic Signal Processing Symposium, Kolmarden, Sweden , July 13-15, 2000, pp.383-386.
[25] Tenenbaum, J. M., Accommodation in Computer Vision, Ph.D Thesis, Elect. Eng. Dept., Stanford University, 1970.
[26] Viola, P. and M. Jones, "Rapid object detection using a boosted cascade of simple features," Computer Vision and Pattern Recognition, vol.1, pp.511-518, 2001.
[27] Wang, J. G. and T. N. Tan, "A new face detection method based on shape information," Pattern Recognition Letters, vol.21, pp.463-471, 2000.
[28] Yang, J., D. Zhang, A. F. Frangi, and J. Yang, "Two-dimensional PCA: a new approach to appearance-based face representation and recognition," IEEE Trans. on Pattern Analysis and Machine Intelligence, vol.26, no.1, pp.131-137, 2004.
[29] You, Z. and A. K. Jain, "Performance evaluation of shape matching via chord length distribution," Computer Vision, Graphics and Image Processing, vol.28, pp.185-198, 1984.
[30] 王彥鈞, 利用 LDA 進行特定人物辨識之方法, 國立台灣科技大學,機械工程系, 碩士論文, 2008.
[31] 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, 國立中興大學, 應用數學系, 碩士論文, 2009.
[32] 林文章, 不同場景的膚色偵測與臉部定位, 國立中央大學, 資訊工程研究所, 碩士論文, 2009.
[33] 林咸仁, 改良線性鑑別式分析在少量訓練樣本下之人臉辨識研究, 國立成功, 大學資訊工程學系, 碩士論文, 2002.
- 77 -
[34] 林俊男, 一種加強影像邊緣銳化的影像放大方法, 元智大學, 資訊工程研究所, 碩士論文, 2002.
[35] 傅京孫, 模式識別及其應用, 科學出版社, 北京, 1983.
[36] 藍寅峻, 以假設邊界為基礎的適應性數位影像放大技術, 國立臺灣大學, 資訊工程學系, 碩士論文, 1999.
[37] 鐘仁厚, 基於模糊邏輯之臉部表情辨識, 國立中央大學, 電機工程學系, 碩士論文, 2008.
指導教授 曾定章(Din-Chang Tseng) 審核日期 2012-1-16 推文 plurk
funp
live
udn
HD
myshare
netvibes
friend
youpush
delicious
baidu
網路書籤 Google bookmarks
del.icio.us
hemidemi
myshare