当前位置: X-MOL 学术IEEE Trans. Aerosp. Electron. Sys. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A Generalized Low-Rank Double-Tensor Nuclear Norm Completion Framework for Infrared Small Target Detection
IEEE Transactions on Aerospace and Electronic Systems ( IF 5.1 ) Pub Date : 2022-02-01 , DOI: 10.1109/taes.2022.3147437
Lizhen Deng 1 , Dongyuan Xu 1 , Guoxia Xu 2 , Hu Zhu 3
Affiliation  

Infrared small target detection is a research hotspot in computer vision technology that plays an important role in infrared early warning systems. Specifically, infrared images with strong background clutter and noise pose a challenge to target detection technology. In this article, we propose a method for infrared small target detection based on the double nuclear norm and ring-structural elements over a generalized tensor framework. We use the double nuclear norm instead of the traditional single nuclear norm as the relaxation of the rank function, which solves the problem that the suboptimal solution deviates from the original solution and better approaches the rank minimization. In addition, we use weighted ring structural elements instead of traditional structural elements to make better use of the target information and its surrounding background. Experiments on six sequences of real images show that the proposed method can enhance the target and suppress the background effectively, and ensure a high detection probability and a low false alarm rate.

中文翻译:


一种用于红外小目标检测的广义低阶双张量核范数完成框架



红外小目标检测是计算机视觉技术的研究热点,在红外预警系统中发挥着重要作用。具体来说,背景杂波和噪声较强的红外图像对目标检测技术提出了挑战。在本文中,我们提出了一种基于广义张量框架上的双核范数和环结构元素的红外小目标检测方法。我们使用双核范数代替传统的单核范数作为秩函数的松弛,解决了次优解偏离原解的问题,更好地逼近秩最小化。此外,我们使用加权环形结构元素代替传统的结构元素,以更好地利用目标信息及其周围背景。对六序列真实图像的实验表明,该方法能够有效增强目标并抑制背景,并保证较高的检测概率和较低的误报率。
更新日期:2022-02-01
down
wechat
bug