Short Outburst Radiator Classification Based on LS-SVM

A short outburst radiator classification algorithm based on LS-SVM (Least-Square Support Vector Machine) is presented in this paper. The theory of feature extracting and LS-SVM are introduced. And then, the classifier used in short outburst radiator is designed with cross-validation and grid-search to obtain training parameters. A multi-classifier with three coding schemes is also designed. Compared with C-SVM, the simulations show that the performance of LS-SVM is better than C-SVM, and the accurate recognition rate is above 90%, which demonstrates that the classifier is feasible and effective.