摘要
敌我识别在现代战争中是十分关键的一环,有效的敌我识别可以避免攻击己方目标,能够实现更精准地打击。文章以指挥信息系统为背景,针对军事样本数据少的问题,提出了基于度量学习的敌我识别方法,用基于小样本学习分类技术实现在少样本条件下的敌我识别。相比以往基于概率推理的图像分类技术,能够更准确地实现敌我目标的分类。
IFF is a key link in modern war. Effective IFF can avoid attacking our targets and achieve a more accurate attack.Taking the command information system as the background, aiming at the problem of less military sample data, this paper proposes a friend or foe recognition method based on metric learning and uses the classification technology based on small sample learning to realize the friend or foe recognition under the condition of fewer samples. Compared with the previous image classification technology based on probability reasoning, it can more accurately classify the enemy and our targets.
作者
章倩
王梓祺
ZHANG Qian;WANG Ziqi(l.The 28th Research Institute,China Electronic Technology Group Corporation,Naiijing 210007,China;Command and Control Engineering College,Army Engineering University of PLA,Nanjing 210007,China)
出处
《长江信息通信》
2022年第6期120-122,共3页
Changjiang Information & Communications