跳到主要內容

臺灣博碩士論文加值系統

訪客IP:44.211.49.158
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:賴昆煒
研究生(外文):Kun-Wei Lai
論文名稱:基於Radon轉換之人臉辨識系統
論文名稱(外文):A face recognition system based on the Radon transform
指導教授:李棟良李棟良引用關係
指導教授(外文):作者未提供
學位類別:碩士
校院名稱:銘傳大學
系所名稱:電腦與通訊工程學系碩士班
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:2012
畢業學年度:100
語文別:中文
論文頁數:50
中文關鍵詞:LDAPCARadon TransformOpenCV
外文關鍵詞:PCALDARadon TransformOpenCV
相關次數:
  • 被引用被引用:5
  • 點閱點閱:1011
  • 評分評分:
  • 下載下載:149
  • 收藏至我的研究室書目清單書目收藏:1
近年來生物認證技術逐漸普及,例如人臉辨識、虹膜辨識、指紋辨識、掌紋辨識、語音辨識等等。在這之中以人臉辨識最為廣泛使用,因為使用者並不需要穿戴額外的裝置,也不需要和受測裝置有任何接觸,只需要透過基本的攝影照相裝置,就可以得到辨識所需要的資料,因此,人臉辨識可以說是生物認證技術中最方便的方法了。
本論文提出以Radon Transform為基礎的即時人臉辨識系統。本論文在人臉偵測方面,利用了OpenCV的人眼偵測結果,來定位出人臉所在位置以及界定的臉部大小,再透過Radon Transform來判斷此人臉是否為正面影像。而在人臉辨識部分則是以PCA來降低資料的維度,接著再利用LDA的特性來提高特徵向量的鑑別性,藉此增加不同的類別中特徵向量間的相異性,最後將此影像特徵向量與資料庫內的資料向量做cosθ夾角的比對,當cosθ的值越大,代表此影像為資料庫的候選者之一。
實驗的部分,本論文使用自製人臉資料庫,一共15人,有300張影像,分別在四種不同的光源下拍攝,並且拍攝五種不同臉部角度的影像。其中10人為人臉資料庫樣本和內部測試樣本,並以不同數量的影像建立不同的人臉資料庫,剩下的影像當作內部測試的樣本,結果顯示內部資料庫辨識率能達到100%。另外5人共100張影像當作入侵偵測,結果顯示最佳的辨識結果能達到98%的正確偵測率(True Positive Rate)和100%的錯誤排除率(False Negative Rate)。由實驗結果顯示,本論文所提出以Radon為基礎的人臉辨識系統能達到理想的辨識結果。
In recent years, the bio-authentication technology such as the face recognition, iris recognition, fingerprint recognition, palm print recognition and voice recognition becomes popular. Among the face recognition is the most widely be used, because user don''t need to wear other devices and don''t need to contact with these devices. We can take the all information needed for face recognition only by use the simple camera, Therefore, the face recognition could be the most convenient recognition of the bio-authentication technology.
We proposed the real-time face recognition system based on the Radon transform. In the face detection, we use the eyes detection result of the OpenCV to find the where is the face and what size of the face. And then we use the Radon Transform to decide the human face is front face or not. In the face recognition, we use the PCA to reduce the data dimension and then we use the LDA to improve the discrimination of feature vectors to increase difference between different class feature vectors. Finally, we take this image of feature vectors to use the cosθ compare with the database. If the value of the greatest of the cosθ, it means this image is the human in the database.
In the experiment section, our face database is created by ourselves to perform all the simulations. The face database is contains 15 individuals, 300 images from four different light sources and five different perspective. We use 10 individuals (different image number) as training patterns, other images as testing patterns, the results show the best recognition rate can achieve 100%。The images of other 5 individuals (100 images) are used as invader test images, the best recognition result can achieve 98% true positive rate and 100% false negative rate. The experiment results showed that we proposed the face recognition system based on the Radon transform can achieve good recognition results.
摘要 i
ABSTRACT ii
致謝 iii
目錄 iv
圖目錄 vi
表目錄 vii
第一章 緒論 1
第一節 研究動機 1
第二節 研究目的 2
第三節 論文架構 2
第二章 人臉偵測與辨識之相關方法 3
第一節 相關文獻探討 3
一、 人臉偵測 3
二、 人臉辨識 4
第二節 人臉偵測方法介紹 5
一、 Viloa等人提出的Adaboost介紹 5
(一) 層疊分類器 5
(二) 積分影像(Integral Image) 6
(三) 矩形特徵、弱分類器、Adaboost演算法 7
二、 OpenCV介紹 9
第三節 人臉辨識方法介紹 10
一、 Radon Transform介紹 10
二、 主成份分析(Principle Component Analysis - PCA) 12
三、 線性鑑別分析(Linear Discriminate Analysis - LDA) 15
第三章 研究方法介紹 19
第一節 Adaboost人眼偵測 20
第二節 Radon轉換與傾斜修正 20
一、 影像擷取與正規化 20
二、 影像灰階化與遮罩 21
三、 Radon轉換與二值化 22
四、 傾斜修正 23
第三節 特徵擷取 23
第四節 資料建立與辨識 23
第四章 實驗結果 25
第一節 人臉資料庫 25
第二節 實驗結果 25
一、 內部資料庫之辨識結果 27
(一) 實驗一 27
(二) 實驗二 27
(三) 實驗三 29
(四) 實驗四 31
(五) 實驗五 33
二、 入侵偵測結果 33
(一) 實驗一 34
(二) 實驗二 35
(三) 實驗三 36
(四) 實驗四 37
(五) 實驗五 38
第五章 結論與未來展望 40
第一節 結論 40
第二節 未來展望 40
參考文獻 41
[1]Ming-Hsuang Yang, David J. Kriegman, narendra Ahuja, "Detecting Faces in Images: A Survey", IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, 2002, pp. 34-58.
[2]Andrea F, Abate, Michele Nappi, Daniel Riccio, Gabrele Sabatino, "2D and 3D face recognition: A survey", ScienceDirect Pattern Recognition Letters , vol. 28, 2007 pp. 1885-1906
[3]D. V. Jadhav, and R. S. Holambe, “Rotation, illumination invariant polynomial kernel Fisher discriminant analysis using Radon and discrete cosine transforms based features for face recognition,” Pattern Recognition Letters, Vol. 31, pp. 1002-1009, 2010
[4]D. V. Jadhav, and R. S. Holambe, “Feature extraction using Radon and wavelet transforms whit application to face recognition,” Neurocomputing, Vol. 72, pp. 1951-1959, 2009.
[5]Paul Viola, Michael Jones, "Robust Real-time Object Detection", Second International Workshop on Statistical and Computational Theories of Vision - Modeling, Learning, Computing, and Sampling, 2001.
[6]H. Araujo, J.Dias, "An introduction to the log-polar mapping", Proceedings of the Workshop on Cybernetic Vision 1996, 1997, pp. 139-144.
[7]I. Craw, D. Tock, and A. Bennett, "Finding Face Features," Proc. Second European Conf. Computer Vision, pp. 92-96, 1992.
[8]A. Lanitis, C.J. Taylor, and T.F. Cootes, "An Automatic Face Identification System Using Flexible Appearance Models," Image and Vision Computing, vol. 13, no. 5, pp. 393-401, 1995.
[9]G. Yang and T. S. Huang, "Human Face Detection in Complex background, "Pattern Recognition, vol. 27, no. 1, pp. 53-63, 1994.
[10]Lin-Lin Huang, Akinobu Shimizu, Yoshihiro Hagihara, Hidefumi Kobatake, "Face detection from cluttered images using a polynomial neural network", Neurocomputting vol. 51, 2003, pp. 197-211.
[11]Lin-Lin Huang, Akinobu Shimizu, Yoshihiro Hagihara, Hidefumi Kobatake, "Gradient feature extraction for classification-based face detection", Pattern Recognition vol. 36, issue 11, 2003, pp. 2501-2511.
[12]K.C. Yow and R. Cipolla, "Feature-Based Human Face Detection," mage and Vision Computing, vol. 15, no. 9, pp. 713-735, 1997.
[13]T.K. Leung, M.C. Burl, and P. Perona, "Finding Faces in Cluttered  Scenes Using Random Labeled Graph Matching, "Proc. Fifth IEEE Int’l  Conf. Computer Vision, pp. 637-644, 1995.
[14]Y. Dai and Y. Nakano, "Face-Texture Model Based on SGLD and Its  Application in Face Detection in a Color Scene, "Pattern Recognition, vol. 29, no. 6, pp. 1007-1017, 1996.
[15]F. Hjouj and D. W. Kammler, “Identification of Reflected, Scaled, Translated, and Rotated Objects From Their Radon Projections,” IEEE Trans. Image Processing, vol. 17, no. 3. pp. 301–310, March 2008
[16]Y. Wan, and N. Wei, “A Fast Algorithm for Recognizing Translated, Rotated, Reflected, and Scaled Objects From Only Their Projections,” IEEE Signal Processing Letters, Vol. 17, No. 1, pp. 71-74, Jan. 2010.
[17]X. Wang, B. Xiao, J.-F. Ma, and X-L Bi, “Scaling and rotation invariant analysis approach to object recognition based on Radon and Fourier–Mellin transforms,” Pattern Recognition, Vol. 40, pp. 3503-3508, 2007.
[18]K. Jafari-Khouzani and H. Soltanian-Zadeh, “Rotation-Invariant Multiresolution Texture Analysis Using Radon and Wavelet Transforms,” IEEE Trans. Image Processing, vol. 14, no. 6. pp. 783–795, June 2005.
[19]梁振升,「以OpenCV實現即時人臉偵測與辨識系統」,私立銘傳大學,電腦與通訊工程學系碩士論文,指導教授:李棟良,2010年。
[20]http://www.opencv.org.cn/index.php/%E9%A6%96%E9%A1%B5
[21]A.M. Martinez and R. Benavente, "The AR face database," CVC Tech. Report #24, 1998.
[22]Yi Li and Baochang Zhang, “Bagging Base Efficient Kernel Fisher Discriminant Analysis for Face Recognition,” Pattern Recognition, 2006.
[23]Kazuhiro Hotta, “Local normalized linear summation kernel for fast and robust recognition,” Pattern Recognition, Vol.43, pp.906-913, 2010.
QRCODE
 
 
 
 
 
                                                                                                                                                                                                                                                                                                                                                                                                               
第一頁 上一頁 下一頁 最後一頁 top
無相關期刊