156 Small Target Detection in Sea Clutter Based on LS-SVM

Because of sea clutter, small targets detection in sea has a certain difficulties. Many scholars try to solve these problems by neural network. However the defects of neural network restricted its development. Support vector machine shows its superiority in convergence process and the select of hidden nodes is avoided. This paper uses the particle swarm algorithm to optimize the least squares support vector machine, and overcome the experienced dependence when some important parameters are selected. The time needed for convergence is shortened effectively. This method is applied to the targets detection in sea, and the specific process model is given.