Temperature-Aware Dynamic Fusion Network for Few-Shot Segmentation of Infrared Images
- 25 September 2025
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-14
Abstract
Infrared remote sensing is indispensable for aerial target detection due to its all-weather thermal radiation sensing capability. Integrating visible light’s high resolution with infrared’s robustness to environmental interference represents an advanced strategy to improve segmentation accuracy. Current challenges of the strategy include: 1) degradation of infrared images due to background noise; 2) cross-modal feature mismatches caused by temperature disparities between thermal and spectral data; 3) insufficient annotated datasets restricting model generalization. To address these challenges, this paper proposes a temperature-aware dynamic fusion network for few-shot infrared image segmentation algorithm. The method significantly improves segmentation accuracy through two key innovations: temperature-guided dual-modal feature fusion and thermodynamic few-shot transfer mechanisms. Specifically, in temperature-guided dual-modal feature fusion, Dual-stream Pyramid Enhancement Module (DPEM) is designed to suppress noise and boost contrast in infrared image. Building upon this, Temperature-guided Dual-modal Encoder is constructed to dynamically fuse visible light and Infrared features. In thermodynamic few-shot transfer, Thermodynamic Adversarial Alignment can obtain dual-modal features with discriminative information in few-shot transfer. Then, Thermodynamic Pseudo-Labeling Enhancement is created to improve pseudo-label quality by Temperature Confidence Propagation. Numerous experimental results performed on a multi-scenario dataset with 12,800 visible-infrared aircraft images have validated the method’s superiority, achieving SOTA performance with minimum gains of 7.07% mPA, 6.48% mF1, and 7.93% mIoU over existing methods like GANFuse and TransFusion. This framework advances explainable infrared visual understanding in complex environments.Keywords
Funding Information
- Key Research and Development Program of Shaanxi (2025SF-YBXM-546)
This publication has 39 references indexed in Scilit:
- Modeling optical imaging pipeline and learning contrastive-based representation for hybrid-corrupted image restorationMultimedia Systems, 2025
- Vision transformers in multi-modal brain tumor MRI segmentation: A reviewMeta-Radiology, 2023
- ResNet 342021
- Learning Gradient Fields for Shape Generation2020
- GANFuse: a novel multi-exposure image fusion method based on generative adversarial networksNeural Computing and Applications, 2020
- Deep semantic segmentation of natural and medical images: a reviewArtificial Intelligence Review, 2020
- Thermal Conductivity of Amorphous MaterialsAdvanced Functional Materials, 2019
- A review: Deep learning for medical image segmentation using multi-modality fusionArray, 2019
- FusionGAN: A generative adversarial network for infrared and visible image fusionInformation Fusion, 2019
- A novel multi-modality image fusion method based on image decomposition and sparse representationInformation Sciences, 2018