基于深度学习的红外弱小目标检测算法研究综述  

Survey of infrared dim small target detection algorithm based on deep learning

作　　者：王恒慧 曹东[1] 赵杨 杨阳[1] WANG Heng-hui;CAO Dong;ZHAO Yang;YANG Yang(Computation Aerodynamics Institute of China Aerodynamics Research and Development Center,Mianyang 621000,China)

机构地区：[1]中国空气动力研究与发展中心计算空气动力研究所,四川绵阳621000

出　　处：《激光与红外》2022年第9期1274-1279,共6页Laser & Infrared

摘　　要：目标检测技术是安防监控、预警探测、遥感成像等装备的核心要素,也是当前深度学习研究领域的热点之一。红外探测系统通过被动接收物体发射的红外电磁波进行成像,具备温度灵敏度高、探测距离远、被动探测隐蔽性强等优点,在目标探测领域有广泛的应用。文中从红外弱小目标图像的特点出发,针对基于深度学习的视觉图像目标检测算法进行分类描述,并对深度学习在红外弱小目标检测中的有效手段进行总结,最后对未来的发展趋势做出展望。Target detection technology is the core element of security monitoring, early warning detection, remote sensing imaging and other equipment, and is also one of the hotspots in the field of deep learning.Infrared detection system through passive reception of infrared electromagnetic waves emitted by the object for imaging, with high temperature sensitivity, detection distance, passive detection of concealment and other advantages, in the field of target detection has a wide range of applications.In this paper, from the characteristics of infrared dim small target images, the classification and description of deep learning based on visual image target detection algorithms are classified and described, and the effective means of deep learning in infrared weak small target detection are summarized, and finally an outlook on future development trends is made.

关 键 词：红外弱小目标 目标检测 深度学习 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]