Images containing faces are essential to intelligent vision-based human computer intcraction, and research efforts in face processing incluse face recognition, face tracking, pose estimation, and expression recognition. However, all of these researching directions involve in one problem-----face detection and location, in other words, before this face processing, we must know faces'locations and scales. Consequently, to build an automated face processing system which analyzes the information contained in face images, robust and efficient face detection algorithms are require.Recently, human face detection, as a key technology in human face information processing , is becoming a big problem that gathers more attention in the field of pattern recognition and computer vision. It is widely applied in commercial and law area,such as mug shots retrieval, real-time video surveillance in security system and cryptography in bank and so on. Face recognition has direct, friendly characteristics and it is no psycholigical obstacle for users. This dissertation mainly studies the approaches to frontal face detection and recognition. The main research works and contributions are as the following,(1) Colour segmentation. on the basis of the common color space YCrCb, this paper put forward a new color space YCgCb which has better colour clustering effect , in the color space This paper constructs a combine model using a table combined by light(Y) and chroma(Cg,Cb),.Propose a rapid and effective threshold estimation methods, The experiment expresses that complexion model can examine a person face position availably.(2) Location of facial characteristics. It includes the position of human eyes and mouths. Eye position: We determine the horizontal position of human eyes by the gray's derivatve features because of the eye's characteristics that great changes of gray-scale have been taken from the skin, the white of eye,Pupil to skin.The paper make use of the otsu method to threshold the candidate area of human eyes and then identify the locations around the human eyes using the integral projection.The method is simple, effective and a small amount of calculations. Mouth position: people's mouth can be targeted effectively by the tricomponent and chromaticity component of RGB, meanwhile, I can locate accurately the human face according to the triangular relationship of the human eye and mouth.(3) Location of facial characteristics based on wavelet transform. I put forward an improvementalgorithm because the above methods on the human eye's opening and closing have a certain demands. The high frequency details of human face can be obtained through the wavelet transform and then using the method of projection can posite the features of human face.This algorithm is strong enough to facial expression,decoration and deflection. |