Modular FDA and Its Application in Human Face Recognition

Fisher linear discriminating analysis (FDA) is one of accepted and important technique for feature extraction widely used in the areas of images recognition such as human face recognition. Face recognition is essentially a typical small-sample pattern recognition problem in sparse hyper-high dimensional space. The key to solve the problem is how to obtain the significant features for classification. In this paper, a new criterion called Modular Fisher criterion is proposed, which is based on the image segmentation. After an image is segmented into four different regions, the classical Fisher criterion is used in the sub-image. The ORL face image database is made use of to simulate, and when the training sample is only one, the error recognition rate of 9.17 percent is achieved. The experimental results show the validity of the algorithm, which can not only accelerate recognition speed by reducing dimensionality but also reduce the memory capacity.