航空兵器Issue(1)：55-61,7.DOI:10.19297/j.cnki.41-1228/tj.2017.01.009
基于Hu矩与改进PNN的飞机姿态识别算法
An Algorithm of Aircraft Pose Recognition Based on Hu Moments and IPNN
摘要
Abstract
In military fields, to recognize aircraft pose correctly and quickly, a novel method based on Hu moments and improved probabilistic neural network (IPNN) is proposed.For noise-free grayscale images, this method firstly executes binarization-processing, and then calculates Hu moments of the images.The 360° flight pose of aircraft is divided into ree categories according to three views of engineering, namely, the positive view, the side view and the top view.The PNN network is corstructed taking four dimensional Hu moments as the inputs of PNN and the three categories represenfed by three views as the output of PNN.The experiments results show that the average recognition rate关键词
飞机/姿态/概率神经网络/Hu矩/分类Key words
aircraft/pose/probabilistic neural network/Hu moment/classification分类
信息技术与安全科学引用本文复制引用
成杰,李新德..基于Hu矩与改进PNN的飞机姿态识别算法[J].航空兵器,2017,(1):55-61,7.基金项目
国家自然科学基金项目(6508000076) （6508000076）
航空科学基金项目(20140169002) （20140169002）
