Face detection by facial features with color images and face recognition using PCA

Human face detection plays an important role in applications such as video surveillance, human computer interface, face recognition, and face image database management. The face recognition by a CCD camera has the merit of being linked with other recognition systems such as an iris recognition to implement a multimodal recognition system. This paper is concerned with a new approach to face recognition that is automatically distinguished from moving pictures. Based on the research about recognition of color image by a CCD camera, we first find the proper value of color images in order to distinguish the tone of skin from other parts of face. Then, we look for the skin color among the regions of skin color converting RGB into Y, Cb, Cr to find skin parts of face. This new method can be applied to real-time biometric systems. We have developed the approach to face recognition with eigenface, focusing on the effects of eigenface to represent human face under several environment conditions. Finally an error rate is compared when face recognition is processed with facial features through the PCA (principal component analysis).