An Analysis on Face Recognition using Principal Component Analysis Approach.

Face recognition is a type of biometric recognition based on human facial feature information. Human face images or videos can be automatically collected using a high definition camera. Advanced technologies then can be used for face recognition by tracking on collected images to detect human faces. The facial recognition algorithm can cut out the main facial area after detecting the face and find the key facial feature points, and input it into the recognition algorithm after processing. To extract and compare the facial features, the recognition algorithm is used to the complete the final classification. This research is to study the face recognition using PCA (Principal Component Analysis. The PCA-based is used to eigenface recognition, Hotelling transform in PCA is used to obtain the main components of the face distribution, i.e. the feature vectors (eigenfaces). The face images from the training library and the face images to be recognized are projected onto this space separately to match the recognized image output based on the principle of minimum geometric distance. The face angle, face mask, and face expression factors are selected for testing against face recognition. Hence questions and hypotheses are formulated to verify whether the recognition rate of face recognition is influenced by these factors for this recognition method. Based on the results of the analysis it was that face angle, face masking have a positive effect on face recognition. Furthermore, according to the analysis, it can be concluded that face masking has the highest significance for face recognition.