Abstract
The key to surveillance systems is object detection. Many approaches to this task have been implemented over multiple decades, but still lag efficient performance in dynamic backgrounds. Systems developed so far requires complex computations and human beings intervention for parameter adjustments. The proposed method is based on a neural-fuzzy model. The neural stage, based on a one-to-one self-organizing map (SOM) architecture, deals with the dynamic background for object detection as well as shadow elimination. The fuzzy inference mamdani system mimics human behavior to automatically adjust the main parameters involved in the SOM detection model, making the system independent of the scenario. The model developed provides robustness over real video scenes.
