基于形状无关纹理和Boosting学习的人口统计学分类

杨之光; 艾海舟

doi:10.3724/SP.J.1146.2006.01328

基于形状无关纹理和Boosting学习的人口统计学分类

doi: 10.3724/SP.J.1146.2006.01328

基金项目:

国家自然科学基金 (60332010, 60673107)资助课题

计量
- 文章访问数: 3005
- HTML全文浏览量: 80
- PDF下载量: 1272
- 被引次数: 0
出版历程
- 收稿日期: 2006-09-06
- 修回日期: 2007-01-30
- 刊出日期: 2008-03-19

Demographical Classification by Shape Free Texture and Boosting Learning

摘要

摘要: 基于形状无关纹理和boosting学习，该文提出了对性别和年龄分类的方法，其中年龄被划分为儿童、青年、中年和老年4类。检测到人脸后，利用人脸配准的结果规范化人脸图像获得形状无关纹理。在此基础上提取Haar型特征、LBP直方图和Gabor Jet 3种特征，通过boosting学习分别训练分类器。实验表明，LBP直方图特征能够鲁棒地区分儿童和老人，Haar型特征用作区分青年和中年人则更为有效，而Gabor Jet特征更适于性别分类。
- 人口统计学分类; 人脸图像处理; Boosting
Abstract: In this paper, a gender and age classification method, in which age is classified into four classes: child, youth, midlife and agedness, based on shape free texture and boosting learning is introduced. After a face is detected, face alignment extracts 88 facial landmarks by which the face image is normalized to a shape free texture. Further more, three kinds of local feature, Haar like feature, LBP histogram and Gabor jet are extracted from the shape free texture; and boosting learning method is used for training classifiers. The experimental results show that, LBP histogram can be used for robust recognition of children and old people, Haar like feature is more efficient for discriminating young and middle aged people, and Gabor Jet fits for gender classification best.

HTML全文

参考文献(1)

Craw I, Tock D, and Bennett A. Finding face features. Proc. ECCV, Santa Margherita Ligure, Italy, 1992: 92-96.[2]Yuille A L, Hallinan P W, and Cohen D S. Feature extraction from faces using deformable templates. IJCV, 1992, 8(2), 99-111.[3]Young Ho Kwon and Niels da Vitoria Lobo. Age classification from facial images. Proc. CVPR, Seattle. Washington, USA, 1994: 762-767.[4]Turk M A and Pentland A P. Face recognition using eigenfaces. Proc. CVPR, Hawaii, USA, 1992: 586-591.[5]Lanitis A, Taylor C J, and Cootes T F. Modeling the process of ageing in face images. Proc. ICCV, Kerkyra, Greece , 1999: 131-136.[6]Moghaddam B and Yang M H. Gender classification with support vector machines. PAMI, 2002, 24(5): 707-711.[7]Shakhnarovich G, Viola P, and Moghaddam B. A unified learning framework for real time face detection and classification. Automatic face and gesture recognition, Washington, USA, 2002: 16-26.[8]Wu Bo, Ai Haizhou, and Huang Chang. LUT-based adaboost for gender classification. AVBPA, Guildford, 2003: 104-110.[9]Huang Chang and Ai Haizhou, et al.. Boosting nested cascade detector for multi-view face detection. ICPR, Cambridge, UK, 2004: 23-26.[10]Zhang Li and Ai Haizhou, et al.. Robust face alignment based on local texture classifiers. ICIP, Genoa, Italy, September 11-14, 2005: 354-357.[11]Yang Zhiguang, Li Ming, and Ai Haizhou. An experimental study on automatic face gender classification. ICPR, Hong Kong, China, 2006: 1099-1102.[12]Viola P and Jones M. Rapid object detection using a boosted cascade of simple features. CVPR, Kauai, Hawaii, USA, 2001: 511-518.[13]Ojala T, Pietikainen M, and Harwood D. A comparative study of texture measures with classification based on feature distribution[J].Pattern Recognition.1996, 29(1):51-59[14]Laurenz Wiskott, Jean-Marc Fellous, Norbert Kruger, and Christoph von der Malsburg. Face recognition by elastic bunch graph matching. PAMI, 1997, 19(7): 775-779.[15]Wu Bo, Ai Haizhou, Huang Chang, and Lao Shihong. Fast rotation invariant multi-view face detection based on real adaboost. Automatic face and gesture recognition, Seoul, Korea, 2004: 79-84.[16]Ojala T, Pietikainen M, and Maenpaa M. Multiresolution gray-scale and rotation invariant texture classification with local binary patterns. PAMI, 2002, 24(7): 971-987.

施引文献

资源附件(0)

访问统计