Sensor based Environment Adaptive Learning System for Person Authorization

This paper presents a robust distributed architecture for adaptive and intelligent systems, called EAS (environment adaptive learning scheme), with self-learning capability to use under dynamic and uneven environments. Our proposed system adopts the concepts of situation-awareness with the evolutionary computations where the working environments are modeled and identified as environmental situations. We have used ART2 for environment modeling while unsupervised learning algorithm for environment identification. Environment adaptive algorithm, for its adaptive criteria, is used to explore action configuration for each identified situation to implement our concept. We have achieved very encouraging experimental results for CCD camera visual sensor based face detection, recognition system and ECG sensor system.