Target type recognition algorithm for SAR image based on multi-feature fusion classifier of KPFD
Yingying Kong, Weiyang Chen, Henry Leung · 2015
Due to the presence of speckle, the target recognition algorithm of SAR image is different from other algorithms. There exists nuances in the detail of type recognition. This paper proposes Multi-feature fusion classifier of KPFD. Based on data from MSTAR database, the results of experiment show the new algorithm is more effective than other 5 kinds of recognition algorithm and recently recognition algorithm. In addition, when the KPFD recognition algorithm is combined with feature fusion classifier in the decision level and measure level, the feature fusion classifier brings good performance on the identification of the types of tank by using Naive Bayesian Classification algorithm(NBC). The recognition rate is up to 87%.