A comparative study on face detection algorithms

Due to massive growth of video information and their usage in surveillance applications, a more sophisticated and an intelligent examination systems development is far more critical and exceptionally important. Peoples are one of most common and very specific objects in video which were tracked with the aid of face detection algorithms. Although face detection algorithms are inherently complex due to the complex environment (illumination, pose, age, color coding, occlusion etc.), we have seen the development of some intelligent and accurate face detection algorithms to achieve the intended goal of face localization. In this paper, we have made an attempt to study four such well known algorithms namely Color based, Normalized Pixel Difference (NPD), Viola-Jones and Independent Component Analysis (ICA) based face detection methods and explore their merits and suitability by considering different situations. We have also described detection technique with some application domain along with different challenges in this field.