Λεπτομέρειες βιβλιογραφικής εγγραφής
| Τίτλος: |
RShDet: An Adaptive Spectral-Aware Network for Remote Sensing Object Detection Under Haze Corruption. |
| Συγγραφείς: |
Zhang, Wei, Wang, Yuantao, Yang, Haowei, Mao, Xuerui |
| Πηγή: |
Remote Sensing; Apr2026, Vol. 18 Issue 7, p1020, 30p |
| Θεματικοί όροι: |
Haze, Frequency-domain analysis, Machine learning, Image enhancement (Imaging systems), Object recognition (Computer vision), Remote sensing |
| Περίληψη: |
Highlights: What are the main findings? RShDet is a novel frequency-domain adaptive framework for haze-degraded remote sensing object detection that unifies image enhancement and object detection within a spectral-aware learning paradigm. By dynamically modeling object-relevant spectral characteristics and enabling adaptive cross-task interaction, the framework improves detection robustness and reliability under severe haze degradation. RShDet establishes a new state of the art in haze-degraded remote sensing object detection, enabling significantly more reliable object localization under severe atmospheric degradation. These gains are consistently validated across both synthetic and real-world hazy benchmarks, including a +2.16% mAP50 improvement over DHCNet on Hazy-DOTA. What are the implications of the main findings? The findings demonstrate that frequency-domain joint optimization is a highly effective paradigm for haze-robust remote sensing object detection. By enabling adaptive spectral-aware learning and facilitating deeper integration between enhancement and detection processes, this approach provides a principled solution for mitigating atmospheric degradation and improving detection reliability under adverse visual conditions. The proposed Hazy-DOTA dataset, together with the frequency-aware enhancement and cross-task integration framework, establishes a reproducible benchmark and methodological foundation for future research on haze-robust object detection across diverse remote sensing platforms, including satellite and UAV systems. Remote sensing (RS) object detection faces intrinsic challenges arising from the overhead imaging paradigm and the diversity of climatic conditions. In particular, atmospheric phenomena such as clouds and haze cause severe visual degradation, making reliable object detection difficult. However, most existing detectors are developed under clear-weather conditions, which limits their generalization capability in realistic haze-degraded RS scenarios. To alleviate this issue, an adaptive spectral-aware network for RS object detection under haze interference is proposed, termed RShDet, which is designed to handle both high-altitude RS imagery and low-altitude Unmanned Aerial Vehicle (UAV) scenarios. Firstly, the Object-Centered Dynamic Enhancement (OCDE) module dynamically adjusts the spatial positions of key-value pairs through query-agnostic offsets, enabling the network to emphasize object-relevant regions while suppressing haze-induced background interference. Secondly, the Dynamic Multi-Spectral Perception and Filtering (DSPF) module introduces a multi-spectral attention mechanism that adaptively selects informative frequency components, thereby enhancing discriminative feature representations in hazy environments. Thirdly, the Frequency-Domain Multi-Feature Fusion (FDMF) module employs learnable weights to complementarily integrate amplitude and phase information in the frequency domain, enabling effective cross-task feature interaction between the enhancement and detection branches. Extensive experiments demonstrate that RShDet consistently achieves superior detection performance under hazy conditions across both synthetic and real-world benchmarks. Specifically, it achieves improvements of 2.4% mAP50 on Hazy-DOTA, 1.9% mAP on HazyDet, and 2.33% mAP on the real-world foggy dataset RTTS, surpassing existing state-of-the-art methods. [ABSTRACT FROM AUTHOR] |
|
Copyright of Remote Sensing is the property of MDPI and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.) |
| Βάση Δεδομένων: |
Complementary Index |