Information Security System

1. Information Security System Biometrics of Next Generation: An overview Nguyễn Văn Đại - 51100706 TrươngDuyLinh - 51101863 Faculty of Computer Science & Engineering HCMC University of Technology

2. Content Faculty of Computer Science & Engineering HCMC University of Technology 2

3. Abstract • The growing demands for stringent security in applications. • Biometric recognition is being increasingly adopted and rapidly growing. • What is biometric technology? (physiological or behavioral characteristics). • Biometric is not new. • Biometric recognition technology faces new challenges. Faculty of Computer Science & Engineering HCMC University of Technology 3

4. Introduction • Biometric technology type: • The choice depends on the nature and requirements of the identification application. Faculty of Computer Science & Engineering HCMC University of Technology 4

5. Introduction Small compare: Fingerprint (50% of market share) has many advantages: • Solutions are mature at an affordable price. • Good performance. • Good ergonomics. • Data are stable. Facial recognition (12% of market share) • Recent technology • Few terminals • Very good ergonomics • Contactless • Medium performance. • Lighting is sensitive • Re enrolment is necessary after some period of time. IRIS (9% of market share): • High cost. • Very low false acceptance rate but a residual false reject rate still high. • Contactless • Ergonomics : Medium • Data are stable. No technology is 100% reliable Faculty of Computer Science & Engineering HCMC University of Technology 5

6. Introduction Voice Biometric Technology: • Identify a person based on their spoken. • Depends on the shape and size of vocal tracts, lips, ... • Used in verification mode. Signature Biometric Technology: • Behavioral biometric modality that is used in daily business transactions. • Some problem: shape, speed, acceleration, … • Very few automatic signature verification systems have been deployed. Faculty of Computer Science & Engineering HCMC University of Technology 6

7. Expectations Performance: • Is of utmost interest in the deployment of biometric systems. • The accuracy of a biometric system is not static, it is data dependent and influenced by several factors. • A biometrics authentication system can make two types of errors: false match & false non-match. • Can not reduce both these errors simultaneously. Faculty of Computer Science & Engineering HCMC University of Technology 7

8. Expectations Cost: • Estimated from its direct and indirect components (hardware components & software modules). • The indirect cost in the usage of biometric system. • return on investment or the cost-benefit analysis Interoperability • The system be interoperable among different biometrics technologies. • The biometric system should be highly interoperable. • Reduce the need for additional software development and bring all the associated advantages (cost savings and efficiency). Faculty of Computer Science & Engineering HCMC University of Technology 8

9. Expectations User Convenience: • A biometrics system should be user friendly. • Health and hygienic can influence user acceptance. Security: • Biometric systems are vulnerable to potential security breaches from spoof and malicious attacks. Faculty of Computer Science & Engineering HCMC University of Technology 9

10. First Generation Biometrics • Include a variety of fingerprint, iris, and face recognition systems • Continuing advances in the sensing technologies, computational speed, operating environment, and storage capabilities. • The major limitations: achievable performance, security, and privacy. Faculty of Computer Science & Engineering HCMC University of Technology 10

11. Second generation biometrics systems • Systems must confront these challenges and develop novel techniques for sensing, signal/image representation, and matching. Faculty of Computer Science & Engineering HCMC University of Technology 11

12. Data Acquisition Environment The performance of matching algorithm critically depends on the quality of biometric data. We now outline two such challenges facing the second generation biometrics system to improve the data acquisition environment using new sensing technologies. Faculty of Computer Science & Engineering HCMC University of Technology 12

13. Biometric System Security • The security ensured by the deployed biometric systems can itself be compromised. A number of studies have analyzed the likelihood of such security breaches and potential approaches to counter these vulnerabilities Image: Typical attack points in a biometric system Faculty of Computer Science & Engineering HCMC University of Technology 13

14. Large­ scale Applications • Biometric systems that can effectively and efficiently operate in ultra large-scale applications, i.e. those capable of supporting hundreds of millions of registered users, have a number of potential opportunities. • High accuracy and throughput under varying operating conditions and user composition. • Sensor interoperability. • Rapid collection of biometric data in harsh operating environments with virtually no failure to enroll rate. • High levels of privacy and template protection. • Secure supporting information/operating systems. Faculty of Computer Science & Engineering HCMC University of Technology 14

15. Soft Biometrics • Metrics are those human characteristics that provide some information about the individual, but lack the distinctiveness and permanence to sufficiently differentiate any two individuals. • Examples of soft biometrics used in law enforcement include scars, marks, tattoos, color of eye, and hair color (a) Image sample for knuckle biometrics (b) conjunctive vasculature (c) tattoo image. Faculty of Computer Science & Engineering HCMC University of Technology 15

16. Application Perspective • Biometric systems are often felt invasive, since the sensors directly interact with the human body to capture person-specific data that is considered privileged. • The stigma of forensic and criminal investigations has been known to influence the user acceptance of the first generation biometric systems. • Biometric traits are part of human body and behavior and therefore, it can threaten the personal privacy of some users. Faculty of Computer Science & Engineering HCMC University of Technology 16

17. Faculty of Computer Science & Engineering HCMC University of Technology 17

18. Faculty of Computer Science & Engineering HCMC University of Technology 18

19. References • Biometrics of Next Generation: An Overview (Anil K.Jain, Ajay Kumar) • Introduction to biometrics – slide 6 (Q.Thomazeau) Faculty of Computer Science & Engineering HCMC University of Technology 19

20. Thanks for your listen!