Violence Detection in Surveillance Scenes
Borivoje (Borko) Furht, Ph.D.
Borko Furht, Ph.D., is a professor in the Department of Electrical & Computer Engineering and Computer Science at Florida Atlantic University (FAU) in Boca Raton, Florida. He is also Director of the NSF-sponsored Industry/University Cooperative Research Center on Advanced Knowledge Enablement at FAU. Before joining FAU, he was a vice president of research and a senior director of development at Modcomp (Ft. Lauderdale), a computer company of Daimler Benz, Germany, a professor at University of Miami in Coral Gables, Florida, and a senior researcher in the Institute Boris Kidric-Vinca, Yugoslavia. Professor Furht received Ph.D. degree in electrical and computer engineering from the University of Belgrade. His current research is in multimedia systems, multimedia big data, video coding and compression, 3D video and image systems, wireless multimedia, cloud computing, and social networks. He is presently Principal Investigator and Co-PI of several multiyear, multimillion dollar projects. He received total funding of $15 million from government agencies including NSF, NIH, Department of Navy, DoD, and NASA, and private industries including IBM, Google, Apple, LexisNexis, Motorola, and others. He is the author of numerous books and articles in the areas of multimedia, computer architecture, real-time computing, and operating systems. He is a founder and editor-in-chief of the Journal of Multimedia Tools and Applications (Springer) and he recently co-founded Journal of Big Data (Springer). He has received several technical and publishing awards, and has consulted for many high-tech companies including IBM, Hewlett-Packard, Xerox, General Electric, JPL, NASA, Honeywell, and RCA. He has also served as a consultant to various colleges and universities. He has given many invited talks, keynote lectures, seminars, and tutorials. He served as Chairman and Director on the Board of Directors of several high-tech companies and as an expert witness for Cisco, Qualcomm, Adobe, and Bell Canada.
This REU project is focused on detecting and localizing regions of violence in surveillance scenes. The participant will learn how to extract candidate regions from surveillance video, apply computer vision techniques to identify potential regions of violence, and develop descriptors to distinguish violent versus non-violent regions.
The training will center on understanding the design, implementation, and application of the Criminal Aggression Recognition Engine (CARE), an OpenCV-based violence detection system, implemented in Java, which relies on machine learning to distinguish violent versus non-violent events. The system can be trained using video datasets to classify action in live feeds as violent or non-violent based on a configurable set of extracted features. The REU participant will apply the CARE system to a range of video datasets to evaluate its performance in distinguishing violent versus non-violent activity across a suite of extracted features. The results of this experimentation will be used to guide the design and evaluation of new features based on modifications to existing extraction algorithms. The project will provide a meaningful experience for the participant, while contributing to Borko Furht’s ongoing work in this area.