A Ship Recognition Method Based on Affinity Propagation
Weiya Guo, Xuezhi Xia · 2013
Accurate target classification is the keystone of the ship targets recognition in sea battlefields. Aiming at the deficiencies of supervised and unsupervised classified methods, we present a novel scheme called semi-supervised ship target recognition based on affinity propagation (AP). In order to circumvent the problem of choosing initial points, the method introduces affinity propagation clustering to construct classification model simply and effectively. Based on the idea of semi-supervised learning, a few restrictions of labeled flows and priori manifold distribution of sampled space are abstracted. Also, manifold similarity is defined. Hence, the semi-supervised method can not only largely reduce the complexity of marking sampled flows, but also nicely improve the performance of the classified. Theoretical analysis and experimental results show that that the proposed method is robust and can get better than KNN or SVM or HDR method. With the acquirement of high recognition rate of ship targets in the sea battlefields, undoubtedly, this approach is a feasible and efficient method.