Ear recognition based on wavelet transform and Discriminative Common Vectors

Discriminative common vectors is one of the most successful methods which overcome the small sample size case in Fisher¿s linear discriminant analysis. But when we directly use DCV to reduce the dimensions of the ear images, the computational expense of training is still relatively large. A new method is proposed in this paper that the low frequency sub-images are obtained by utilizing two-dimensional wavelet transform firstly and the features are extracted by applying DCV to the sub-images. The experimental results show that the proposed Wavelet+DCV method achieve better performance than PCA+LDA method and the computation burden is also reduced greatly.