Face Detection Algorithm Based on Skin Color and Edge Density

Face detection is an important part of many image analysis processes like video surveillance, facial expression recognition etc. This paper presents a fast and easy to implement algorithm for face detection. It is based on the skin detection from Y, Cb and Cr components augmented by facial elements estimate from edge density on the candidate area. Our method is invariant to the position of the face since it does not try to detect facial features like eyes or lips, but rather estimates their presence from the characteristics of edge density on the considered area. On different test examples our method achieved very good results so it can be considered a promising tool for many applications where face detection is necessary. Key-words: Image processing, Face detection, Skin color detection, Edge detection