Face Detection Technique Based on Skin Color and Facial Features

Abstract

Face detection is an essential first step in face recognition systems with the purpose of localizing and extracting the face region from the background. Apart from increasing the efficiency of face recognition systems, face detection technique also opens up the door of opportunity for application areas such as content based image retrieval, video encoding, video conferencing, crowd surveillance and intelligent human computer interfaces. This thesis presents the design of face detection approach which is capable of detecting human faces from complex background. A skin color modeling process is adopted for the face segmentation process. Image enhancement is then used to improve the face candidate before feeding to the face object classifier based on Modified Hausdroff distance. The results indicate that the system is able to detect human faces with reasonable accuracy