Abstract:

With the development of artificial intelligence theory and technology， face detection and recognition， as an important branch of computer vision， has become the focus of academic research in recent years. This paper studies and implements the Adaboost face detection algorithm and the face recognition algorithm based on eigenfaces. Using the cascading characteristics of Adaboost detectors， a face classification algorithm based on gradient direction pyramid and support vector machine is introduced， which removes part of the false detection results； a segmentation algorithm based on the skin color model is introduced to accurately perform the face area. Face recognition are experimented on the second-generation ID card， Feret face database， and internet image resources. The results show that the algorithm has a significant effect on frontal faces， with fast recognition speed and high accuracy.

Key words:

face detection, face recognition, Adaboost, gradient direction pyramid, support vector machine