红外图像的目标检测研究

打开文本图片集

摘要：红外搜索与跟踪系统被广泛应用于现代国防中针对远距离目标的检测、跟踪及预警。但是，该统检测的目标一般为尺寸较小且模糊的点状结构通常难以辨别，一种基于图像中目标与背景局部结构的差异性，采用NRAM模型实现的方法可以对单帧红外图像中的小目标进行有效检测。

关键词：红外小目标检测；NRAM模型；稀疏性；低秩性

doi：10.3969/J.ISSN.1672-7274.2024.01.012

中图分类号：TN 215            文献标志码：A            文章编码：1672-7274（2024）01-00-04

Research on Target Detection of Infrared Image

YU Zhiqi

（The University of Hong Kong， Hong Kong 999077， China）

Abstract： Infrared search and tracking system is widely used in modern national defense for long-range target detection， tracking and early warning. However， the target detected by the system is generally a small and blurred dot structure， which is usually difficult to identify. A method based on the difference between the local structure of the target and the background in the image，which is implemented by the NRAM model， can effectively detect small targets in a single frame infrared image.

Key words： infrared small target detection; NRAM model; sparsity; low-rank

1   研究背景

红外搜索与跟踪系统被广泛应用于针对飞机、船舶等远距离目标的检测、跟踪及预警。（剩余5457字）