跳到主要內容

臺灣博碩士論文加值系統

(18.97.14.87) 您好!臺灣時間:2025/02/09 09:20
字體大小: 字級放大   字級縮小   預設字形  
回查詢結果 :::

詳目顯示

: 
twitterline
研究生:陳致豪
研究生(外文):Chen, Chihhao
論文名稱:以膚色偵測加速AdaBoost人臉偵測
論文名稱(外文):Speedup AdaBoost Face Detection by Skin Color Detection
指導教授:黃樹林黃樹林引用關係張創然張創然引用關係
指導教授(外文):Hwang, ShulinChang, Chuangjan
口試委員:阮聖彰
口試委員(外文):Ruan, Shanqjang
口試日期:2011-07-08
學位類別:碩士
校院名稱:明志科技大學
系所名稱:電子工程研究所
學門:工程學門
學類:電資工程學類
論文種類:學術論文
論文出版年:2011
畢業學年度:99
語文別:中文
論文頁數:59
中文關鍵詞:膚色偵測人臉偵測動態影像Adaboost
外文關鍵詞:skin color detectionface detectiondynamic imagesAdaboost
相關次數:
  • 被引用被引用:6
  • 點閱點閱:934
  • 評分評分:
  • 下載下載:298
  • 收藏至我的研究室書目清單書目收藏:0
人臉偵測(Face Detection)的研究源自於人臉辨識(Face Recognition),意旨在影像中判斷是否有人臉存在,並進一步確認人臉的位置。在影像系統普及的現代,隨著電腦視覺(Computer Vision)的興起,人臉偵測的研究不斷地迅速發展。本文提出了適用於動態影像下的多姿態人臉偵測法,主要分為下列四部份:(1)實做Viola提出基於AdaBoost的人臉偵測法,並使用OpenCV官方所提供的三個分類器(Cascade)進行效能分析。(2)利用膚色偵測(Skin Color Detection)提出一個適用於動態影像(Dynamic Images)的兩階段式人臉偵測法,實驗結果顯示,如此能有效降低檢索區塊的維度,除了能縮短偵測時間,也能減少錯誤的偵測。(3)提出以雙眼取代全臉特徵的偵測方法,實驗結果顯示,雖然實際偵測率會下降15~17%,但是偵測速度可以提昇3倍多。(4)針對複雜的背景、不同膚色的人種、以及人臉的多姿態進行實驗,同時也對常見的攝影問題,如偵測距離、模糊度、明亮度、對比度此四種屬性進行測試,以各種實驗來驗證本系統的穩健性。
Study on human face detection is originated from human face recognition, and it aims to recognize human face in images and identify the position of human face. In modern time, with the prevalence of imaging system and the rise of computer vision, researches on human face detection develop rapidly. This study proposes a human face detection method based on dynamic images. The paper is organized into four parts. 1) It first implements the human face detection proposed by Viola upon AdaBoost, and analyze efficacy by cascade provided by OpenCV. 2) It then develops two-stage human face detection for dynamic image by detection of the color of skin. After reducing the dimension of searching area, it can reduce detection, and avoid the errors. 3) It proposes the detection to replace the whole face characteristics by eyes. Experimental result showed that although actual detection rate will reduce by 15%, detection speed can increase by 3 times. 4) Experiments were conducted on complicated background, races of different colors of skin and multiple postures of human face, and tested the common issues of photography, such as detection distance, opacity, brightness and contrast. The experimental results confirmed that the system is stable human face detection.
明志科技大學碩士學位論文指導教授推薦書 i
明志科技大學碩士學位論文口試委員審定書 ii
明志科技大學學位論文授權書 iii
誌謝 iv
中文摘要 v
Abstract vi
目錄 vii
表目錄 x
圖目錄 xi
第一章 緒論 1
1.1研究動機與目的 1
1.2論文架構 3
第二章 背景技術介紹 4
2.1人臉偵測的研究方法簡介 4
2.1.1基於知識的方法(Knowledge-Based Methods) 4
2.1.2基於特徵的方法(Feature-Based Methods) 4
2.1.3基於模板匹配的方法 ( Template Matching ) 5
2.1.4基於統計理論的方法 (Statistical-based Methods) 5
2.2結合Adaboost演算法的人臉偵測 6
2.2.1 AdaBoost 演算法 6
2.2.2 Haar like矩形特徵 7
2.2.3積分影像 9
2.2.4串接式決策分類器 (Cascade Classifier) 10
2.3色彩空間與膚色偵測 12
2.3.1 RGB色彩空間 12
2.3.2 YIQ色彩空間 13
2.3.3 YCbcr色彩空間 14
2.3.4 HSV色彩空間 14
2.3.5膚色偵測技術 15
2.3.6色彩空間的選擇 16
2.4 Open CV 17
第三章 Adaboost人臉偵測與膚色偵測 18
3.1實作基於AdaBoost的人臉偵測 18
3.1.1訓練階段 19
3.1.1.1正樣本與負樣本 19
3.1.1.2 CreateSamples 19
3.1.1.3 HaarTraining 20
3.1.2.偵測階段 22
3.1.2.1 cvHaarDetectObjects 23
3.2 AdaBoost的人臉偵測法之改進分析 24
3.2.1兩階段式人臉偵測流程 25
3.2.2高複雜度的人臉偵測 26
3.2.3膚色分析與建模 27
3.2.4膚色偵測 30
3.2.5矩形比例與大小過濾 30
第四章實驗架構與分析 34
4.1人臉偵測的效能評估方法 35
4.2實驗環境 36
4.3官方分類器效能實驗與分析 37
4.4兩階段式人臉偵測實驗分析 39
4.5以雙眼分類(eye)器替換人臉(alt2)分類器的實驗數據 40
4.6系統穩健性實驗 43
4.6.1偵測距離實驗 43
4.6.2模糊度實驗 46
4.6.3對比度實驗 48
4.6.4亮度實驗 51
第五章 結論與未來展望 55
5.1結論 55
5.2未來展望 56
參考文獻 57

[1] R. Xiao, M.-J. Li, H.-J. Zhang, (2004). Robust multipose face detection in images. IEEE Transactions on Circuits and Systems for Video Technology, 14(1), 31- 41.
[2] 俞玉莲,王耀明(2005)。彩色图像的层次式人脸检测算法。上海师范大学数理信息学院,上海。
[3] Xinge You, Dan Zhang, & Qiuhui Chen. (2006). Face Representation By Using Non-tensor Product Wavelets. 18th International Conference on Pattern Recognition. (1),503-506.
[4] Viola, P., & Jones, M. (2001). Rapid object detection using a boosted cascade of simple features. IEEE Computer Society Conference on Computer Vision and Pattern Recognition. (1),I-511- I-518.
[5] M.-H. Yang, Kriegman, D. J., & Ahuja, N. (2002). Detecting faces in images: a survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 24(1), 34-58.
[6] Freund, Y., & Schapire, R. E. (1995). A decision-theoretic generalization of on-line learning and an application to boosting. Second European Conference on Computational Learning Theory, 23–37.
[7] Lienhart, R., & Maydt, J. (2002). An extended set of Haar-like features for rapid object detection. International Conference on Image Processing. (1), I-900- I-903.
[8] 赵楠,查红彬(2005)。基于 AdaBoost 算法的人脸检测。北京大学物理学院物理学系,北京。
[9] "RGB color model". Retrieved July 1, 2011, from http://en.wikipedia.org/wiki/RGB_color_model
[10] "YIQ". Retrieved July 1, 2011, from http://en.wikipedia.org/wiki/YIQ
[11] "Ycbcr". Retrieved July 1, 2011, from http://en.wikipedia.org/wiki/Ycbcr
[12]"HSL and HSV". Retrieved July 1, 2011, from http://en.wikipedia.org/wiki/HSL_and_HSV
[13] Jones, M. J, & Rehg, J. M. (1999). Statistical color models with application to skin detection. IEEE Computer Society Conference. 1 (274-280).
[14] Son Lam Phung, Bouzerdoum, A., & Chai, D. (2002b). A novel skin color model in YCbCr color space and its application to human face detection. 2002 International Conference on Image Processing. 1(I-289- I-292).
[15] Yang, M.-hsuan, & Ahuja, N. (1999). Gaussian Mixture Model for Human Skin Color and Its Applications in Image and Video Databases. ITS APPLICATION IN IMAGE AND VIDEO DATABASES. 3656(458-466)
[16] Shin, M. C., Chang, K. I., & Tsap, L. V. (2002). Does colorspace transformation make any difference on skin detection?. Sixth IEEE Workshop on Applications of Computer Vision. 275- 279.

[17] Hammami, M., Tsishkou, D., & Liming Chen. (2004). Adult content Web filtering and face detection using data-mining based kin-color model. IEEE Internation Conference on Multimedia & Expo. 1( 403-406).
[18] "OPEN CV". Retrieved July 1, 2011, from https://code.ros.org/gf/project/opencv/scmsvn/
[19] "PIE Database". Retrieved July 1, 2011, from http://www.ri.cmu.edu/research_project_detail.html?project_id=418&menu_id=261
[20] "CBCL SOFTWARE". Retrieved July 1, 2011, from http://cbcl.mit.edu/software-datasets/FaceData2.html
[21] "The Color FERET Database". Retrieved July 1, 2011, from http://face.nist.gov/colorferet/
[22]"Ara Nefian Face Recognition Page". Retrieved July 1, 2011, from http://www.anefian.com/research/face_reco.htm
[23] D.-S. WEI, L.-Q. LI, Improvement of Adaboost face detection. Graduate University of Chinese Academy of Sciences, Beijing.
[24] 周彥佑,黃樹林(2010)。以區域採樣最佳化提升人臉偵測於IP-CAM上之應用效能。明志科技大學電機工程所,新北市。
[25] Yu-Ting Pai, Li-Te Lee, Shanq-Jang Ruan, Yen-Hsiang Chen, Mohanty, S. P., & Kougianos, E. (2008). Honeycomb Model Based Skin Color Detector for Face Detection. 15th International Conference on Mechatronics and Machine Vision in Practice, 11-16.
[26] 許慕平,馮濤,仲澄,楊恒歡,荊銳 (2010)。复杂背景下一种基于多直方图的自适应人脸定位算法。上海第二工業大學學報,27(1)。
[27] 趙懷勛,徐鋒(2011)。基于HCbCr的人臉檢測方法。COMPUTER APPLICATIONS AND SOFTWARE,28(3)。

QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top