Design And Implementation Of Color-based Face Detection And Location

In recent years, more and more people and organizations began to study the face recognition technology, appropriate; face recognition technology has become a hot research direction. This phenomenon has mainly two reasons:first, the frequent economic activity continued social and economic development brought about by the rapid flow of leads to the population, resulting in a series of security problems; secondly, the rapid development of computer technology, more and more people began to study computer-related technologies, some of the technology has matured and has been put on the market. For the reasons above two aspects, as image processing and pattern recognition in the field of face recognition technology, it has become a hot area of research. Face recognition is to identify the identity of the person on the basis of biometric-based technology. Common biometrics includes palmprint recognition, fingerprint recognition, voice recognition, iris recognition, and so on. This paper studied human detection and recognition technology consists of three parts: face detection, facial feature extraction, and face recognition. These three aspects are successively interconnected, indispensable. Among them, face detection and localization refers to identify a face from a given photo or image that you want to find the location of the face, requiring split the face region from color or black and white picture; human face image feature extraction and this link requires recognition of face detection and localization session segmented face image normalization, and the results related treatment.Similarly, in all areas of automatic face recognition technology can be subdivided into much research-related technologies. Color-based face detection and location technology are the main contents of this major study. In recent years, with the development of face recognition technology, face detection, there have been a number of key technologies, mainly face detection method based on learning, knowledge-based face detection methods, as well as a template matching method, so this article first, these types of methods are reviewed, introduces the basic theory of face detection and recognition, on this basis, but also introduces the key technologies of monitoring and face recognition, focusing on core lists several more typical approach; then we cut the key elements of this study, based on the skin color of the face detection setting to focus the discussion, including, in a given image, using techniques based on the color of the photo face detection and feature extraction, and face parts shown.This paper presents a detection algorithm based on skin color regions, the basic idea is the use of fuzzy C-means image segmentation, image chip area, according to the area of skin color pixel ratio to determine whether the entire area of the skin area. This method is simple, can improve the detection rate. Fuzzy C-Means method acts only on the color attribute, is based on the color attributes, often given in the entire color image region segmentation, to produce certain parts, are more sensitive to light. Therefore, in order to improve the quality and efficiency of testing, we do not intend to adopt the overall use of fuzzy C-means, but in a given image, the introduction of this kind of fuzzy clustering properties of local spatial distribution of the pixel. This solution can make the face area is separated in a unified region to improve accuracy. After going perfect face detection to locate, in this part, we in the geometric model of the human face, the characteristics of the human eye using location positioning. After the positioning of the eyes, according to the human face geometry, we can locate the corresponding face of the other parts, such as the mouth and nose.