Research On Two-Classifier-Based Face Detection

Face recognition have gotten a lot of concern from researchers in recent years, because there is a great foreground in many fields,such as video surveillance, security verification and home enterainment.And it has become a popular research topic in the field of bilology feature identification.In fact,face recognition is the most convenient and direct means,compared with other ways of biology feature identification.Face detection has been intensively researched in recent years due to their importance in apolications such as natural human computer interfaces, content-based retrieval,object based video compression,digital video process and video surveillance.And it has been a most interesting and challenging topics in the fields of Artificial Intelligence and Computer Vision.The purpose of face detection from a given image or video is positioning of the region face.Face detection is an indispensable pre-processing work in the proceeding of the face recognition.It is also a non-rigid object recognition of exploration.Because of huge computation,and how quickly and efficiently detect human faces,face detection has become an important topic.Statistical learning theory is a kind of new technology in machine learning field. Its main idea is to control learning machine's generlization ability by controlling its model complexity.Support vector machine is a new kind of learning machine based on statistical learning theory,which has advantages.It solves small-sample problems by using structural risk minization to take the place of empirical risk minization. Because of such advantages,SVM becomes a hot spot of machine learning theory, and is applied successfully.In this thesis,the current main task is to summarize the detection method.Based on the previous,in order to improve the detection speed,we represent a feature-based and image-based method of combining the rapid method of face detection.The thesis introduces the support vector machine.In order to filter out background information on a large number of non-human face,the thesis introduces rectangular eigenvector and rough classifier. Face detection and localization is a highly practical application.The particular demand of application must be taken enough attention on the process of detection and localization of human face,which is such a complicated high-dimension pattern. Intergrated methods are frequently employed to achieve the accuracy and real-time ability of the system.