Total views : 231

Recognition of Abnormal Human Behavior using Kinect Case Study: Tehran Metro Station

Affiliations

  • Computer Science, Artificial Intelligence, Semnan University, Semnan, Iran, Islamic Republic of
  • Information Technology, Semnan University, Semnan, Iran, Islamic Republic of
  • Business Administration - Computer Information Systems, Colorado State University, Fort Collins, United States

Abstract


Objectives: The ability of detecting people and understanding their behaviors for recognition or detection of special event has attracted significant research interest in recent years. Methods: The main purpose of this study focuses on abnormal human behavior detection by using Microsoft’s Kinect camera. By utilizing Kinect, the system tracks all persons in Metro station and calculates their position relations, which provide a more effective way to analyze human’s behavior. Results: It first applies Kinect technology of Microsoft in the field of intelligent monitoring. By taking use of the depth information, the system recognizes unusual behavior and by face detection and analysis, the system of face-hiding behavior. The time people spend in Metro station is recorded and spending too much time is also regarded as an abnormal behavior. Applications: Experiments show that the system is very robust for abnormal behavior recognition and provides a new way for preventing Metro station crimes.

Keywords

Abnormal Human, Behavior, Kinect, Recognition.

Full Text:

 |  (PDF views: 283)

References


  • Nguyen NT, Phung DQ, Venkatesh S, Bui H. Learning and detecting activities from movement trajectories using the hierarchical hidden Markov model. Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition. 2005 Jul; 2:955–960.
  • Dai P, Di H, Dong L, Tao L, Xu G. Group interaction analysis in dynamic context. IEEE Transactions on Systems, Man and Cybernetics, Part B (Cybernetics). 2009 Feb; 39(1):34–42.
  • Recognition of high-level group activities based on activities of individual members. 2008. Available from: http://ieeexplore.ieee.org/document/4544065/
  • Yao A, Gall J, Van Gool L. Coupled action recognition and pose estimation from multiple views. International Journal of Computer Vision. 2012 Oct; 100(1):16–37.
  • Martin R, Sujatha S. Bottom-up approach of modeling human decision making for building intelligent agents. Indian Journal of Science and Technology. 2016 Jan; 9(4):1–5.

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution 3.0 License.