The Research And Design Of Multi-pose Face Recognition Algorithm

As a direct and effective biometric identification technology that conforms to human lifestyle,face identification is gradually becoming a mainstream and reliable identity authentication method in the process of technological development. Face recognition technology research has become more and more mature. Many various recognition systems are springing up. In order to improve the performance of the system, a variety of techniques are used. In this paper, based on other face recognition research, I conduct a research of creating a more efficient and useful face recognition system, combining the comparison of different methods,and it is expected to achieve a more practical value in multi-pose images processing and face Recognition.The human face plays an important role in our social interaction, conveying people’s identity. Using the human face as a key to security, biometric face recognition technology has received significant attention in the past several years due to its potential for a wide variety of applications in both law enforcement and non-law enforcement. Face recognition is to identify the biological characteristics of this kind of identification technology based on the original biological characteristics. After the collection of biological characteristics, computer is used to conduct digital image processing, template matching method and the face recognition process. General process of face recognition means face detection, facial feature extraction,and the finally is face recognition. In addition, in order to enhance the accuracy of the recognition system, most of the systems pre-treat the obtained images at first.The structure of this paper mainly based on researching these processes, and several effective methods and comparative description for the analysis. In based on the demand of detection, I pre-treat the color image, establish the skin color distribution model, combine with the face of the basic characteristics and eye location in face detection, and then obtain the face region by the verification. Finally, I adopt PCA algorithm to extract facial features,combined with the improved neural network recognition decision, ultimately I get the results of feedback, and complete the full functionality of the face recognition system.The ideological function of artificial neural networks mainly based on computer, to achieve the corresponding effect for the input parameters and the output results of nonlinearrelationship. For the neural network, image parameter input and output appears to be entirely without a fixed order, and there are no corresponding rules.After analysis of different face recognition methods, we could finally get a more efficient and reliable face recognition system for complex color images. This paper focuses on the research of camera images or static images for noise filtering, light compensation and image enhancement processing and analysis, basing on the combination of face color space of skin color model face detection. Face recognition technology research has become more and more mature and various recognition system. In order to improve the performance of the system,we can use a variety of techniques.