Plenary Lecture
Current Biometric Trends - Recognition of Vein Patterns and Soft Biometry
Professor Ryszard S. Choraś
Head of the Department of Applied informatics and System Engineering
Faculty of Telecommunications, Computer Science and Electrical Engineering
University of Technology and Life Sciences
Poland
E-mail: choras@utp.edu.pl
Abstract: The core of all biometrics systems have five key modules: sensors (image/data acquisition), feature extractor, biometric database, matcher and decision-maker. The sensor reads the biometric information from the user. Feature extractor module extracts features from the biometric data. Matcher module indicates the similarity between extracted features from user sample and a enrolled template. Biometric database maintains the templates of the enrolled users. Decision-maker interprets the result. Vein pattern is unique for each human being even in the case of identical twins. Moreover, it is a highly stable pattern over time. Vein recognition is a method of biometric, that uses pattern recognition techniques based on images of blood vessel. Blood vessel patterns (identified only on a live body) are unique to each individual. Vein recognition is a strong immunity to forgery. Existing vein recognition methods can be roughly divided into two categories. The first category are methods that determine feature points within the images and match through the spatial relationships among feature points. The other category of methods combine the global features of vein images. The vein pattern is distinctive for various individuals. The current available approaches for vein recognition are all based on texture extraction based on visible and/or infrared image of pattern vein. In this lecture we considered various image vein templates and various vein recognition methods. Soft biometric features are those that provide certain information about a person, but do not provide 100% certainty about the person's identification and/or are not sufficient to distinguish between two persons. A system that is completely based on soft biometric traits cannot provide the required accuracy in the recognition of individuals. However, soft biometric traits can be used to improve the performance of a traditional biometric system in many ways. In order to utilize soft biometrics, there must be a mechanism to automatically extract features from the user during the recognition phase. The biometrics system should be able to automatically measure the soft biometric characteristics like e.g. gender, and ethnicity without any interaction with the user. The biometric recognition system is divided into two subsystems. One subsystem is based on traditional biometric identifiers, the secondary biometric system, is based on soft biometric traits like gender and/or ethnicity. Main Biometric Modalities for Gender Estimation are Face, Iris, Fingerprint, Hand geometry, Voice, Handwriting, Ear, Gait.
The most informative features are:
- Geometric features,
- Texture features,
- Statistical features.
For ethnicity classification, the most informative features are around the eyes, nose, and lip. In this plenary lecture some aspects biometrics vein recognition and soft biometrics (especially gender and ethnicity applications) and features extracted from their modalities will be presented.
Brief Biography of the Speaker: Prof. Ryszard S. Choraś is currently Full Professor in the Institute of Telecommunications and Computer Sciences of the University of Technology & Life Sciences, Bydgoszcz, Poland. His research experience covers image processing and analysis, image coding, feature extraction and computer vision. At present, he is working in the field of image retrieval and indexing, mainly in low- and high-level features extraction and knowledge extraction in CBIR systems. He is the author of Computer Vision. Methods of Image Interpretation and Identification (2005) and more than 203 articles in journals and conference proceedings. He is the member of the Polish Cybernetical Society, Polish Neural Networks Society, IASTED, and the Polish Image Processing Association. Professor Choras is a member of the editorial boards of Machine Vision and Graphics, International Journal of Biometrics (IJBM), International Journal of Biology and Biomedical Engineering, Recent Patents On Signal Processing (Bentham Open). He is the editor-in-chief of WSEAS Transaction on Signal Processing Journal, Image Processing and Communications, An International Journal and Associate editor-in-chief Computer Science Journals (CSC Journals) Image Processing (IJIP). He is also the Chairman of the Image Processing and Communications Conference (2009, 2010, 2011, 2012, 2013, 2014, 2015, 2016, 2017) and editor books Image Processing and Communications Challenges published in Advances in Intelligent Systems and Computing Springer Verlag Series. He has served on numerous conference committees, e.g., as Visualization, Imaging, and Image Processing (VIIP) , IASTED International Conference on Signal Processing, Pattern Recognition and Applications (SPPRA) and International Conference on Computer Vision and Graphics in Warsaw, ICINCO\ICATE Conference.