Ziyi Chen

Efficient fusion of infrared and visible images is of critical importance for real-time applications such as autonomous driving. While deep learning-based fusion methods have demonstrated significant improvements in fusion quality in recent years, current network architectures still exhibit unsatisfactory computational complexity and processing speed. To reduce computational complexity and improve fusion efficiency, mask-based methods or approaches driven by downstream tasks often prioritize key regions. However, such methods tend to overemphasize target objects, potentially overlooking contextually significant elements. To address this limitation and achieve more effective fusion, we propose BEFuse, a decoupled two-stage training strategy with an end-to-end inference framework. BEFuse extracts shallow features and gradient information from images in the first stage via a cross-modal image segmentation subnetwork. In the fusion stage, we use the Hadamard product to map features to an implicit quadratic feature space, combining feature similarity and gradient mask information, allowing automatic adjustment of loss weights and improving fusion accuracy. Experiments on four datasets (MSRS, TNO, RoadScene, and M3FD) show that BEFuse outperforms existing methods in both fusion quality and computational speed.

Remote sensing imagery (RSI) often involves complex backgrounds and multiscale targets, while variations such as rotation and scaling significantly degrade the performance of existing object detection algorithms and hinder effective modeling of spatial relationships between objects. To address these challenges, we propose SFE-CapsNet. First, we fuse scalar features extracted by convolutional neural networks (CNNs) with vector features from capsule networks through structural reorganization to form virtual capsules, approximating the dynamic routing process via a single fully connected layer. This design preserves object pose and texture information while streamlining information flow. Second, we introduce a capsule attention module (CAM) that generates attention masks to dynamically enhance target-relevant features and suppress background noise, strengthening multilevel feature representations. Integrated with a feature pyramid network (FPN) architecture, our approach achieves precise detection of targets at varying scales. Experimental results demonstrate that multilevel feature fusion and the capsule attention mechanism significantly improve detection accuracy and robustness, achieving 77.65% mean average precision (mAP) on the DOTA dataset and 97.63% mAP on HRSC2016, highlighting its effectiveness and efficiency in complex scenes.

Cross-view geo-localization (CVGL) aims to match images of the same location captured from different viewpoints, such as those captured by Unmanned Aerial Vehicles (UAVs) and satellite platforms. The task is particularly challenging due to significant variations in scale, viewpoint, and illumination. Most existing methods employ symmetric sampling strategy to construct drone–satellite image pairs for deep metric learning, but neglect the potential of incorporating multi-view drone images to enhance the viewpoint robustness of features. To address this, we propose leveraging multi-view images to learn Domain-Invariant Discriminative Embeddings (DIDE) for CVGL. DIDE introduces an Inter-view Feature Aggregation Module (IFAM), which dynamically integrates multi-view drone information into robust embeddings. These are used in contrastive learning with satellite embeddings within batches to learn view-invariant discriminative features, while representation learning further improves scene discrimination across batches. To reduce the domain gap, DIDE constructs and aligns drone and satellite prototypes for effective cross-domain feature alignment. Furthermore, we adopt a parameter-efficient transfer learning strategy that leverages the capabilities of pre-trained foundation models while fine-tuning only dual adapters, significantly reducing the trainable parameters. DIDE achieves the state-of-the-art on University-1652 and University-160k, competitive results on SUES-200, and demonstrates strong cross-dataset transferability, with fewer training parameters and lower computational cost.

The spatial resolution of remotely sensed images has seen significant improvements; higher resolution facilitates the understanding of remote sensing images and improves the accuracy of building footprint extraction. Thus, this paper introduces a novel approach that combines super‐resolution reconstruction with semantic segmentation deep models for building footprint extraction. To begin, a super‐resolution reconstruction deep model is employed to enhance the resolution of the target images. Subsequently, training samples are generated using various data augmentation techniques based on the reconstructed images. Finally, a semantic segmentation deep model is trained to perform the final building footprint extraction. For validation, our proposed framework was tested on two publicly available aerial image datasets: the WHU building dataset and the Massachusetts building dataset. The results obtained from these datasets demonstrated the effectiveness of our method, achieving high Intersection over Union (IoU) scores of 90.57% and 75.02%, respectively. Furthermore, our approach outperformed seven other competing methods, showcasing its superior performance.

The large scale and extensive coverage of point cloud data make large-scale airborne laser scanning(ALS) point cloud semantic segmentation a highly challenging task. Although Transformers have shown impressive performance in large-scale point cloud semantic segmentation task, their quadratic complexity limits the processing capacity. To alleviate this issue, we propose Local and Global Mamba (LGMamba)—a novel State Space Model (SSM)-based network for large-scale point cloud semantic segmentation. Specifically, we propose Local Mamba module to extract fine-grained local features by effectively capturing local dependencies. Then, we propose Global Mamba module to refine the learned local features by capturing the global long distance dependencies of whole scenes. The validation of our method on the DALES datasets was conducted. Extensive experimental results demonstrate the effectiveness of LGMamba, with mean Intersection over Union (mIoU) of 82.3% and overall accuracy (OA) of 97.7% on DALES.

Earth observation satellites can capture optical images under different temporal, climatic conditions, and platforms exhibit substantial differences in color and brightness, leading to poor visual experiences when synthesizing large-area optical satellite images. The related issue of color balancing has attracted considerable attention from researchers, yet challenges such as a lack of research data and sensitivity to model parameters persist. To address these problems, this article publishes a publicly open dataset and presents a semantic segmentation-enhanced color balance network (SECBNet). First, to mitigate the scarcity of research data, we develop a publicly available remote sensing image color balance dataset, Zhu Hai color balance image (ZHCBI), to support related research activities. Second, to improve semantic consistency between the color-balanced images and the target images, we design a dual-branch U-Net architecture guided by segmentation results and propose a novel segmentation feature loss function. Finally, to address issues of seams and unnatural transitions between blocks in segmented processing, we introduce a postprocessing module based on weighted averaging. We conducted comparative experiments and analyses with existing mainstream color balancing algorithms on the ZHCBI dataset. The results demonstrate that our proposed method achieves state-of-the-art color balancing quality, with significant improvement in visual effects and a higher peak signal-to-noise ratio (PSNR) (23.64 dB) compared with other mainstream methods.