(IJCSIS) International Journal of Computer Science and Information Security, Vol. 9, No. 6, June 2011 AUTOMATED FACE DETECTION AND FEATURE EXTRACTION USING COLOR FERET IMAGE DATABASE Dewi Agushinta R. 1 , Fitria Handayani S. 2 1 Information System, 2 Informatics Gunadarma University Jl. Margonda Raya 100 Pondok Cina, Depok 16424, Indonesia 1 dewiar@staff.gunadarma.ac.id, 2 dearest_v3chan@student.gunadarma.ac.id Abstract—Detecting the location of human faces and then extracting the facial features in an image is an important ability with a wide range of applications, such as human face recognition, surveillance systems, human-computer interfacing, biometric identification, etc. Both face detection and face features extraction methods have been reported by the researchers, each with a separate process in the field of face recognition. They need to be connected through adapting the face detection results to be the input face in the extraction process by turning the minimum face size results from the detection process, and the way of face cropping process from the extraction process. The identification and recognition of human face features that has developed in this research is the combination of face features detecting and extracting process in 150 frontal single still face images from color FERET facial image database with additional extracted face features and face features’ distances. Keywords- face detection, face extraction, face features, face recognition, feret I. INTRODUCTION Biometrics encompasses automated methods of recognizing an individual based on measurable biological (anatomical and physiological) and behavioral characteristics. As a biometric, facial recognition is a form of computer vision that uses faces to attempt to identify a person or verify a person’s claimed identity [1]. Facial recognition systems are more widely used than just a few years ago. Most of the pioneering work in face recognition was done based on the geometric features of a human face [2]. Detecting the location of human faces and then extracting the human face features in an image is an important ability with a wide range of applications, such as human face recognition, surveillance systems, human computer interfacing, video-conferencing, etc [3]. Detecting human faces and extracting the facial features in an unconstrained image is a challenging process. There are several variables that affect the detection performance, such as different skin coloring, gender, and facial expressions. One of the researches in face recognition system is face image detection from a single still image where the face image part is detected by skin color model analysis [4], [5]. The determination of face region in the research was not complete, because some parts of the face were not included in the extraction process. If one component of face features is detected, the position of other components would be obtained and the component could be extracted. Other research in face recognition system is the face features extraction which has obtained eye and mouth region and the distances between eye and mouth [6]. In this paper, based on those researches, we develop a system that separates a face image into face components, and then extracts components in the regions of eyes, nose, mouth, and face boundary from a frontal single still face images from color FERET facial image database with additional extracted face features and face features’ distances. II. METHODOLOGY The techniques in image processing that is used in face recognition are varied [7], yet they have the same three basic stages, as illustrated in Fig. 1. First, the system gets the input of an image. Then, the system will detect the face region. After that, it will extract the face image in order to get the face features, for instance eye, nose and mouth. Finally, the system uses the detected face image to obtain the information of each face feature that has been extracted. Figure 1. Configuration of a generic face recognition system. 313 http://sites.google.com/site/ijcsis/ ISSN 1947-5500