In this thesis, a study on face detection and standardization problem with video signal is presentd. 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 those researching directions involve in one problem– face detection and location.The research on face detection has lasted for more than twenty years. But, up to now, due to the complexity of the purpose such as the diversity of face paterns, variable lighting condition and so on, many researches can not resolve the problem completely even if they have studied it for long time. In this thesis, the author has done some work on the face detection.The work includes:(1) Movement detection and skin segmentationMovement detection finds out the movement region; Skin segmentation method segments'skin'and'non-skin'in an input image, depress the calculation.(2) Multiple templates matchingOperator burden is too large for templates matching, which can not satisfy real-time system. So the author proposes binary tamplate mathing. It can do traditional template matching could do, but the operator burden is nearly zero.(3) Neural netwok (NN)NN has large amount of nodes and greate remembrance. Author proposes a method, this combined with face detection and eyes location, based on BP-NN. Also operator bueden is large, but it is very easy to transplant to FPGA. |