A Simulation Method of Generating Multi-view Image of Aircraft Targets

Abstract

Abstract. Aircraft target detection in airport remote sensing image by deep learning is a hot topic of current research, however, training deep network model needs a large amount of datasets, and there is no website or the specialized agencies to collect remote sensing images of airport. To solve the problem, a multi-view images generation method for aircraft targets is proposed in this paper. Firstly, the aircraft viewpoint transformation detection imaging model was established, and the coordinate transformation is deduced in detail; secondly, a multi-view image simulation algorithm structure was designed, and the realization of the algorithm is introduced in detail. Experimental results show that the method has good simulation effect on remote sensing images of airport and have high use value.