摘要
针对航迹数据量有限条件下,如何精确、快速地实现机型分类识别问题,提出一种基于动力学特征的机型分类识别方法。首先,分析飞机的运动特性,提取动压均值特征,并结合航迹数据中飞行高度、对地速度的均值特征,构建动力学特征数据集;其次,搭建轻量化BP神经网络作为分类器,使用训练数据集对网络进行训练,得到机型分类识别网络模型,实现机型分类识别。实验结果表明:在航迹数据量有限条件下,动力学特征的引入使机型分类识别的平均正确率提高了16.57%;与现有研究方法相比,分类识别所用时间为对比方法的13.3%。
To realize aircraft type classification and recognition accurately and quickly under the condition of limited track data, an aircraft type classification and recognition method based on dynamic characteristics is proposed in this paper. Firstly, the aircraft motion characteristics are analyzed to obtain the mean characteristics of dynamic pressure, which constitute the dynamic feature data set together with the mean characteristics of flight altitude and ground velocity;Secondly, the light weight BP neural network is trained with the training set to obtain the model for aircraft type classification and recognition. The experimental results show that under the condition of limited track data, the introduction of dynamic features improves the average accuracy of aircraft type classification and recognition by 16.57%. Further, the classification and recognition time is 13.3% of that of the existing research methods.
作者
王硕
吴楠
黄洁
王建涛
WANG Shuo;WU Nan;HUANG Jie;WANG Jiantao(Information Engineering University,Zhengzhou 450001,China)
出处
《信息工程大学学报》
2022年第5期530-536,共7页
Journal of Information Engineering University
关键词
特征提取
动压
分类识别
BP神经网络
feature extraction
dynamic pressure
classification recognition
BP neural network