Improved Kernel Discriminative Common Vectors for Face Recognition
C. Lakshmi, M. Ponnavaikko · 2009
Kernel discriminative common vector is one of the most effective non-linear techniques for feature extraction from high dimensional data including images and text data. This paper presents a new algorithm called "improved kernel discriminative common vector" (IKDCV) to further improves the overall performance of KDCV by integrating the boosting parameters and KDCV techniques. The proposed method possesses several appealing properties. First, like all kernel methods, it handles non-linearity in a disciplined manner. Second by introducing the pair-wise class discriminant information into discriminant criterion, it further increases the classification accuracy. Third, by calculating significant discriminant information, within class scatter space, it also effectively deals with the small sample size problem. Fourth, it constitutes a strong ensemble based KDCV framework by taking advantage of the boosting parameters and KDCV techniques. This new method is applied on extended YaleB face database and achieves better recognition performance by means of solving overlapping between classes. Experimental results demonstrate the promising performance of the proposed method as compared to the other linear and non-linear methods.