Detection-driven exposure-correction network for nighttime drone-view object detection.

Xi, Yue; Jia, Wenjing; Miao, Qiguang; Feng, Junmei; Ren, Jinchang; Luo, Heng

Abstract

Drone-view object detection (DroneDet) models typically suffer a significant performance drop when applied to nighttime scenes. Existing solutions attempt to employ an exposure-adjustment module to reveal objects hidden in dark regions before detection. However, most exposure-adjustment models are only optimized for human perception, where the exposure-adjusted images may not necessarily enhance recognition. To tackle this issue, we propose a novel Detection-driven Exposure-Correction network for nighttime DroneDet, called DEDet. The DEDet conducts adaptive, non-linear adjustment of pixel values in a spatially fine-grained manner to generate DroneDet-friendly images. Specifically, we develop a Fine-grained Parameter Predictor (FPP) to estimate pixel-wise parameter maps of the image filters. These filters, along with the estimated parameters, are used to adjust pixel values of the low-light image based on non-uniform illuminations in drone-captured images. In order to learn the non-linear transformation from the original nighttime images to their DroneDet-friendly counterparts, we propose a Progressive Filtering module that applies recursive filters to iteratively refine the exposed image. Furthermore, to evaluate the performance of the proposed DEDet, we have built a dataset NightDrone to address the scarcity of the datasets specifically tailored for this purpose. Extensive experiments conducted on four nighttime datasets show that DEDet achieves a superior accuracy compared with the state-of-the-art methods. Furthermore, ablation studies and visualizations demonstrate the validity and interpretability of our approach. Our NightDrone dataset can be downloaded from https://github.com/yuexiemail/NightDrone-Dataset.

Citation

XI, Y., JIA, W., MIAO, Q., FENG, J., REN, J. and LUO, H. 2024. Detection-driven exposure-correction network for nighttime drone-view object detection. IEEE transactions on geoscience and remote sensing [online], 62, article number 5605014. Available from: https://doi.org/10.1109/TGRS.2024.3351134

• Journal Article Type: Article
• Acceptance Date: Dec 30, 2023
• Online Publication Date: Jan 8, 2024
• Publication Date: Dec 31, 2024
• Deposit Date: Apr 16, 2024
• Publicly Available Date: Apr 16, 2024
• Journal: IEEE transactions on geoscience and remote sensing
• Print ISSN: 0196-2892
• Electronic ISSN: 1558-0644
• Publisher: Institute of Electrical and Electronics Engineers (IEEE)
• Peer Reviewed: Peer Reviewed
• Volume: 62
• Article Number: 5605014
• DOI: https://doi.org/10.1109/tgrs.2024.3351134
• Keywords: Adverse illumination conditions; Differentiable image filters; Drone-view object detection (DroneDet); Exposure correction
• Public URL: https://rgu-repository.worktribe.com/output/2204926