Fight detection based on Hidden Markov Model

This paper presents a novel approach to detect human fights based on Hidden Markov Model (HMM). We present two HMM models to the problem. The first one is Fight Model, the second is Ordinary Model. According to the motion analysis between people in the scene, features for human behavior have been proposed. Given the observation value of features in time sequence, the probability can be evaluated by two Models. The larger value means that the Model is the suitable one to describe what happens in the scene. Experimental result demonstrats that the method is robust and efficient in detecting human fight behaviors.

[1]  A. Sathyanarayana,et al.  Driver behavior analysis and route recognition by Hidden Markov Models , 2008, 2008 IEEE International Conference on Vehicular Electronics and Safety.

[2]  Guangyou Xu,et al.  Human action recognition in smart classroom , 2002, Proceedings of Fifth IEEE International Conference on Automatic Face Gesture Recognition.

[3]  M. P. S Bhatia,et al.  Data clustering with modified K-means algorithm , 2011, 2011 International Conference on Recent Trends in Information Technology (ICRTIT).

[4]  Yong Xu,et al.  Fighting Detection Based on Optical Flow Context Histogram , 2011, 2011 Second International Conference on Innovations in Bio-inspired Computing and Applications.

[5]  Yangsheng Xu,et al.  Modeling human actions from learning , 2004, 2004 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) (IEEE Cat. No.04CH37566).

[6]  Lawrence R. Rabiner,et al.  A tutorial on hidden Markov models and selected applications in speech recognition , 1989, Proc. IEEE.

[7]  Nikos Fakotakis,et al.  Fusion of acoustic and optical sensor data for automatic fight detection in urban environments , 2010, 2010 13th International Conference on Information Fusion.