A Hybrid Data Mining Technique For Face Recognition System Using Enhanced Independent Component Analysis

Face recognition is one of the most reliable and adaptive systems, which has the benefits of being passive, non-intrusive system for verifying personal identity, and is impossible to deceive as the process involves unique identification methods. The problems in existing technologies are overall accuracy, sensitivity to changes in lighting, camera angle, pose and increased computational time. The proposed Face Recognition system uses a hybrid Data Mining approach which produces promising results for high accuracy as well as reduced computational time. The proposed face recognition system employs hybrid approach with Neural network and Genetic algorithm with an ensemble enhancement of the Independent Component Analysis (EICA). ICA is more adaptive model used for decision making and is insensitivity to large illumination and facial expressions in the recognition process.