Comparative Study of Radial Basis Function Neural Network with Estimation of Eigenvalue in Image Using MATLAB
Radial Basis Functions (RBFs) are very important in neural network. In this paper various Radial Basis Functions of neural network such as Generalized Inverse Multi Quadratic, Generalized Multi Quadratic and Gaussian are compared with matrix of images. Mathematical calculation, comparative study and simulation of Eigen value of matrix show that Gaussian RBF performs better result and gives lesser error compared to the other Radial Basis Functions of neutral network.
KeywordsGaussian Generalized-multi-quadratic Generalized-inverse-multi-quadratic Neural network Radial basis function
The authors are so grateful to the anonymous referee for a careful checking of the details and for helpful comments and suggestions that improve this paper.
- 1.Schölkopf, B., Sung, K.-K., Burges, C.J.C., Girosi, F., Niyogi, P., Poggio, T., Vapnik, V.: Comparing support vector machines with Gaussian kernels to radial basis function classifiers. IEEE Trans. Signal Process. 45, 2758–2765 (1997)Google Scholar
- 3.Luo, F.L., Li, Y.D.: Real-time computation of the eigenvector corresponding to the smallest eigen value of a positive definite matrix. IEEE Trans. Circuits Syst. 1(41), 550–553 (1994)Google Scholar
- 8.Math works.: MATLAB 7.6.0 (R2008a) (2008)Google Scholar