Course - Biometrics - IMT6071
Biometrics
Assessments and mandatory activities may be changed until September 20th.
About
About the course
Course content
- Fingerprint recognition
- Vein recognition
- Face recognition including three dimensional data
- Iris recognition
- Multimodal biometrics
- Attack mechanisms against biometric system (components)
- Privacy Enhancing Technologies including homomorphic encryption
- Sample quality assessment technologies
Students will learn the capabilities of GPU-consuming Deep Learning approaches to solve biometric tasks. They will also be motivated to investigate, whether low-resource consuming hand-crafted algorithms can solve a task equally well in a sustainable spirit.
Learning outcome
Knowledge:
- The candidate possesses knowledge at the most advanced frontier in the field of biometrics.
- The candidate has mastered academic theory and scientific methods in biometrics.
- The candidate is capable of considering suitability and use of different methods and processes in research in the field of biometrics.
- The candidate is capable of contributing to development of new knowledge, theories, methods, interpretations and forms of documentation in biometrics.
Skills:
- The candidate is capable of formulating problems, planning and completing research projects in biometrics.
- The candidate is capable of doing research and development at a high international level. The candidate is capable of handling complex academic tasks.
- The candidate can challenge established knowledge and practice in biometrics.
More specifically after the course, the candidate should have the following capabilities:
- Developed a systematic understanding of biometric systems and their capabilities.
- Mastered multiple modality-specific feature extraction and have the ability to evaluate their suitability for given acquisition characteristics
- Developed in-depth insights into statistical methods and tools for biometrics and their performance evaluation
- The ability to synthesize multi-modal analysis methods and solve score normalisation problems in fusion systems
- The ability to appraise and differentiate threats to biometric reference data, judging and realizing adequate protection mechanisms accordingly
- The ability to perform in-depth assessment of biometric component placement within a security system demonstrated
- The ability to design and defend a biometric security system when provided with a threat scenario
General competence:
- The candidate is capable of identifying relevant - and possibly new - ethical problems and exercising research in biometrics with academic integrity.
- The candidate is capable of managing complex interdisciplinary tasks and projects.
- The candidate is capable of disseminating the results of research and development in biometrics through approved national and international publication channels.
- The candidate is capable of taking part in debates in international forums within the field of biometrics.
- The candidate is capable of considering the need for, taking initiative to and engaging in innovation in the field of biometrics.
More specifically the candidate:
- will have the competence to demonstrate the ability to design a biometric system suitable for a given scenario
- judge the relevance of ethical and privacy issues investigate for a given scenario
- technical solutions and evaluate them in a critical analysis.
- synthesize new ideas during evaluation phase
- communicate with peers in the biometric community in terms of reviewing research topics
- manage team work
Learning methods and activities
- Lectures
- Assignments
- Seminar(s)
Additional information: Seminar with term paper presentation
Compulsory requirements: None
About the course in general: In case of doubt the English version is the main description
Further on evaluation
Re-sit: The whole course must be repeated.
Assessment forms: Candidates must provide a research report (term paper) on a topic that is chosen by the candidate in coordination with the lecturer. The term paper should not focus on a survey of methods but rather address original research and be submitted to a scientific conference (e.g. NISK, BIOSIG)
Specific conditions
Admission to a programme of study is required:
Information Security and Communication Technology (PHISCT)
Course materials
[1] A. Jain, P. Flynn, A. Russ: Handbook of Biometrics, Springer, 2008
[2] S. Li and A. Jain: Handbook of Face Recognition, Springer, 2011
[3] D. Maltoni , D. Maio, A. Jain, S Prabhakar: Handbook of Fingerprint Recognition, Springer, 2009
[4] S. Marcel, M. Nixon, J. Fierrez, N. Evans: Handbook of Biometric Anti-Spoofing - Presentation Attack Detection, Springer, 2019
[5] P. Tuyls, B. Skroic, T. Kevenaar: Security with Noisy Data, Springer, 2007
[6] A. Uhl, C. Busch, S. Marcel, R. Veldhuis: Handbook of Vascular Biometrics, Springer, 2020
[7] C. Rathgeb, R. Tolosana, R. Vera-Rodriguez, C. Busch: Handbook of Digital Face Manipulation and Detection, Springer, 2021
Subject areas
- Informatics
Contact information
Course coordinator
Department with academic responsibility
Department of Information Security and Communication Technology