CIBIM 2011 - CIBIM 2011 2011 IEEE Workshop on Computational Intelligence in Biometrics and Identity Management

CIBIM 2011
2011 IEEE Workshop on Computational Intelligence in Biometrics and Identity Management
Biometric technology is the technology of the 21st century which uses measurable physiological or behavioural characteristics to reliably distinguish one person from another. The technology is fast gaining popularity as means of personal identification and verification for different commercial, government and law enforcement applications. Since biometric information cannot be captured in precisely the same way twice, biometric matching is always a “fuzzy comparison”. This feature makes computational intelligence (CI), which is primarily based on artificial intelligence, neural networks, fuzzy logic, evolutionary computing, etc., an ideal solution for addressing biometric problems. The main objective of this workshop is to bring together the leading researchers to exchange the latest theoretical and experimental CI solutions in biometrics and identity management. This event will provide an interdisciplinary forum for research scientists, system developers and students from around the world to discuss the latest advances in the field of Computational Intelligence and its application to real world problems in biometrics and identity management. The submission needs to deal with computational intelligence in biometrics.

Topics
Topics of interest include but are certainly not limited to:

CI-based biometric algorithms, techniques and systems
Machine learning, neural-networks and artificial intelligence methods in biometrics and identity management
Biometric solutions for physical and logical securities
Biometric smart ID, RFID ePassport, biometric authentication and identity management
Biometric information privacy and data security
Covert and unconstrained biometrics
Multiple biometrics and multi-modal biometrics information fusion
Biometric anti-spoofing and liveness detection
Mobile biometric devices and embedded biometric systems
Biometric performance, assurance, and interoperability testing
Symposium Co-Chairs
Qinghan Xiao (qinghan.xiao-AT-drdc-rddc.gc.ca), Defence R&D, Canada
David Zhang (csdzhang-AT-comp.polyu.edu.hk), Hong Kong Polytechnic University, China
Fabio Scotti (fscotti-AT-dti.unimi.it), University of Milan, Italy

Program Committee
Hervé Chabanne, Morpho & Télécom ParisTech, France
Ke Chen, University of Manchester, UK
Eliza Du, Indiana University-Purdue University Indianapolis, USA
Jianjiang Feng, Tsinghua University, China
Eric Granger, École de technologie supérieure, Montreal, Canada
Kevin Jia, IGT, USA
Adams Wai-Kin Kong, Nanyang University, Singapore
Wenxin Li, Peking University, China
Hugo Proença, University of Beira Interior, Portugal
Evangelia Micheli-Tzanakou, Rutgers University, USA
Seref Sagıroglu, Gazi University, Ankara, Turkey
Mario Savastano, National Research Council of Italy
Jie Tian, Chinese Academy of Sciences, China
Jeffrey Voas, Science Applications International Corporation, USA
Jia-Ching Wang, National Cheng Kung University, Tainan, Taiwan
Yong Xu, Harbin Institute of Technology, China
Xin Yang, Chinese Academy of Sciences, China
Svetlana N. Yanushkevich, University of Calgary, Alberta, Canada

Special Sessions
#1. Adaptive Classification Systems for Biometric Recognition
The recognition of individuals based on their biometric traits provides a powerful alternative to traditional authentication schemes presently applied in a multitude of security and surveillance systems. However, the performance of state-of-the-art neural and statistical classifiers employed in biometric recognition systems typically decline in practice because they face complex operational environments that change over time, and because they are designed a priori using limited and unbalanced data samples. In fact, biometric systems are typically designed with a limited set of training samples, and with static classification environments in mind. For accurate and timely recognition, biometric systems should allow for efficient adaptation in response to emerging knowledge and data.
In recent years, adaptive classification systems have been proposed to efficiently maintain up-to-date biometric models, and sustain a high level of accuracy in real-world biometric applications. These systems have the ability to evolve their parameters and architecture over time in response to new or changing input features, data samples, classes (i.e., individuals) and/or environments. Moreover, these systems play a central role in self-adapting and human-centric frameworks, where biometric systems are gradually designed and updated as the operational environment unfolds. Significant challenges must be overcome before such techniques can be successfully deployed for real-world biometric applications. The purpose of this session is to provide a scientific forum for researchers, engineers, system designers to present and discuss recent advances in the area of adaptive classification systems for biometric recognition and related technologies.

Topics
Suggested topics include as they apply to biometric recognition, but are not limited to:

Adaptive Pattern Recognition Methods, Systems and Technologies
Intelligent and Evolutionary Systems
Neural and Statistical Classifiers
Multi-Classifier Systems
Incremental Learning of Features, Data Samples and Classes
On-Line, Adaptive and Life-Long Learning
Selection and Fusion in Ensembles of Classifiers
Evolutionary Computation
Feature Extraction and Selection
Adaptation of Biometric Systems in Static and Dynamically-Changing Environments
Ambiguity and Novelty Detection
Methodologies for Evaluation of Adaptive Biometric Systems
Special Session Organizer and Chair
Eric Granger, Université du Québec, Montreal, Canada (eric.granger-AT-etsmtl.ca)

#2 Decision-making Support for Biometric Systems
Decision-making support system (DMSS) has been known as an enabler of improving quality of decision. Biometric decision-making support is a potential application domain of DMSS because of the number of influencing factors and complexity of biometric systems. The aim of this session is to provide a scientific forum for researchers, engineers and computer scientists to discuss and report recent advantages in the area of artificial intelligence techniques for enhancing application of biometrics in civil, law enforcement, biomedical and other applications.

Topics
Original research in the area of biometric systems and applications is solicited, which may include, but is not limited to:

Artificial intelligence methods in biometrics
Agent based authentication systems
Reliability of biometric evidence
Bayesian and Dempster-Shafer decision-making for biometric systems
Fusion levels (rank, decision, sensor, feature and match-score)
Multibiometric system applications
All other aspects of decision-making in biometric application
Special Session Organizers and Chairs
Svetlana N. Yanushkevich, Biometric Technologies Laboratory, University of Calgary, Canada (syanshk-AT-ucalgary.ca)
Vlad Shmerko, Biometric Technologies Laboratory, University of Calgary, Canada