Dissertation > Excellent graduate degree dissertation topics show

Algorithms Research on Feature Extraction and Classifiers of High-Dimensional and Small Sample Size Data

Author: WangWeiDong
Tutor: YangJingYu
School: Nanjing University of Technology and Engineering
Course: Pattern Recognition and Intelligent Systems
Keywords: pattern recognition small sample size problem feature extraction classifiers virtual samples machine learning dual-space algorithms perturbation analysis face recognition
CLC: TP391.41
Type: PhD thesis
Year: 2007
Downloads: 860
Quote: 1
Read: Download Dissertation

Abstract


The "small sample size" (SSS) problem arises from the small number of availabletraining samples compared to the dimensionality of the sample space. Therefore, a criticalissue of applying linear discriminant analysis (LDA) and quadratic discriminantanalysis(QDA) is both the singularity and instability of the covarianee matrix. This paperprovided some comprehensive solutions to the problem.First, this paper proposed two methods to produce virtual samples. One of them chose agroup of normal orthogonal vectors to produce virtual samples in a subspaee, which were usedto optimize covariance matrices of per class. The other method got many virtual samples byperturbing training samples. These virtual samples made the covariance matrices of per classbecome non-singularity, so the classifier of quadratic discriminant analysis could be useddirectly. Second, the intelligence feature extraction and classifiers algorithms and a dual-spacealgorithm were put forward to solve the problem of the singularity and instability. This paper,applying the machine-learning theory to design intelligence the feature extraction andclassifiers algorithms, proposed the intelligence feature extraction and classifiers algorithmswith learning function. They could automatically learn from testing samples and renew theirknowledge. Also, the paper put forward a dual-space algorithm, which projected the testingsamples difficult to recognize in one single subspace into another single subspace andclassified. Then, the recognition results of two subspaces were tested together. Moreover, thispaper presented a novel classifier based on orthogonal projection for small sample sizeproblem. The character of the classifier is that it need not calculate the covariance matrices’inverses and the recognition accuracy is better than RDA and nearest neighbor classifier.Based on perturbation features of eigenvalue and eigenvector, the paper pointed outeigenvector of morbid eigenvalue may be perturbed to a great degree. The proposed algorithmcan simplify projection matrix, improve the efficiency of features extraction and then makethe recognition ratio robust.

Related Dissertations

  1. Research on Algorithms of 2D Face Template Protection,TP391.41
  2. The Design of Identification System for Power Marketing Services,TP391.41
  3. Under the conditions of variable lighting and occlusion of face recognition technology and its application,TP391.41
  4. Research on Application of Fractional Differential to Digital Image Processing and Support Vector Machine to Face Recognition,TP391.41
  5. Social network multimedia data mining,TP311.13
  6. Research on Feature Extraction and Classification Algorithm of Face Image,TP391.41
  7. Improvement of Manifold Learning for Face Recognition,TP391.41
  8. Research on Kernel-based Discriminant Analysis Algorithm,TP391.41
  9. Based on face and fingerprint multi- modal identification system research,TP391.41
  10. Face Recognition Based on Gabor Wavelettransform,TP391.41
  11. Research on the Face Recognitionalgorithm Based on Weighted Maximum Margin Criterion,TP391.41
  12. 3D Face Recognition Using Ifnormation Fusion,TP391.41
  13. Research on Technology of Face Recognition Based on Grey Relational Analysis,TP391.41
  14. Research and Implementation of Face Recognition Algorithm Based on ASM,TP391.41
  15. Research on Multi-View Face Recognition Based on Statistical Models,TP391.41
  16. Implementations of Face Recognition System Based on ADSP-BF561,TP391.41
  17. Research on the Methods of Face Recognition for Mobile Phone,TP391.41
  18. Research about Access Control System Based on Face Recognition,TP391.41
  19. Research of Face Recognition Based on Video Surveillance,TP391.41
  20. Research on Face Detection and Recognition Based on Video,TP391.41
  21. Face Recognition Based on Bayesian Decision Fusion of the Algorithm,TP391.41

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile