Seminar on Biometric Systems

Higher education teachers: Dobrišek Simon

Subject description

Prerequisites:

• Completed undergraduate study programme in the field of electrical engineering or related engineering or natural and mathematical sciences.
• Enrolment in the 2nd year of the Master’s study programme for Electrical Engineering (2nd cycle).
• The basics of pattern recognition, machine learning, probability theory, and computer programming.

Content (Syllabus outline):

Introduction to Biometric Systems: identifiable biometric characteristics (physiological, behavioural), system components and phases of system operation (enrolment, verification, identification).
Acquisition of Physiological (face, fingerprint, iris, hand palms and geometry) and Behavioural (voice, mimic, handwriting, and gait) Characteristics: contact and noncontact measurement, frequently used sensors. Testing the quality and genuineness of acquired data.
Design of Uni-modal and Multi-modal Biometric Systems: sources of biometric information, levels and methods of biometric information fusion. Comparison of uni- and multi-modal systems.
Evaluation of Biometric Systems: average enrolment and recognition time, biometric system errors (matching and decision errors), enrolment error, data acquisition error.
Testing of Biometric Systems: test plan, person group, testing enrolment, verification and identification processes. Forgery tests. Databases for automated and repeatable tests.
Biometric Standards and Privacy Issues. Ethical and Cultural Issues associated with biometric system applications.
Seminars: development of uni- and multi-modal biometric systems: biometric systems in security (identification and travel documents, e-commerce, e-security systems) and others (smart rooms and environments, user-adapted content search) applications.

Objectives and competences:

• To acquaint students with the principles and basic components of biometric systems.
• To present and elaborate examples of biometric systems for the automated recognition of people.
• To expand knowledge from the field of pattern recognition.

Intended learning outcomes:

Knowledge and understanding:

• After completing this course, the student will be able to demonstrate a knowledge and understanding of the:
• principles of the construction of biometric systems for the automated recognition of people,
• problems of quality assurance in biometric systems and the protection of biometric data,
• ethical and cultural issues associated with the use of biometric systems.

The use of knowledge:

• The student will be able to use the acquired knowledge to develop and construct biometric systems for the automated recognition of people (identification and travel documents, e-commerce, e-security systems, smart rooms and environments, user-adapted multi-media content search, identification of the writers of historical documents, criminal investigations support).

Transferable skills:

• the use of literature and other resources in the field of biometric systems;
• the use of open source development tools, data sets and programming environments: the students carry out the projects in one of the programming languages ​​C/C++, Python, C#, Java, or using MATLAB, use one of the biometric databases (NIST SRE, XM2VTS, FRGC, Banca, LFW, PolyU, etc.), and use tools like OpenCV, ORANGE and WEKA;
• communication skills: oral presentation of seminar projects, preparing seminar project reports;
• problem solving: problem analysis, algorithm design, implementation and testing of a program;
• group work: the organization and management of groups, active participation in groups.

Learning and teaching methods:

• lectures,
• seminars,
• seminar projects.

Study materials

1. N. Pavešić: Razpoznavanje vzorcev (3. izdaja), Založba FE in FRI, 2012.
2. K. Jain, A. A. Ross, K. Nandakumar, Introduction to Biometrics, Springer, 2011.
3. R. M. Bolle et al.: Guide to Biometrics, Springer, 2004.
4. D. Maltoni et al.: Handbook of Fingerprint Recognition, Springer, 2003.