Eye Detection based on Viola & Jones Detector, Skin Color, and Eye Template

Abstract

In this paper, we propose a new method of eye detection based on Viola & Jones detector, skin color and eye template. Our method is processed into three steps: In the first step, faces are detected by using the Viola & Jones detector. In the second step, face images are segmented into skin regions and non-skin regions with a threshold calculated by a combination of rules proposed on elements of the three color spaces RGB, HSV and YCbCr. The third step is the matching of an eye template with non-skin regions. Our method is tested on the FEI database available online, as well as a set of personal images and other from the Internet. The obtained results are satisfactory in terms of the quality of detection and rapidity.

A New Local Descriptor Based on Strings for Face Recognition

“…A face recognition system consists of three critical steps: detection of a face in an image based on the position of the eyes [3,4], feature extraction stage, and the classification stage. Generally, a key issue in face recognition is finding an efficient facial image representation as a feature vector.…”