SAR Image Recognition Using Synergetic Neural Networks Based on Immune Clonal Programming

A method for SAR image recognition algorithm is proposed, which makes use of the global optimal search ability and the quick local search ability of Immune Clonal Programming (ICP) [1] to obtain the prototype vectors in Synergetic Neural Networks (SNN) [2]. As a result, the recognition performance of SNN is improved. Moreover, a study has been made of multi-class recognition using SNN, a bottleneck problem of SNN, and the strategy of One-Against-One [3] is introduced in this paper. Simulation result shows the recognition accuracy rate of SNN is satisfied.