Infrared and visible image fusion aims to describe the same scene from different aspects by combining complementary information of multi-modality images. The existing Generative adversarial networks (GAN) based infrared and visible image fusion methods cannot perceive the most discriminative regions, and hence fail to highlight the typical parts existing in infrared and visible images. To this end, we integrate multi-scale attention mechanism into both generator and discriminator of GAN to fuse infrared and visible images (AttentionFGAN). The multi-scale attention mechanism aims to not only capture comprehensive spatial information to help generator focus on the foreground target information of infrared image and background detail information of visible image, but also constrain the discriminators focus more on the attention regions rather than the whole input image. The generator of AttentionFGAN consists of two multi-scale attention networks and an image fusion network. Two multi-scale attention networks capture the attention maps of infrared and visible images respectively, so that the fusion network can reconstruct the fused image by paying more attention to the typical regions of source images. Besides, two discriminators are adopted to force the fused result keep more intensity and texture information from infrared and visible image respectively. Moreover, to keep more information of attention region from source images, an attention loss function is designed. Finally, the ablation experiments illustrate the effectiveness of the key parts of our method, and extensive qualitative and quantitative experiments on three public datasets demonstrate the advantages and effectiveness of AttentionFGAN compared with the other state-of-the-art methods.

With the progression of remote sensing technologies, extracting road networks from satellite imagery has emerged as a pivotal research domain in both Geographic Information Systems and Intelligent Transportation Systems. Recognizing the difficulty in balancing lightweight network design with extraction accuracy, the challenge of synergistically preserving global road connectivity and local details, and the hardship in effectively integrating low-level features with high-level representations to achieve full coupling between road details and semantic understanding in road extraction from remote sensing images, this study introduces a Hierarchical Perception Lightweight Network for road extraction (HPLNet). This innovative network integrates shallow perception part and deep perception part, aiming to optimize the trade-off between inference efficiency and extraction accuracy. In shallow perception, directional stripe convolutions capture road details, while deep perception integrates a spatial-channel semantic awareness network to bridge local and global information, boosting road semantic feature extraction. Moreover, to extend the model’s reception at both pixel and semantic levels, each network component strategically introduces parameter-free channel shift operations. HPLNet attains state-of-the-art efficiency in balancing parameter footprint and inference latency: its parameter count is merely 22% of that of U-Net, while its inference speed is 18% faster than FCN. Concurrently, it delivers competitive segmentation metrics on the Massachusetts dataset, achieving an IoU of 64.32% and an F1 score of 79.96%. Experimental results demonstrate that the proposed network achieves superior performance in both segmentation accuracy and model complexity, thereby offering an efficient solution for real-time deployment on edge devices.

The Tibetan Plateau, known as the “Asian Water Tower”, contains thousands of lakes that are sensitive to climate variability and human activities. To investigate their long-term and short-term dynamics, we developed a daily surface-water mapping dataset covering the period from 2000 to 2024 based on MODIS daily reflectance time series (MOD09GQ/MYD09GQ and MOD09GA/MYD09GA). A hybrid methodology combining per-pixel spectral indices, superpixel segmentation, and fusion of Terra and Aqua results was applied, followed by temporal interpolation to produce cloud-free daily water maps. Validation against Landsat classifications and the 30 m global water dataset indicates an overall accuracy of 96.89% and a mean relative error below 9.1%, confirming the robustness of our dataset. Based on this dataset, we analyzed the spatiotemporal evolution of 1293 lakes (no less than 5 km²). Results show that approximately 87.7% of lakes expanded, with the fastest growth reaching +43.18 km²/y, whereas 12.3% shrank, with the largest decrease being −5.91 km²/y. Seasonal patterns reveal that most lakes reach maximum extent in October and minimum extent in January. This study provides a long-term, cloud-free daily water mapping product for the Tibetan Plateau, which can serve as a valuable resource for future research on regional hydrology, ecosystem vulnerability, and climate–water interactions in high-altitude regions.

Illegal drone intrusion near airport runways poses a critical threat to civil aviation safety, creating an urgent need for runway-side vision systems that can detect intruding UAVs early enough for safety warning and collision-risk mitigation. However, the development of such detectors is severely hindered by the scarcity of annotated real-world data in this high-security scenario. To address this bottleneck, we present SynthAirDrone, the first high-fidelity synthetic dataset for UAV intrusion detection in airport runway environments, together with an intelligent data generation framework integrating scene-aware placement and multi-criteria quality assessment. The proposed method uses sky-region segmentation to guide physically plausible drone placement, and combines perspective-aware scaling, Poisson image editing, and a four-dimensional quality scoring system—covering sky overlap, lighting consistency, size plausibility, and edge continuity—to improve visual plausibility and semantic consistency. The resulting dataset comprises 6500 high-quality images, all annotated in YOLO-compatible format. Using the lightweight YOLOv11n model, we show that models trained solely on SynthAirDrone exhibit non-trivial cross-domain transfer to Anti-UAV, while mixed training with limited real data provides the strongest real-world performance under the present setting. Ablation studies further confirm that a quality threshold of $\tau =0.6$ achieves the best trade-off between diversity and fidelity. Overall, this work delivers a reproducible and efficient synthetic data solution for UAV detector development in high-security, data-scarce airport-runway scenarios.

Pansharpening is a crucial topic in remote sensing, and numerous deep learning-based methods have recently been proposed to explore the potential of deep neural networks (DNNs). However, existing approaches are often sensitive to spatial translation errors between high-resolution panchromatic (HRPan) and low-resolution multispectral (LRMS) images, leading to noticeable artifacts in the fused results. To address this issue, we propose an unsupervised pansharpening method that is robust to translation misalignment between HRPan and LRMS inputs. The proposed framework integrates a shift-invariant module to estimate subpixel spatial offsets and a diffusion-based generative model to progressively enhance spatial and spectral details. Moreover, a multi-scale detail injection module is designed to guide the diffusion process with fine-grained structural information. In addition, a carefully formulated loss function is established to preserve the fidelity of fusion results and facilitate the estimation of translation errors. Experiments conducted on the GaoFen-2, GaoFen-1, and WorldView-2 datasets demonstrate that the proposed method achieves superior fusion quality compared with state-of-the-art approaches and effectively suppresses artifacts caused by translation errors.