Josef Bigun – författare

Automatic person authentication, the identification and verification of an individual as such, has increasingly been acknowledged as a significant aspect of various security applications. Various recognition and identification systems have been based on biometrics utilizing biometric features such as fingerprint, face, retina scans, iris patterns, hand geometry, DNA traces, gait, and others.This book originates from an international summer school on biometrics, held in Alghero, Italy, in June 2003. The seven revised tutorial lectures by leading researchers introduce the reader to biometrics-based person authentication, fingerprint recognition, gait recognition, various aspects of face recognition and face detection, topologies for biometric recognition, and hand detection. Also included are the four best selected student papers, all dealing with face recognition.

Image analysis is a computational feat which humans show excellence in, in comp- ison with computers. Yet the list of applications that rely on automatic processing of images has been growing at a fast pace. Biometric authentication by face, ?ngerprint, and iris, online character recognition in cell phones as well as drug design tools are but a few of its benefactors appearing on the headlines. This is, of course, facilitated by the valuable output of the resarch community in the past 30 years. The pattern recognition and computer vision communities that study image analysis have large conferences, which regularly draw 1000 parti- pants. In a way this is not surprising, because much of the human-speci?c activities critically rely on intelligent use of vision. If routine parts of these activities can be automated, much is to be gained in comfort and sustainable development. The - search ?eld could equally be called visualintelligence because it concerns nearly all activities of awake humans. Humans use or rely on pictures or pictorial languages to represent, analyze, and develop abstract metaphors related to nearly every aspect of thinking and behaving, be it science, mathematics, philosopy, religion, music, or emotions. The present volume is an introductory textbook on signal analysis of visual c- putation for senior-level undergraduates or for graduate students in science and - gineering. My modest goal has been to present the frequently used techniques to analyze images in a common framework–directional image processing.

Image analysis is a computational feat which humans show excellence in, in comp- ison with computers. Yet the list of applications that rely on automatic processing of images has been growing at a fast pace. Biometric authentication by face, ?ngerprint, and iris, online character recognition in cell phones as well as drug design tools are but a few of its benefactors appearing on the headlines. This is, of course, facilitated by the valuable output of the resarch community in the past 30 years. The pattern recognition and computer vision communities that study image analysis have large conferences, which regularly draw 1000 parti- pants. In a way this is not surprising, because much of the human-speci?c activities critically rely on intelligent use of vision. If routine parts of these activities can be automated, much is to be gained in comfort and sustainable development. The - search ?eld could equally be called visualintelligence because it concerns nearly all activities of awake humans. Humans use or rely on pictures or pictorial languages to represent, analyze, and develop abstract metaphors related to nearly every aspect of thinking and behaving, be it science, mathematics, philosopy, religion, music, or emotions. The present volume is an introductory textbook on signal analysis of visual c- putation for senior-level undergraduates or for graduate students in science and - gineering. My modest goal has been to present the frequently used techniques to analyze images in a common framework–directional image processing.

The excellently received call for papers of the 13th Scandinavian Conference on Image Analysis, June 29-July 2 (SCIA 2003) resulted in the selected articles of this proceedings. Additionally the volume also contains invited contributions from - Ivar Austvoll, Stavanger University College (NO), - Lars B? a? ath, Halmstad University (SE), - Ewert Bengtsson, Uppsala University (SE), - Rasmus Larsen, Technical University of Denmark (DK), - Jussi Parkkinen, University of Joensuu (FI), - Pietro Perona, California Institute of Technology (US) which brings the total number of articles to 152. The theme of the papers are dominated by the categories - Feature extraction - Depth and surface - Medical image processing - Shape analysis - Segmentation and spatial grouping - Coding and representation - Motion analysis - Texture analysis - Color analysis - Indexing and categorization which also represent the topical groupings of this book. The particularly strong response to the feature extraction, depth and surface, and medical image processing themes makes us believe that these areas are c- rently expansive, partly because of the rich set of problems which remain to be addressed.

We presented an approach for the visual discrimination of children (3-5 years old) from adults using stride-based properties of their walking style. Trajectories of marked head and ankle positions for six children and nine adults were used to compute the relative stride and stride frequency for each walker at di?erent speeds. The distinction between child and adult for these features is quite strong and reduces the task of categorization to a linear discrimination test. Using a trained two-class linear perceptron, we were able to achieve a correct classi?- tion rate of 93-95% for our dataset. Given that only two motion features were used to characterize and di?erentiate children from adults, the result is quite - couraging. The use of natural modes as a means of visual categorization provides a useful bottom-up framework for the classi?cation and recognition of humans in motion. References 1. A. Baumberg and D. Hogg. Learning ?exible models from image sequences. In Proc. European Conf. Comp. Vis. , pages 299-308, 1994. 2. W. Boda, W. Tapp, and T. Findley. Biomechanical comparison of treadmill and overground walking. In Proc. Can. Soc. for Biomech. , pages 88-89, 1994. 3. C.Bregler and J. Malik. Tracking people with twists and exponential maps. In Proc. Comp. Vis. and Pattern Rec. , pages 8-15, 1998. 4. I. Chang and C. Huang. The model-based human body motion analysis system. Image and Vision Comp. , 18(14):1067-1083, 2000. 5. D. Gavrila. Pedestrian detection from a moving vehicle. In Proc. European Conf.

Biometric authentication refers to identifying an individual based on his or her distinguishing physiological and/or behavioral characteristics. It associates an individual with a previously determined identity based on that individual s appearance or behavior. Because many physiological or behavioral characteristics (biometric indicators) are distinctive to each person, biometric identifiers are inherently more reliable and more capable than knowledge-based (e.g., password) and token-based (e.g., a key) techniques in differentiating between an authorized person and a fraudulent impostor. For this reason, more and more organizations are looking to automated identity authentication systems to improve customer satisfaction, security, and operating efficiency as well as to save critical resources. Biometric authentication is a challenging pattern recognition problem; it involves more than just template matching. The intrinsic nature of biometric data must be carefully studied, analyzed, and its properties taken into account in developing suitable representation and matching algorithms. The intrinsic variability of data with time and environmental conditions, the social acceptability and invasiveness of acquisition devices, and the facility with which the data can be counterfeited must be considered in the choice of a biometric indicator for a given application. In order to deploy a biometric authentication system, one must consider its reliability, accuracy, applicability, and efficiency. Eventually, it may be necessary to combine several biometric indicators (multimodal-biometrics) to cope with the drawbacks of the individual biometric indicators.

Biometric authentication refers to identifying an individual based on his or her distinguishing physiological and/or behavioral characteristics. It associates an individual with a previously determined identity based on that individual s appearance or behavior. Because many physiological or behavioral characteristics (biometric indicators) are distinctive to each person, biometric identifiers are inherently more reliable and more capable than knowledge-based (e.g., password) and token-based (e.g., a key) techniques in differentiating between an authorized person and a fraudulent impostor. For this reason, more and more organizations are looking to automated identity authentication systems to improve customer satisfaction, security, and operating efficiency as well as to save critical resources. Biometric authentication is a challenging pattern recognition problem; it involves more than just template matching. The intrinsic nature of biometric data must be carefully studied, analyzed, and its properties taken into account in developing suitable representation and matching algorithms. The intrinsic variability of data with time and environmental conditions, the social acceptability and invasiveness of acquisition devices, and the facility with which the data can be counterfeited must be considered in the choice of a biometric indicator for a given application. In order to deploy a biometric authentication system, one must consider its reliability, accuracy, applicability, and efficiency. Eventually, it may be necessary to combine several biometric indicators (multimodal-biometrics) to cope with the drawbacks of the individual biometric indicators.

This book constitutes the refereed proceedings of the First International Conference on Audio- and Video-based Biometric Person Authentication, AVBPA'97, held in Crans-Montana, Switzerland, in March 1997.The 49 revised papers presented were carefully reviewed and selected by the program committee for inclusion in the book; also included are four invited contributions. The papers are organized in sections on facial features localisation, lip and facial motion, visual non-face biometrics, face-based authentication, text-dependent speaker authentication, text-independent authentication, audio-video features and fusion, and systems and applications.