Abstract: In order to improve discriminant ability of the discriminative common vectors(DCV) algorithm,a novel kernel discriminative common vectors(KDCV) algorithm is developed.The kernel function is used firstly to project the original samples into an implicit space called feature space by nonlinear kernel mapping,then the equivalent solution model of the KDCV algorithm are established by the theory of reproducing kernel in the feature space.Finally,according to the new solution model of the KDCV algorithm,the discriminant vectors in the null space of the kernel within-scatter matrix are extracted by performing the Gram-Schmidt orthogonalization twice.Experimental results on face database demonstrate the effectiveness of the proposed method.