Thesis Title: Automated Behavioural Recognition From Visual Data

Research on video surveillance systems for understanding human behaviour has grown steadily in recent years. However, most people find having a camera in their homes as invasion of their privacy; even if the video recording is not transmitted to another place. For this reason, there is a need to develop a system, capable of making decisions at the camera end with simple pre-processing techniques. 

The focus of this work is to identify simple and computationally inexpensive detection and classification techniques. The system will be capable of automatically classifying the occupants behaviour and also adopt to long-term changes in the environment.

Director of Studies

Dr Kofi Appiah

Research Groups/Centres and Projects

Computational Intelligence and Applications Research Group (CIA)

Publications

• Suad Albawendi, Kofi Appiah, Heather Powell and Ahmad Lotfi "Overview of Behavioural Understanding System with filtered Vision Sensor", Proceedings of the International Conference on Interactive Technologies and Games iTAG 2015.