Foley-Sammon Optimal Discriminant Vectors Using Kernel Approach
IEEE Transactions on Neural Networks, Vol. 16, No.1, pp. 1-9, 2005，-0001，（）：
A new nonlinear feature extraction method called kernel Foley-Sammon optimal discriminant vectors (KFSODV) is presented in this paper. This new method extends the well-known Foley-Sammon optimal discriminant vectors (FSODV) method from linear domain to a nonlinear domain via the kernel trick that has been used in support vector machine (SVM) and other commonly used kernel based methods. The proposed method also provides an effective technique to solve the so-called "small sample size" (SSS) problem which exists in many classification problems such as face recognition. We give the derivation of KFSODV and the comparison to other commonly used kernel based methods. The experimental results on both simulated and real data confirm that the KFSODV method is superior to other commonly used kernel based methods in terms of the performance of discrimination.
版权说明：以下全部内容由邹采荣上传于 2005年05月18日 00时31分16秒，版权归本人所有。