| Images containing faces are essential to intelligent vision-based human computer interaction, and research efforts in face processing include 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, face detection algorithms are absolutely necessary.The research on face detection has lasted for more than twenty years. But, up to now, due to the complexity of the face detection, many researchers can not resolve the problem completely even if they have studied it for long time. In the thesis, some work on the face detection has been done. Research work includes the following contents:Firstly, current typical algorithms on face detection are concluded and analyzed, and then a coarse-to-fine detection algorithm is introduced to detect the number, size and position of face in color images. By analyzing and comparing the cluster of face skin in different color space, a mixed skin model is proposed. The model is combined by Gauss model based on the color space of YCbCr and Region model based on the color space of rgb. And the same time, it is validated and analyzed. After analyzing the connected component and using the filter based on mathematical morphology, the candidate facial regions can be obtained. In the selective and validated stage, we mainly utilize faces' geometrical characters to roughly filter in these candidate regions. Finally, we further verify these regions by detecting mouth and output the result.Secondly, in the thesis, the features' localization is researched and studied. In localizing eyes, the chromatic and luminance of eyes are mainly taken into account, and an algorithm to locate eyes, which is based on matching chromatic and luminance... |
PDF Full Text Request