Machine generated contents note: ch. 1 Introduction to Behavioral Biometrics --
1.1. Behaviometrics --
1.1.1. How It Works --
1.1.2. Major Benefits --
1.2. What Is Special about Behavioral Biometrics Data Acquisition? --
1.3. Behavioral Biometrics Features --
1.4. Classification of Behavioral Biometrics Traits --
1.5. Properties of Few Behavioral Biometrics --
1.5.1. Signature --
1.5.1.1. Constraints of Signature Recognition --
1.5.1.2. Merits of Signature Recognition --
1.5.1.3. Demerits of Signature Recognition --
1.5.1.4. Applications of Signature Recognition --
1.5.2. Keystroke Dynamics --
1.5.2.1. Merits of Keystroke Recognition --
1.5.2.2. Demerits of Keystroke Recognition --
1.5.2.3. Application of Keystroke Recognition --
1.5.3. Gait --
1.5.3.1. Merits of Gait Recognition --
1.5.3.2. Demerits of Gait Recognition --
1.5.3.3. Application of Gait Recognition --
1.5.4. Voice --
1.5.4.1. Differences between Voice and Speech Recognition --
1.5.4.2. Merits of Voice Recognition --
1.5.4.3. Demerits of Voice Recognition --
1.5.4.4. Applications of Voice Recognition --
1.6. Behavioral Biometrics Data Acquisition Device --
1.7. Behavioral Biometrics Recognition Systems --
1.7.1. Accomplishment of Behavioral Biometrics Systems --
1.7.2. Initial Processing and Analysis of Biometric Traits --
1.7.3. Framework --
1.8. Generalized Algorithm --
1.9. Performance Measurement --
1.9.1. Benchmark Definition --
1.9.2. Robustness Analysis --
1.9.3. Discussion --
1.10. Evaluation of Behavioral Biometric Systems --
1.10.1. Discussion --
1.11. Comparison and Analysis --
1.12. Human Measurement and Evaluation on the Basis of Behavioral Biometric Features --
1.12.1. Verification and Identification --
1.12.2. Error Sources in Behavioral Biometrics --
1.13. Types of Basic Verification Errors and Their Rates --
1.13.1. Error Graphical Representation --
1.13.2. Further Study of Errors --
1.14. Open Issues --
1.14.1. Collection of Sensitive Information --
1.14.2. Negative Reaction to Obtrusive Equipment --
1.14.3. Consent and Secondary Use for Data Collected with Unobtrusive Equipment --
1.14.4. Sensitivity to Change of Application Configuration --
1.14.5. Spoofing Attacks --
1.15. Future Trends --
1.16. Application Area --
1.17. Behavioral Biometrics Used in Real-Time Application --
1.18. Conclusions --
References --
ch. 2 Signature Recognition --
2.1. Brief History of Handwriting Analysis --
2.2. Automated Systems for Signature Recognition --
2.3. Offline and Online Signatures --
2.4. Types of Forgeries --
2.5. Databases for Signature System Evaluation --
2.5.1. SVC2004 --
2.5.2. GPDS-960 --
2.5.3. MCYT-100 --
2.5.4. BIOMET --
2.6. Commercial Software --
2.6.1. SOFTPRO --
2.6.2. ParaScript --
2.6.3. SQN Banking Systems --
2.7. A Review to Signature Recognizers --
2.7.1. Data Acquisition --
2.7.2. Preprocessing --
2.7.3. Feature Extraction --
2.7.3.1. Graphology Based --
2.7.3.2. Shape Context Based --
2.7.3.3. Contour Based --
2.7.3.4. Projection Based --
2.7.3.5. Curvature Based --
2.7.3.6. Radon Transform Based --
2.7.3.7. Hough Transform Based --
2.7.3.8. Texture Based --
2.7.3.9. Wavelet Transform Based --
2.7.4. Classification --
2.7.4.1. Template Matching --
2.7.4.2. Statistical Classification --
2.8. Assessment of Biometric Signature Systems --
2.9. Example Studies on Signature Recognition --
2.9.1. Online System --
2.9.1.1. Results --
2.9.1.2. Identification --
2.9.1.3. Verification --
2.9.1.4. Discussion --
2.9.2. Offline System --
2.9.2.1. Results --
2.9.2.2. Identification --
2.9.2.3. Verification --
2.9.2.4. Discussion --
References --
ch. 3 Keystroke Dynamics --
3.1. History of Keystroke Dynamics --
3.2. Keystroke Analysis --
3.2.1. Data Acquisition --
3.3. Variability of Users, User Behavior, and Hardware --
3.4. Authentication and Identification --
3.4.1. On Biometrics Context of Keystroke Dynamics --
3.5. Characteristics of Keystroke Dynamics --
3.5.1. Universality --
3.5.2. Uniqueness --
3.5.3. Permanence --
3.5.4. Collectability --
3.5.5. Performance --
3.5.6. Acceptability --
3.5.7. Circumvention --
3.5.8. Summary --
3.6. Approaches to Keystroke Dynamics --
3.6.1. Taxonomies of Approaches --
3.6.2. Input Text Approach Taxonomy --
3.6.3. Simple Typing Features --
3.7. Advanced Approaches --
3.8. Fixed Text for All Users --
3.8.1. Dataset --
3.8.2. Proposed Algorithm --
3.9. Fixed Text for Each User (BioPassword/AdmitOneSecurity) --
3.9.1. Computer-Access Security Systems Using Keystroke Dynamics --
3.9.2. AdmitOneSecurity --
3.10. Non-Fixed Text with Regard to Key --
3.10.1. Proposed Algorithm --
3.10.2. Experimental Results and Discussion --
3.11. Non-Fixed Text with No Regard to Key --
3.11.1. Dataset --
3.11.2. Proposed Algorithm --
3.12. Continuous Authentication --
3.13. Perspectives --
3.14. Modern Trends and Commercial Applications for Keystroke Dynamics --
3.14.1. Errors Made by Users and Their Correction Methods --
3.14.2. Pressure-Sensitive Keyboards --
3.14.3. Mobile Phone Keyboards --
3.14.4. ATM Hardware --
3.14.5. Random Numbers Generation --
3.14.6. Timing Attacks on Secure Communications --
3.14.7. Examples of Commercial Applications --
3.15. Legal Issues --
3.16. Conclusions --
References --
ch. 4 Gait Analysis --
4.1. Human Gait Recognition --
4.2. Features of Gait Analysis --
4.3. Applications of Gait Analysis --
4.4. Gait Cycle --
4.5. Describing a Stance --
4.6. Why Does Gait Change from Person to Person or from Time to Time? --
4.7. A Brief Review of the Literature on Human Gait Recognition --
4.8. Research Challenges --
4.8.1. External Factors --
4.8.2. Internal Factors --
4.9. Gait Databases for Research --
4.9.1. CASIA-A --
4.9.2. CASIA-B --
4.9.3. CMU MoBo --
4.9.4. USF Dataset --
4.9.5. Southampton Dataset --
4.9.6. 3D Dataset --
4.9.7. UMD Dataset --
4.9.8. TUM-IITKGP Dataset --
4.9.9. OU-ISIR Database --
4.10. Gait Recognition Using Partial Silhouette-Based Approach --
4.10.1. Motivation of the Partial Silhouette-Based Approach --
4.10.2. Dynamic Features of Gait --
Why Partial Silhouette? --
4.10.3. Partial Silhouette-Based Methodology --
4.10.4. Preprocessing for Removing Noise --
4.10.5. Gait Cycle Detection and Extraction of Landmark Frames --
4.11. Extraction of Partial Silhouette --
4.11.1. Bounding Box --
4.11.2. Image Segmentation --
4.11.3. Feature Extraction --
4.11.4. Classification --
4.11.5. Training --
4.11.6. Testing --
4.12. Experimental Verification --
4.12.1. Results of Full versus Partial Silhouettes --
4.13. Comparison with Other Methods --
4.14. Effectiveness of Partial Silhouette Method in the Presence of Noise --
4.15. Time Complexity of the Partial Silhouette-Based Method --
4.16. Conclusions --
References --
ch. 5 Voice Recognition --
5.1. Voice Recognition --
5.1.1. Advantages of Voice Recognition over Other Biometric Traits --
5.1.2. Main Steps in Voice Recognition Systems --
5.2. Signal Acquisition and Preprocessing --
5.2.1. Biological Background --
5.2.2. Preprocessing Stage --
5.2.3. Feature Extraction --
5.3. Toeplitz Matrix Minimal Eigenvalues Algorithm --
A Survey --
5.3.1. Linear Predictive Coding and Burg's Model --
5.3.2. Mel Frequency Cepstral Coefficients --
5.4. Classification Using NNs --
5.4.1. Probabilistic NNs --
5.4.2. Radial Basis Function NNs --
5.5. Achievements in Similar Works --
5.6. Achievements in Voice Recognition --
5.6.1. The Simplest Case, Uttered Words Recognition --
5.6.1.1. Input Samples and Preprocessing Stage --
5.6.1.2. Experiments and Result --
5.6.2. Voiceprint and Security Systems --
5.6.2.1. Performance of the Speaker Identification Security System --
5.6.2.2. Multilevel Security for the Spoken Words and Speaker --
5.6.3. Text-Independent Speaker Identification --
5.6.3.1. Database and Preprocessing --
5.6.3.2. First Attempt --
5.6.3.3. Another Attempt --
5.6.4. What about Speaker Verification? --
5.6.4.1. Identification Treatment --
5.6.4.2. Verify the Speaker --
Claiming It Correctly --
5.6.4.3. True Rejection and False Acceptance --
5.6.4.4. Extra Testing Data for Verification --
5.7. Conclusions --
References.
