This Paper explores the Technique to identify the psychology of a human being through Multi- feature Face Recognition system by Neural Network. The system works in three modes: Training, Classification and Matching – Emotion. In the training mode face images are normalized according to different emotions, extract appropriate face feature by using Principle and independent component analysis (PCA) & (ICA). The extracted features are then trained in parallel to partition in different face classes using Back- propagation neural network (BPNN). In classification, the trained face images are fed with new face image. Scored based strategy works with both PCA BPNN and ICA BPNN to classify given new face image according to trained face classes. In matching mode, the face image(s) is to be matched with the set of particular emotion and identify the emotion of the current face image.
Emotion recognition, Multi feature face recognition, Neural Network, Scored Based Strategy
