Face Recognition Of The Global And Local Feature Fusion Based On PCA

Face recognition technology is a most challenging and hot subject in the field of pattern recognition and vision of the machine .Up to now, there have many methods recognition of face been proposed, experienced and even put into practical application. However, it can be found out that the methods of recognition each have advantages and minuses. How to full play the advantage and avoid the shortcoming of the methods is our goal. Data fusion inspired by the people in recent years, this article is from this point of view; put forward a new method of face recognition.In this article, feature extraction is studied. First, using wavelet transform and Gabor transform extraction global and local feature. Second, PCA carried out through the integration of features. Certify this method by some experiment. The main research work and innovation are as follows:1. In this article. based on the impact of local face features on face recognition—"the eyebrow and the eye" is selected as the best representation of the local feature. So, we choose the "the eyebrow and the eye" as the local area to identify local feature. This is not only enhanced the capacity of local feature of symptoms, but also reduce the amount of calculation.2. Use different methods extracting the feature according to the characteristics of the different feature. Using wavelet transform and Gabor transform extraction global and local feature. In this way, this extract is able to reflect the characteristics of the face. Face recognition can be made to identify good results.3. Using PCA combination of the best representation local feature and global feature. In this way, taking into account the global and local information. So that the parameter as far as possible represent the face. Enhance the ability of face classification.Finally, sum up the thesis, analyze the current research in the need to be further improved, pointed out the direction for future work.