Description

We propose an automatic battery power management scheme which makes use of human interaction with the system. Here we stimulate and test the problem using a captured video sequence. We intend to detect the face of the person who is sitting in front of the screen. Haar face detector is implemented to detect the face of the person in all the input frames. The detected coordinates are used to crop the input data. These datas are subjected to invariant moment feature extraction which has the capability to recognize shape based features. These features are labeled depending on the user attention level into four states (Interactive State, Attentive State, Inattentive State, and Away State). These features are then classified using probabilistic Bayesian classifier. Depending on the status of the classified frame, the input video sequence will provide a message, whether the laptop should be switched off or not.