PASSNet: Perception-Aware Semantic Strengthening Network for Infrared Small Target Detection
IEEE Transactions on Aerospace and Electronic Systems · 2026 · 62 · 11578-11598
DOI: 10.1109/taes.2026.3697158 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Aerospace and Electronic Systems, 2026. Verfasst von Anning Pan und 4 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref). Gefördert von: National Natural Science Foundation of China, Applied Basic Research Key Project of Yunnan.
Abstract (Originalsprache)
Infrared small target detection (IRSTD) is vital in military surveillance, aviation safety, and wildlife monitoring due to its strong anti-interference capability under complex backgrounds. However, the intrinsic properties of small targets, including weak semantics, low contrast, and limited structural details, make accurate detection challenging. Existing deep learning-based methods suffer from unstable spatial perception in shallow layers, semantic dilution in deeper encoder stages, and a lack of high-level semantic guidance during decoding, which collectively lead to inaccurate localization and blurred boundaries. Moreover, most existing IRSTD methods fail to maintain a favorable balance between accuracy and efficiency under heterogeneous resource and frame-rate constraints. To address these issues, we propose the perception-aware semantic strengthening network (PASSNet), an efficient and effective IRSTD method that integrates enhancement mechanisms at three progressive levels: shallow spatial perception, deep layer channel semantics, and decoding reconstruction. Specifically, the spatial perception enhancement module leverages parallel multiscale convolutions with spatial attention to stabilize shallow perception and better distinguish small targets from background clutter in shallow encoder layers. The dynamic semantic aggregation module employs graph-based interchannel modeling to strengthen the deep semantic consistency of the encoder and mitigate feature dilution. Moreover, the saliency-guided decoder module performs nonlinear, saliency-aware channel reweighting to enhance structure reconstruction and reduce edge blurring. Extensive experiments on three widely used public datasets demonstrate that PASSNet achieves superior performance in both detection accuracy and inference speed. In addition, the BSIRST_v2 dataset further validates PASSNet's promising generalization for multitarget detection in real-world scenarios.
Förderung
- National Natural Science Foundation of China
- Applied Basic Research Key Project of Yunnan
Zitieren
APA
Pan, A., Yang, J., Zhang, F., Deng, S., & Yang, Y. (2026). PASSNet: Perception-Aware Semantic Strengthening Network for Infrared Small Target Detection. IEEE Transactions on Aerospace and Electronic Systems, 62, 11578-11598. https://doi.org/10.1109/taes.2026.3697158
Vancouver
Pan A, Yang J, Zhang F, Deng S, Yang Y. PASSNet: Perception-Aware Semantic Strengthening Network for Infrared Small Target Detection. IEEE Transactions on Aerospace and Electronic Systems. 2026;62:11578-11598. doi:10.1109/taes.2026.3697158
BibTeX
@article{Pan2026,
author = {Pan, Anning and Yang, Jing and Zhang, Fuqing and Deng, Shen and Yang, Yang},
title = {PASSNet: Perception-Aware Semantic Strengthening Network for Infrared Small Target Detection},
journal = {IEEE Transactions on Aerospace and Electronic Systems},
year = {2026},
volume = {62},
pages = {11578--11598},
doi = {10.1109/taes.2026.3697158}
}
EUVIMED