Vision based human activity tracking using artificial neural networks

The tracking and understanding the behavior of human beings is an important problem in a number of industrial applications. At the highest level, the system capable of resolving this problem should be able to recognize the human behavior and understand the intent and motive from the observation alone. This is a difficult task, even for humans to perform, and the mistakes and misinterpretations are very common. In this text, a method for the design of an artificial neural network based intelligent human activity monitoring system is proposed, that can detect and track suspicious activity in a typical surveillance environment. This method makes use of the silhouette pattern of the human blob obtained from segmentation of the scene captured by the camera.