Face Detection and Recognition

Book description

Face detection and recognition are the nonintrusive biometrics of choice in many security applications. Examples of their use include border control, driver's license issuance, law enforcement investigations, and physical access control.Face Detection and Recognition: Theory and Practice elaborates on and explains the theory and practice of face de

Table of contents

  1. Cover
  2. Half Title
  3. Title Page
  4. Copyright Page
  5. Dedication
  6. Contents (1/2)
  7. Contents (2/2)
  8. List of Figures (1/2)
  9. List of Figures (2/2)
  10. List of Tables
  11. Preface
  12. Acknowledgment
  13. 1. Introduction
    1. 1.1. Introduction
    2. 1.2. Biometric identity authentication techniques
    3. 1.3. Face as biometric identity
      1. 1.3.1. Automated face recognition system
      2. 1.3.2. Process ow in face recognition system
      3. 1.3.3. Problems of face detection and recognition
      4. 1.3.4. Liveness detection for face recognition
    4. 1.4. Tests and metrics
    5. 1.5. Cognitive psychology in face recognition
  14. 2. Face detection and recognition techniques
    1. 2.1. Introduction to face detection
    2. 2.2. Feature-based approaches for face detection
      1. 2.2.1. Low-level analysis
        1. 2.2.1.1. Edges
        2. 2.2.1.2. Gray-level analysis
        3. 2.2.1.3. Color information in face detection
        4. 2.2.1.4. Motion-based analysis
      2. 2.2.2. Active shape model
      3. 2.2.3. Feature analysis
      4. 2.2.4. Image-based approaches for face detection
      5. 2.2.5. Statistical approaches
    3. 2.3. Face recognition methods
      1. 2.3.1. Geometric feature-based method
      2. 2.3.2. Subspace-based face recognition
      3. 2.3.3. Neural network-based face recognition
      4. 2.3.4. Correlation-based method
      5. 2.3.5. Matching pursuit-based methods
      6. 2.3.6. Support vector machine approach
      7. 2.3.7. Selected works on face classi ers
    4. 2.4. Face reconstruction techniques
      1. 2.4.1. Three-dimensional face recognition (1/2)
      2. 2.4.1. Three-dimensional face recognition (2/2)
        1. 2.4.1.1. Feature extraction
        2. 2.4.1.2. Global feature extraction
        3. 2.4.1.3. Three-dimensional morphable model
  15. 3. Subspace-based face recognition
    1. 3.1. Introduction
    2. 3.2. Principal component analysis
      1. 3.2.1. Two-dimensional principal component analysis
      2. 3.2.2. Kernel principal component analysis
    3. 3.3. Fisher linear discriminant analysis
      1. 3.3.1. Fisher linear discriminant analysis for two-class case
    4. 3.4. Independent component analysis
  16. 4. Face detection by Bayesian approach
    1. 4.1. Introduction
    2. 4.2. Bayes decision rule for classification
      1. 4.2.1. Gaussian distribution
      2. 4.2.2. Bayes theorem
      3. 4.2.3. Bayesian decision boundaries and discriminant function
      4. 4.2.4. Density estimation using eigenspace decomposition
    3. 4.3. Bayesian discriminant feature method
      1. 4.3.1. Modelling of face and non-face pattern
      2. 4.3.2. Bayes classi cation using BDF
    4. 4.4. Experiments and results
  17. 5. Face detection in color and infrared images
    1. 5.1. Introduction
    2. 5.2. Face detection in color images
    3. 5.3. Color spaces
      1. 5.3.1. RGB model
      2. 5.3.2. HSI color model
      3. 5.3.3. YCbCr color space
    4. 5.4. Face detection from skin regions
      1. 5.4.1. Skin modelling
        1. 5.4.1.1. Skin color modelling explicitly from RGB space
        2. 5.4.1.2. Skin color modelling explicitly from YCbCr space
    5. 5.5. Probabilistic skin detection
    6. 5.6. Face detection by localizing facial features
      1. 5.6.1. EyeMap
      2. 5.6.2. MouthMap
    7. 5.7. Face detection in infrared images
    8. 5.8. Multivariate histogram-based image segmentation
      1. 5.8.1. Method for finding major clusters from a multivariate histogram
      2. 5.8.2.Experiments and results on the color and IR face image datasets
      3. 5.8.3. Utility of facial features
  18. 6. Intelligent face detection
    1. 6.1. Introduction
    2. 6.2. Multilayer perceptron model
      1. 6.2.1. Learning algorithm
    3. 6.3. Face detection networks
    4. 6.4. Training images
      1. 6.4.1. Data preparation
      2. 6.4.2. Face training
        1. 6.4.2.1. Active learning
      3. 6.4.3. Exhaustive training
    5. 6.5. Evaluation of face detection for upright faces
      1. 6.5.1. Algorithm
      2. 6.5.2. Image scanning and face detection
  19. 7. Real-time face detection
    1. 7.1. Introduction
    2. 7.2. Features
    3. 7.3. Integral Image
      1. 7.3.1. Rectangular feature calculation from integral image
    4. 7.4. AdaBoost
      1. 7.4.1. Modifed AdaBoost algorithm
      2. 7.4.2. Cascade classifier
    5. 7.5. Face detection using OpenCV
  20. 8. Face space boundary selection for face detection and recognition
    1. 8.1. Introduction
    2. 8.2. Face points, face classes and face space boundaries
    3. 8.3. Mathematical preliminaries for set estimation method
    4. 8.4. Face space boundary selection using set estimation
      1. 8.4.1. Algorithm for global threshold-based face detection
    5. 8.5. Experimental design and result analysis
      1. 8.5.1. Face/non-face classification using global threshold during face detection
      2. 8.5.2. Comparison between threshold selections by ROC based and set estimation-based techniques
        1. 8.5.2.1. Formation of training-validation-test set
    6. 8.6. Classification of face/non-face regions
    7. 8.7. Class specific thresholds of face-class boundaries for face recognition
    8. 8.8. Experimental design and result analysis
      1. 8.8.1. Description of face dataset
        1. 8.8.1.1. Recognition rates
      2. 8.8.2. Open test results considering imposters in the system
      3. 8.8.3. Recognition rates considering only clients in the system (1/2)
      4. 8.8.3. Recognition rates considering only clients in the system (2/2)
  21. 9. Evolutionary design for face recognition
    1. 9.1. Introduction
    2. 9.2. Genetic algorithms
      1. 9.2.1. Implementation
      2. 9.2.2. Algorithm
    3. 9.3. Representation and discrimination
      1. 9.3.1. Whitening and rotation transformation
      2. 9.3.2. Chromosome representation and genetic operators
      3. 9.3.3. The tness function
      4. 9.3.4. The evolutionary pursuit algorithm for face
  22. 10. Frequency domain correlation filters in face recognition
    1. 10.1. Introduction
      1. 10.1.1. PSR calculation
    2. 10.2. A brief review on correlation filters
    3. 10.3. Mathematical background of correlation filter
      1. 10.3.1. ECPSDF filter design
      2. 10.3.2. MACE filter design
        1. 10.3.2.1. Constrained optimization with Lagrange multipliers
      3. 10.3.3. MVSDF filter design
      4. 10.3.4. Optimal trade-o (OTF) filter design
      5. 10.3.5. Unconstrained correlation filter design
        1. 10.3.5.1. MACH filter design
        2. 10.3.5.2. UMACE filter design
        3. 10.3.5.3. OTMACH filter design
    4. 10.4. Physical requirements in designing correlation filters
    5. 10.5. Applications of correlation filters
    6. 10.6. Performance analysis
      1. 10.6.1. Performance evaluation using PSR values
      2. 10.6.2. Performance evaluation in terms of %RR and %FAR (1/2)
      3. 10.6.2. Performance evaluation in terms of %RR and %FAR (2/2)
      4. 10.6.3. Performance evaluation by receiver operating characteristics (ROC) curves
    7. 10.7. Video correlation lter
    8. 10.8. Formulation of unconstrained video filter
      1. 10.8.1. Mathematical formulation of MUOTSDF
      2. 10.8.2. Unconstrained video filter
    9. 10.9. Distance classifier correlation filter
    10. 10.10. Application of UVF for face detection
      1. 10.10.1. Training approach
      2. 10.10.2. Testing approach
      3. 10.10.3. Face detection in video using UVF
        1. 10.10.3.1. Modification in training approach
      4. 10.10.4. Validation of face detection
      5. 10.10.5. Face classification using DCCF
  23. 11. Subspace-based face recognition in frequency domain
    1. 11.1. Introduction
    2. 11.2. Subspace-based correlation filter
    3. 11.3. Mathematical modelling with 1D subspace
      1. 11.3.1. Reconstructed correlation filter using 1D subspace
      2. 11.3.2. Optimum projecting image correlation filter using 1D
    4. 11.4. Face classification and recognition analysis in
    5. 11.5. Test results with 1D subspace analysis
      1. 11.5.1. Comparative study in terms of PSRs
      2. 11.5.2. Comparative study on %RR and %FAR
    6. 11.6. Mathematical modelling with 2D subspace
      1. 11.6.1. Reconstructed correlation lter using 2D subspace
    7. 11.7. Test results on 2D subspace analysis
      1. 11.7.1. PSR value distribution for authentic and impostor classes
      2. 11.7.2. Comparative performance in terms of %RR
      3. 11.7.3. Performance evaluation using ROC analysis
    8. 11.8. Class-specific nonlinear correlation filter
    9. 11.9. Formulation of nonlinear correlation filters
      1. 11.9.1. Nonlinear optimum projecting image correlation filter
      2. 11.9.2. Nonlinear optimum reconstructed image correlation filter
    10. 11.10. 11.10Face recognition analysis using correlation classifiers
    11. 11.11. Test results
      1. 11.11.1. Comparative study on discriminating performances
      2. 11.11.2. Comparative performance based on PSR distribution
      3. 11.11.3. Performance analysis using ROC
      4. 11.11.4. Noise sensitivity
  24. 12. Landmark localization for face recognition
    1. 12.1. Introduction
    2. 12.2. Elastic bunch graph matching
    3. 12.3. Gabor wavelets
    4. 12.4. Gabor jets
    5. 12.5. The elastic bunch graph matching algorithm
    6. 12.6. Application to face recognition
    7. 12.7. Facial landmark detection
      1. 12.7.1. ASEF correlation filter
      2. 12.7.2. Formulation of ASEF
    8. 12.8. Eye detection
    9. 12.9. Multicorrelation approach
      1. 12.9.1. Design of landmark filter(LF)
      2. 12.9.2. Landmark localization with localization filter
    10. 12.10. Test results
  25. 13. Two-dimensional synthetic face generation using set estimation
    1. 13.1. Introduction
    2. 13.2. Generating face points from intraclass face images
      1. 13.2.1. Face generation using algorithm with intraclass features and related peak signal to noise ratio
    3. 13.3. Generating face points from interclass face images
      1. 13.3.1. Face generation with interclass features
      2. 13.3.2. Rejection of the non-meaningful face and corresponding PSNR test
    4. 13.4. Generalization capability of set estimation method
    5. 13.5. Test of signi cance
  26. 14. Datasets of face images for face recognition systems
    1. 14.1. Face datasets
      1. 14.1.1. ORL dataset
      2. 14.1.2. OULU physics dataset
      3. 14.1.3. XM2VTS dataset
      4. 14.1.4. Yale dataset
      5. 14.1.5. Yale-B dataset
      6. 14.1.6. MIT dataset
      7. 14.1.7. PIE dataset
      8. 14.1.8. UMIST dataset
      9. 14.1.9. PURDU AR dataset
      10. 14.1.10. FERET dataset
      11. 14.1.11. Performance evaluation of face recognition
    2. 14.2. FERET and XM2VTS protocols
    3. 14.3. Face recognition grand challenge (FRGC)
    4. 14.4. Face recognition vendor test (FRVT)
    5. 14.5. Multiple biometric grand challenge
    6. 14.6. Focus of evaluation
  27. Conclusion
  28. Bibliography (1/4)
  29. Bibliography (2/4)
  30. Bibliography (3/4)
  31. Bibliography (4/4)
  32. Index

Product information

  • Title: Face Detection and Recognition
  • Author(s): Asit Kumar Datta, Madhura Datta, Pradipta Kumar Banerjee
  • Release date: October 2015
  • Publisher(s): Chapman and Hall/CRC
  • ISBN: 9781482226577