Video context analysis is an active and vibrant research area, which provides means for extracting, analyzing and understanding behavior of a single target and multiple targets. Over the last few decades, computer vision researchers have been working to improve the accuracy and robustness of algorithms to analyse the context of a video automatically. In general, the research work in this area can be categorized into three major topics: 1) counting number of people in the scene 2) tracking individuals in a crowd and 3) understanding behavior of a single target or multiple targets in the scene.…mehr
Video context analysis is an active and vibrant research area, which provides means for extracting, analyzing and understanding behavior of a single target and multiple targets. Over the last few decades, computer vision researchers have been working to improve the accuracy and robustness of algorithms to analyse the context of a video automatically. In general, the research work in this area can be categorized into three major topics: 1) counting number of people in the scene 2) tracking individuals in a crowd and 3) understanding behavior of a single target or multiple targets in the scene. This book focusses on tracking individual targets and detecting abnormal behavior of a crowd in a complex scene. Firstly, this book surveys the state-of-the-art methods for tracking multiple targets in a complex scene and describes the authors' approach for tracking multiple targets. The proposed approach is to formulate the problem of multi-target tracking as an optimization problem of finding dynamic optima (pedestrians) where these optima interact frequently. A novel particle swarm optimization (PSO) algorithm that uses a set of multiple swarms is presented. Through particles and swarms diversification, motion prediction is introduced into the standard PSO, constraining swarm members to the most likely region in the search space. The social interaction among swarm and the output from pedestrians-detector are also incorporated into the velocity-updating equation. This allows the proposed approach to track multiple targets in a crowded scene with severe occlusion and heavy interactions among targets. The second part of this book discusses the problem of detecting and localising abnormal activities in crowded scenes. We present a spatio-temporal Laplacian Eigenmap method for extracting different crowd activities from videos. This method learns the spatial and temporal variations of local motions in an embedded space and employs representatives of different activitiesto construct the model which characterises the regular behavior of a crowd. This model of regular crowd behavior allows for the detection of abnormal crowd activities both in local and global context and the localization of regions which show abnormal behavior.
Produktdetails
Produktdetails
Synthesis Lectures on Image, Video, and Multimedia Processing
Dr. Myo Thida received her Ph.D. degree from Kingston University, UK, in 2013. She received the B.Eng. and M.Eng. degrees from Nanyang Technological University(NTU), Singapore in 2005 and 2008 respectively. She is currently a Research Scientist with the Institute for Infocomm Research, Singapore. She was the recipient of NTU undergraduate scholarship in 2001 and the M.Eng. research scholarship in 2005. Her current research interests include computer vision and machine learning algorithms which focus on interpreting and understanding human activities in surveillance videos. Dr. Myo Thida won the Best Poster Prize from Springer at International Summer School on Pattern Recognition in 2011. She, together with her team members, also received the 2008 IES Prestigious Engineering Achievement award from Institution of Engineers, Singapore and the 2012 IWA applied research honour award from International Water Association for the development of an intelligent surveillance system for detectingcontaminants in drinking water. Dr. How-lung Eng received his BEng and Ph.D. degrees both in Electrical and Electronic Engineering from Nanyang Technological University, Singapore, in 1998 and 2002, respectively. Currently, he is with the Institute for Infocomm Research, Singapore as a research scientist and Programme Manager of Video Behavioral Analytics Programme. His research interest includes real-time vision, pattern classification and machine learning for abnormal event detection. Dr. How-lung Eng has made several PCT fillings related to video surveillance applications and has actively published his works in the above areas of interest. He was a recipient of the Tan Kah Kee Young Inventors' Award 2000 (Silver, Open Section) for his Ph.D. study, and a recipient of the TEC Innovator Award 2002, the IES Prestigious Engineering Awards 2006 and 2008 and IWA PIA Asia Pacific Regional Award 2012 for his works in the areas of video surveillance and video monitoring to ensure safe drinking water. Dr. Dorothy Monekosso (Ph.D.) is Associate Professor in Medical Electronics and Robotics at the University of The West of England, UK, Head of Electronics and Computing and Research Lead at the Bath Institute of Medical Engineering (BIME), UK. Her research interests include intelligent environments and robotics. Dr. Monekosso is the author of over peer-reviewed publications in scientific journals and conferences and is guest-editor of book collections, published by Springer-Verlag, in the field of ambient intelligence and Intelligent Paradigms in Security. She is a guest editor on special issues of the IEEE Transactions of Automation, Science and Engineering on ""Automation and Engineering from Ambient Intelligence"" 2009 and the Journal of Information Fusion (2013). Dr. Monekosso is a member of the following editorial boards; the International Journal of Robotics and Automation (IJRA), the International Journal on Advances in Intelligent Systems (IJAIS), and International Journal of Advanced Intelligence Paradigms (IJAIP). She has served on a number of conference organizing and programme committees. Her interests include intelligent systems, ambient assisted living, and robotics. Professor Paolo Remagnino is an expert in image and video understanding. He has worked as a researcher at INRIA, Grenoble, and the University of Reading. Since 1998, he has been with Kingston University where he leads the Robot Vision Team (RoViT). His research interests are in image and video understanding, pattern recognition, machine and manifold learning. Prof. Remagnino has published more than 120 papers in international journals and conferences. He has served as associate editor for the IEEE Transactions of Systems, Man and Cybernetics, part A, IEEE Transactions of Automation, Science and Engineering, the Elsevier Image and Vision Computing and Neurocomputing. Prof. Remagnino is a member of the UK EPSRC college and has served as evaluator for Britishand European funding research bodies. Professor Remagnino is currently principal investigator of four grants on the intelligent monitoring of public and private spaces, funded by the EU FP7 programme (PROACTIVE, CARING4U and BREATHE) and a project on advanced video surveillance funded by the U.S. Department of Homeland Security. Prof. Remagnino has an active collaboration on classification of plants and he is associate researcher with the Royal Botanic Gardens, Kew.
Inhaltsangabe
Introduction.- Literature Review.- Tracking Multiple Targets Using Particle Swarm Optimization.- Abnormality Detection in Crowded Scenes.- Conclusion.- Bibliography.- Authors' Biographies.
