Automatic extraction of facial feature plays an important role on the application human face recognition. This thesis paper presents our method for automatically extracting facial features from a given image that consists of a head-and-shoulders view of the person and a complex background scene. Three main steps designed in our approach to extract facial features are face segmentation, location of facial feature region and extraction of facial feature. A skin-color face segmentation method is developed to segment face region. Eye feature, lip color and generalized face geometry and proportions are used to locate the desired regions containing facial features. The extraction of corresponding facial feature is based on different design method applied to the locate region. Experiments confirm the feasibility of the proposed approach, some phenomena observed from the results are discussed, the inherent limits of the approach are given, and some interesting and valuable problems are pointed out.