Multi-Scale Hierarchical Feature Fusion for Infrared Small-Target Detection

Detecting small targets in infrared images presents significant challenges due to their tiny size and complex backgrounds, making this task a hotspot for research. Traditional methods rely on assumption-based modeling and manual design, struggling to handle the variability of real-world scenarios. A...

Full description

Saved in:
Bibliographic Details
Main Authors: Yue Wang, Xinhong Wang, Shi Qiu, Xianghui Chen, Zhaoyan Liu, Chuncheng Zhou, Weiyuan Yao, Hongjia Cheng, Yu Zhang, Feihong Wang, Zhan Shu
Format: Article
Language:English
Published: MDPI AG 2025-01-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/3/428
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850199951716384768
author Yue Wang
Xinhong Wang
Shi Qiu
Xianghui Chen
Zhaoyan Liu
Chuncheng Zhou
Weiyuan Yao
Hongjia Cheng
Yu Zhang
Feihong Wang
Zhan Shu
author_facet Yue Wang
Xinhong Wang
Shi Qiu
Xianghui Chen
Zhaoyan Liu
Chuncheng Zhou
Weiyuan Yao
Hongjia Cheng
Yu Zhang
Feihong Wang
Zhan Shu
author_sort Yue Wang
collection DOAJ
description Detecting small targets in infrared images presents significant challenges due to their tiny size and complex backgrounds, making this task a hotspot for research. Traditional methods rely on assumption-based modeling and manual design, struggling to handle the variability of real-world scenarios. Although convolutional neural networks (CNNs) increase robustness to diverse scenes with a data-driven paradigm, many CNN-based methods are insufficient in capturing fine-grained details necessary for small targets and are less effective during multi-scale feature fusion. To overcome these challenges, we propose the novel Wide-scale Gated Fully Fusion Network (WGFFNet) in this article, which contributes to infrared small-target detection (IRSTD). WGFFNet uses a classic encoder–decoder structure, where the designed stepped fusion block (SFB) embedded in the feature extraction stage captures finer local context across multiple scales during encoding, and along the decoding path, the multi-level features are progressively integrated by a Fully Gated Interaction (FGI) Module to enhance feature representation. The inclusion of a boundary difference loss further optimizes the edge details of targets. We conducted comprehensive experiments on two public infrared small-target datasets: SIRST-V2 and IRSTD-1k. Quantitative and qualitative results demonstrate that our WGFFNet outperforms representative methods when considering various evaluation metrics together, achieving an improved detection performance and computational efficiency for detecting small targets in infrared images.
format Article
id doaj-art-1841f26c770047cc8ca05164a3b1ff2b
institution OA Journals
issn 2072-4292
language English
publishDate 2025-01-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-1841f26c770047cc8ca05164a3b1ff2b2025-08-20T02:12:29ZengMDPI AGRemote Sensing2072-42922025-01-0117342810.3390/rs17030428Multi-Scale Hierarchical Feature Fusion for Infrared Small-Target DetectionYue Wang0Xinhong Wang1Shi Qiu2Xianghui Chen3Zhaoyan Liu4Chuncheng Zhou5Weiyuan Yao6Hongjia Cheng7Yu Zhang8Feihong Wang9Zhan Shu10Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaDetecting small targets in infrared images presents significant challenges due to their tiny size and complex backgrounds, making this task a hotspot for research. Traditional methods rely on assumption-based modeling and manual design, struggling to handle the variability of real-world scenarios. Although convolutional neural networks (CNNs) increase robustness to diverse scenes with a data-driven paradigm, many CNN-based methods are insufficient in capturing fine-grained details necessary for small targets and are less effective during multi-scale feature fusion. To overcome these challenges, we propose the novel Wide-scale Gated Fully Fusion Network (WGFFNet) in this article, which contributes to infrared small-target detection (IRSTD). WGFFNet uses a classic encoder–decoder structure, where the designed stepped fusion block (SFB) embedded in the feature extraction stage captures finer local context across multiple scales during encoding, and along the decoding path, the multi-level features are progressively integrated by a Fully Gated Interaction (FGI) Module to enhance feature representation. The inclusion of a boundary difference loss further optimizes the edge details of targets. We conducted comprehensive experiments on two public infrared small-target datasets: SIRST-V2 and IRSTD-1k. Quantitative and qualitative results demonstrate that our WGFFNet outperforms representative methods when considering various evaluation metrics together, achieving an improved detection performance and computational efficiency for detecting small targets in infrared images.https://www.mdpi.com/2072-4292/17/3/428infrared small-target detectionfeature fusiongate mechanismhierarchical decoding
spellingShingle Yue Wang
Xinhong Wang
Shi Qiu
Xianghui Chen
Zhaoyan Liu
Chuncheng Zhou
Weiyuan Yao
Hongjia Cheng
Yu Zhang
Feihong Wang
Zhan Shu
Multi-Scale Hierarchical Feature Fusion for Infrared Small-Target Detection
Remote Sensing
infrared small-target detection
feature fusion
gate mechanism
hierarchical decoding
title Multi-Scale Hierarchical Feature Fusion for Infrared Small-Target Detection
title_full Multi-Scale Hierarchical Feature Fusion for Infrared Small-Target Detection
title_fullStr Multi-Scale Hierarchical Feature Fusion for Infrared Small-Target Detection
title_full_unstemmed Multi-Scale Hierarchical Feature Fusion for Infrared Small-Target Detection
title_short Multi-Scale Hierarchical Feature Fusion for Infrared Small-Target Detection
title_sort multi scale hierarchical feature fusion for infrared small target detection
topic infrared small-target detection
feature fusion
gate mechanism
hierarchical decoding
url https://www.mdpi.com/2072-4292/17/3/428
work_keys_str_mv AT yuewang multiscalehierarchicalfeaturefusionforinfraredsmalltargetdetection
AT xinhongwang multiscalehierarchicalfeaturefusionforinfraredsmalltargetdetection
AT shiqiu multiscalehierarchicalfeaturefusionforinfraredsmalltargetdetection
AT xianghuichen multiscalehierarchicalfeaturefusionforinfraredsmalltargetdetection
AT zhaoyanliu multiscalehierarchicalfeaturefusionforinfraredsmalltargetdetection
AT chunchengzhou multiscalehierarchicalfeaturefusionforinfraredsmalltargetdetection
AT weiyuanyao multiscalehierarchicalfeaturefusionforinfraredsmalltargetdetection
AT hongjiacheng multiscalehierarchicalfeaturefusionforinfraredsmalltargetdetection
AT yuzhang multiscalehierarchicalfeaturefusionforinfraredsmalltargetdetection
AT feihongwang multiscalehierarchicalfeaturefusionforinfraredsmalltargetdetection
AT zhanshu multiscalehierarchicalfeaturefusionforinfraredsmalltargetdetection