Modern Defense Technology ›› 2020, Vol. 48 ›› Issue (5): 67-73.DOI: 10.3969/j.issn.1009-086x.2020.05.010

• DETECTION AND TRACKING TECHNOLOGY • Previous Articles     Next Articles

Survey of Small Targets Detection Technology in Optical Imaging

WANG Zhi-hu, SHEN Xiao-qing, GUI Wei-long   

  1. China Satellite Maritime Tracking and Control Department,Jiangsu Jiangyin 214431,China
  • Received:2019-06-02 Revised:2020-04-08 Online:2020-10-20 Published:2021-02-01

光学成像小目标检测技术综述

王志虎, 沈小青, 桂伟龙   

  1. 中国卫星海上测控部,江苏 江阴 214431
  • 作者简介:王志虎(1986-),男,江苏徐州人。工程师,博士,主要研究方向为目标检测、自动化测试。通信地址:214431 江苏省江阴市虹桥北路243号 E-mail:jshuzi@126.com

Abstract: Small targets detection in optical imaging is a difficult problem in the field of computer vision.At the same time,with the expansion of imaging distance and the requirements of long range strikes and rapid attack/defense in modern warfare,there will be a lot of realistic needs for small targets detection.Therefore,how to quickly and accurately detect small targets in complex backgrounds is of great significance in areas such as intelligent control,visual navigation,and target monitoring.Reviews the research situation of infrared small targets detection,visible small targets detection and small targets detection based on deep learning in recent years at home and abroad,in order to seek new exploration and breakthrough,as well as provide useful reference for the development of small targets detection technology.

Key words: small target detection, infrared image, visible image, image enhancement, transform domain, deep learning

摘要: 光学成像中的小目标检测是计算机视觉领域的研究难点。同时,伴随成像距离拓展以及现代战争远程打击、快速攻防的要求,必将产生大量小目标检测的现实需求。因此,如何在复杂背景中快速、准确地检测小目标,对智能控制、视觉导航以及目标监测等领域都具有重要意义。从红外小目标检测、可见光小目标检测、基于深度学习的小目标检测3个方面综述了近年来国内外相关领域的研究情况,以期在此基础上寻求新的探索和突破,为小目标检测技术的发展提供有益参考。

关键词: 小目标检测, 红外图像, 可见光图像, 图像增强, 变换域, 深度学习

CLC Number: