Face recognition for authentication by using anthropometric model

Abstract EN

This paper presents an automatic technique for detecting important facial features’ points using a developed anthropometric face model.

The facial features’ points we work on are about the areas of mouth, nose, eyes and eyebrows.

The anthropometric means the scientific study of the measurements and proportions of the human face.

Several processes are performed in order to recognize human personality authenticated or not, these processes are beginning by capturing colored image using fixed digital camera and ending by features isolated into separated sub images and the lengths and distances among them representing authenticated persons information are stored into database..

In authentication stage all the extracted features are compared with stored authenticated facial features in the database, the person is authenticated if a percentage of similarity equal to or greater than 78% is achieved.