Dual-Stream Edge-Target Learning Network for Infrared Small Target Detection

计算机科学 对偶(语法数字) GSM演进的增强数据速率 遥感 红外线的 人工智能 地质学 物理 光学 文学类 艺术
作者
Rui Yao,Wenxin Li,Yong Zhou,Jinqiu Sun,Zihang Yin,Jiaqi Zhao
出处
期刊:IEEE Transactions on Geoscience and Remote Sensing [Institute of Electrical and Electronics Engineers]
卷期号:62: 1-14 被引量:17
标识
DOI:10.1109/tgrs.2024.3488054
摘要

Infrared small target detection (IRSTD) is crucial in both military and civilian applications. However, challenges such as low contrast, low signal-to-noise ratio (SNR), and lack of shape and texture information limit the effectiveness of existing methods in capturing edge details and representing target areas. To address these issues, we propose the dual-stream edge-target learning network (DETL-Net) for IRSTD. This network enhances feature cross-fusion by learning edge details and target regions through a dual-stream framework, significantly improving detection performance. Specifically, we extract multilevel features of the image based on the encoder-decoder structure of U-Net and then reconstruct the feature map. In the decoder, we propose the dual-guided cross-fusion module (DGCFM) to capture edge details of small targets and global contextual features of the target region, achieving complementary advantages. The multiscale context fusion module (MCFM) within DGCFM uses central difference convolution to enhance local contrast and extract rich contextual details, thereby retaining edge information and enhancing overall target representation. In addition, we introduce the cross-dimension interactive aggregation attention module (CIAAM), which dynamically adjusts feature fusion weights across layers to effectively suppress noise and enhance the discrimination of small targets. These modules are sequentially interconnected to progressively refine edge details, and the acquired target features are subsequently utilized for predicting the final target mask via the segmentation head. Experiments on the NUAA-SIRST and IRSTD-1k datasets demonstrate that DETL-Net outperforms state-of-the-art (SOTA) methods. The source code is available at https://github.com/rayyao/DETL-Net.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
完美世界的应助被SLL采纳,获得10
刚刚
共享精神的应助被开心德地采纳,获得10
1秒前
1秒前
泉竹晓筱发布了新的文献求助10
1秒前
2秒前
风朝东发布了新的文献求助10
2秒前
2秒前
2秒前
4秒前
Bai发布了新的文献求助10
4秒前
Owen的应助被难过的谷芹采纳,获得10
5秒前
5秒前
田様的应助被学术混子采纳,获得10
5秒前
Orange的应助被gkam采纳,获得10
5秒前
5秒前
飏yang发布了新的文献求助10
5秒前
6秒前
小牛发布了新的文献求助10
6秒前
汤姆逊波纹疾走完成签到,获得积分10
6秒前
7秒前
小石头完成签到,获得积分10
7秒前
7秒前
嘻嘻哈哈的应助被123采纳,获得10
7秒前
星辰大海的应助被火星上亦绿采纳,获得10
8秒前
8秒前
8秒前
8秒前
liu完成签到,获得积分10
8秒前
8秒前
9秒前
fairytaler的应助被衡阳采纳,获得10
9秒前
syvshc的应助被开心德地采纳,获得10
9秒前
TinLi完成签到,获得积分10
9秒前
风吹麦浪的应助被何文鑫采纳,获得10
9秒前
9秒前
roselau完成签到,获得积分10
9秒前
思源的应助被尹兴亮采纳,获得10
9秒前
shiqiyuan发布了新的文献求助10
10秒前
11秒前
小蘑菇的应助被lawang采纳,获得10
11秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
八旗制度の研究 600
高分子基功能界面水蒸发材料与应用研究 500
外场增强的水汽发电研究 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8187835
求助须知:如何正确求助?哪些是违规求助? 9575825
关于积分的说明 21843558
捐赠科研通 7785715
什么是DOI,文献DOI怎么找? 3435125
关于科研通互助平台的介绍 2559495
邀请新用户注册赠送积分活动 2482008