Analysis of the Influence of Three Factors on Deep Learning Aircraft Target Classification Method

Although there have been many researches on aircraft target classification and recognition by deep learning methods, the researches mainly focus on the classification and recognition of helicopter, propeller aircraft and jet aircraft, and few consider the impact of specific factors on classification and recognition. In view of this situation, this paper uses deep learning method to realize classification and recognition of three types of aircraft targets through simulation, and on this basis, studies and analyzes the influence degree of signal-to-noise ratio, dwell time and pulse repetition frequency on the classification and recognition accuracy of three types of aircraft targets. The experimental results show that although the recognition accuracy will decrease with the decrease of SNR, dwell time and pulse repetition rate, the results can still meet the needs, and provide a certain theoretical basis for applying deep learning to actual radar classification and recognition of aircraft targets.