Overcoming Object Tracking Challenges for Abnormal Behavior Recognition

Khalid Abdul-Aziz Al-Shalfan

This paper is about developing an Intelligent Video System which takes advantage of the latest technology and research enhancement. The goal is to let such a system automatically recognize an abnormal human behavior. We realized it and implemented it with a several scenarios using the SVM model. The results were excellent as we reached the rate of less than 1% of false positives in certain conditions. Many datasets have been developed for such activities like the UT-Interaction dataset and the UCR Video web dataset. As far as our work was concerned, we used a dataset we developed within the university premises containing both simple and complex activities. The experimental results on real-time video streams show the feasibility of our system and its effectiveness in human activity tracking and recognition.

Select Volume / Issues:

• Volume 5,Issue 3,May 2014

Year:

2014

Type of Publication:

Article

Keywords:

Abnormal Behavior; Real-Time Detection; Multi-Camera Tracking; Video Surveillance

Journal:

IJECCE

Volume:

5

Number:

3

Pages:

478-484

Month:

May