基于深度学习算法的坦克装甲目标自动检测与跟踪系统  

Automatic detection and tracking system of tank armored targets based on deep learning algorithm

作　　者：王全东 常天庆 张雷 戴文君 WANG Quandong;CHANG Tianqing;ZHANG Lei;DAI Wenjun(Departmem of Weapon and Control,Army Academy of Armored Forces,Beijing 100072,China)

机构地区：[1]陆军装甲兵学院兵器与控制系,北京100072

出　　处：《系统工程与电子技术》2018年第9期2143-2156,共14页Systems Engineering and Electronics

基　　金：军队院校科技创新工程项目(ZYX14060014)资助课题

摘　　要：实现对目标的自动检测与跟踪是坦克火控系统未来发展的重要方向。首先采用迁移学习的方法将基于深度学习模型的Faster R-卷积神经网络(faster R-convolution neural network,Faster R-CNN)算法应用解决复杂背景下的坦克装甲目标检测问题,与基于人工模型的传统算法相比达到了较高的检测精度。其次,针对坦克火控系统现有目标跟踪算法的不足,通过将Faster R-CNN算法与现有跟踪算法相结合,提出了复合式目标跟踪算法,实现了对坦克装甲目标的自动检测与稳定跟踪。最后,设计了一套目标自动检测与跟踪系统,采用动态扫描凝视成像技术实现了对大范围战场图像的快速、清晰获取,并对所提算法进行了实验测试。同时也指出了深度学习方法在应用于坦克火控系统时仍然存在的部分问题。Realizing automatic detection and tracking of targets is an important development direction of tank fire control system in the future.Firstly,with the method of transfer learning,we apply faster R-convolution neural network（Faster R-CNN）algorithm based on deep learning model to solve the detection problems of tank armored targets under complex background,and we achieve a higher detection accuracy compared with the traditional algorithm.Secondly,aiming at the shortcomings of existing tracking algorithms in the tank fire control system,we propose a composite tracking algorithm by combining Faster R-CNN with existing tracking algorithms to achieve automatic detection and stable tracking of the tank armored targets.Finally,we design an automatic detection and tracking system which uses dynamic scanning and staring imaging technique to realize fast and clear acquisition of battlefield images in a large field of view,and the algorithm of this paper is tested.Moreover,this paper also points out some problems that need to be solved when the deep learning algorithm is applied to the tank fire control system.

关 键 词：目标检测与跟踪 卷积神经网络 坦克火控系统 相关跟踪算法 快速反射镜 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]