LDSDet: Long-Range Context and Dynamic Cross-Modal Alignment for Multimodal Object Detection Under Challenging Illumination.
Saved in:
| Title: | LDSDet: Long-Range Context and Dynamic Cross-Modal Alignment for Multimodal Object Detection Under Challenging Illumination. |
|---|---|
| Authors: | Sun, Shijun1 (AUTHOR), Ma, Shuai1 (AUTHOR), Feng, Xuyang1 (AUTHOR), Sun, Chen1 (AUTHOR), Ding, Baolong1 (AUTHOR), Ran, Yaoyao1 (AUTHOR), Zhang, Yihong1 (AUTHOR) zhangyh@dhu.edu.cn |
| Source: | Remote Sensing. Jun2026, Vol. 18 Issue 11, p1827. 28p. |
| Subjects: | Object recognition (Computer vision), Aerial surveillance, Drone photography, Lighting |
| Abstract: | Highlights: What are the main findings? We propose LDSDet, a unified RGB–IR multimodal object detection framework for UAV perception under challenging illumination conditions, including low-light, nighttime, and uneven lighting. LDSDet integrates three complementary modules: LARC for long-range contextual enhancement, DACF for adaptive cross-modal alignment and interaction, and SSG for suppressing redundant fusion responses. Extensive experiments on DroneVehicle, FLIR-Aligned, and LLVIP demonstrate the effectiveness of LDSDet, which achieves 85.2% mAP50 on DroneVehicle, 45.3 % mAP on FLIR-Aligned, and 67.1 % mAP on LLVIP, showing strong robustness under both daytime and nighttime conditions. What is the implication of the main finding? The proposed framework improves detection robustness under complex UAV imaging conditions, especially in low-light, nighttime, and illumination-imbalanced scenarios. By combining context enhancement, adaptive alignment, and redundancy suppression, LDSDet provides an effective paradigm for reliable RGB–IR multimodal fusion in aerial object detection. The competitive performance and real-time efficiency of LDSDet, particularly its consistent gains under low-light and nighttime conditions, indicate its potential for practical deployment in UAV remote sensing applications, such as urban surveillance, intelligent transportation monitoring, and safety inspection. In the field of remote sensing applications, multimodal object detection has emerged as an important technique for enhancing perception robustness in UAV-based scenarios. Nevertheless, RGB–IR UAV detection remains difficult: Degraded illumination destabilizes shallow representations and weakens local discriminative cues, while spatial inconsistencies and fluctuating modality reliability further hinder cross-modal interaction. In addition, existing methods, which often depend on global illumination estimation or simplistic fusion schemes, struggle to jointly maintain contextual stability, reliable cross-modal interaction, and compact discriminative representations in complex aerial scenes. To address these issues, this paper proposes LDSDet, an RGB–IR multimodal UAV object detector for challenging illumination conditions. Specifically, LDSDet integrates three complementary modules: a Long-range Aware Residual Convolution (LARC) module that enhances contextual perception and stabilizes shallow features; a Dynamic Attention-based Cross-modal Fusion (DACF) block that performs spatially adaptive RGB–IR interaction; and a lightweight SeqShuffleGate (SSG) module that suppresses redundant fusion responses to yield compact and discriminative multimodal representations. Extensive experiments on DroneVehicle, FLIR-Aligned, and LLVIP demonstrate the effectiveness of LDSDet, which achieves 85.2% mAP 50 , 45.3% mAP, and 67.1% mAP, respectively, showing strong robustness under day–night alternation, low-light environments, and complex illumination variations. [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.) | |
| Database: | Engineering Source |
|
Full text is not displayed to guests.
Login for full access.
|
|
| Abstract: | Highlights: What are the main findings? We propose LDSDet, a unified RGB–IR multimodal object detection framework for UAV perception under challenging illumination conditions, including low-light, nighttime, and uneven lighting. LDSDet integrates three complementary modules: LARC for long-range contextual enhancement, DACF for adaptive cross-modal alignment and interaction, and SSG for suppressing redundant fusion responses. Extensive experiments on DroneVehicle, FLIR-Aligned, and LLVIP demonstrate the effectiveness of LDSDet, which achieves 85.2% mAP50 on DroneVehicle, 45.3 % mAP on FLIR-Aligned, and 67.1 % mAP on LLVIP, showing strong robustness under both daytime and nighttime conditions. What is the implication of the main finding? The proposed framework improves detection robustness under complex UAV imaging conditions, especially in low-light, nighttime, and illumination-imbalanced scenarios. By combining context enhancement, adaptive alignment, and redundancy suppression, LDSDet provides an effective paradigm for reliable RGB–IR multimodal fusion in aerial object detection. The competitive performance and real-time efficiency of LDSDet, particularly its consistent gains under low-light and nighttime conditions, indicate its potential for practical deployment in UAV remote sensing applications, such as urban surveillance, intelligent transportation monitoring, and safety inspection. In the field of remote sensing applications, multimodal object detection has emerged as an important technique for enhancing perception robustness in UAV-based scenarios. Nevertheless, RGB–IR UAV detection remains difficult: Degraded illumination destabilizes shallow representations and weakens local discriminative cues, while spatial inconsistencies and fluctuating modality reliability further hinder cross-modal interaction. In addition, existing methods, which often depend on global illumination estimation or simplistic fusion schemes, struggle to jointly maintain contextual stability, reliable cross-modal interaction, and compact discriminative representations in complex aerial scenes. To address these issues, this paper proposes LDSDet, an RGB–IR multimodal UAV object detector for challenging illumination conditions. Specifically, LDSDet integrates three complementary modules: a Long-range Aware Residual Convolution (LARC) module that enhances contextual perception and stabilizes shallow features; a Dynamic Attention-based Cross-modal Fusion (DACF) block that performs spatially adaptive RGB–IR interaction; and a lightweight SeqShuffleGate (SSG) module that suppresses redundant fusion responses to yield compact and discriminative multimodal representations. Extensive experiments on DroneVehicle, FLIR-Aligned, and LLVIP demonstrate the effectiveness of LDSDet, which achieves 85.2% mAP 50 , 45.3% mAP, and 67.1% mAP, respectively, showing strong robustness under day–night alternation, low-light environments, and complex illumination variations. [ABSTRACT FROM AUTHOR] |
|---|---|
| ISSN: | 20724292 |
| DOI: | 10.3390/rs18111827 |
