Facial Feature Extraction Using Enhanced Active Shape Model

Active shape model (ASM) is one of the most popular local texture models for face detection. This paper addresses issues related to face detection and implements an efficient extraction of facial point features which is more suitable for using on mobile devices. The proposed and implemented algorithm in this paper is modified to enhance its performance (1) improving the initialization model using center of the eyes by using a feature map of RGB color information, and (2) building and extending a 2-D profile model for detecting faces in an image. To evaluate the performance of our work, we assess the ratio of success with common face image database. The obtained result shows that our algorithm described in this paper is enough to applicable to mobile environments.