Detecting small objects from UAV-captured aerial imagery is a critical yet challenging task, hindered by factors such as small object size, complex backgrounds, and subtle inter-class differences. Single-modal methods lack the robustness for all-weather operation, while existing multimodal solutions are often too computationally expensive for deployment on resource-constrained UAVs. To this end, we propose TAF-YOLO, a lightweight and efficient multimodal detection framework designed to balance accuracy and efficiency. First, we propose an early fusion module, the Two-branch Adaptive Fusion Network (TAFNet), which adaptively integrates visible and infrared information at both pixel and channel levels before the feature extractor, maximizing complementary data while minimizing redundancy. Second, we propose a Large Adaptive Selective Kernel (LASK) module that dynamically expands the receptive field using multi-scale convolutions and spatial attention, preserving crucial details of small objects during downsampling. Finally, we present an optimized feature neck architecture that replaces PANet’s bidirectional path with a more efficient top-down pathway. This is enhanced by a Dual-Stream Attention Bridge (DSAB) that injects high-level semantics into low-level features, improving localization without significant computational overhead. On the VEDAI benchmark, TAF-YOLO achieves 67.2% mAP₅₀, outperforming the CFT model by 2.7% and demonstrating superior performance against seven other YOLO variants. Our work presents a practical and powerful solution that enables real-time, all-weather object detection on resource-constrained UAVs.

To solve the matching problems caused by the large intensity difference between the multi-source images and the nonlinear radiation distortion, we present a multi-source image matching approach that considers the orientation of the phase sharpness. First, the scale-space of the image pyramid was constructed, and the phase consistency of the image frequency domain was solved to obtain the maximum moment feature, and the KAZE operator is used to extract the feature points. Next, the Log-Gabor even symmetric filter was used to perform Fourier transform, and the improved local phase sharpness feature and phase orientation feature were constructed respectively to replace the gradient amplitude and gradient direction feature of the image. A descriptor of local phase sharpness orientation (LPSO) was established through the log-polar description template, and finally the Euclidean distance was used to measure the similarity to obtain the corresponding points. Multiple sets of typical multi-source images were used as data sources, and the LPSO algorithm was compared with SIFT, LGHD, RIFT and HAPCG algorithms in datasets. Evaluation of the algorithm performances indicates that the proposed method is more accurate and robust in the task of multi-source image matching.

Non-linear radiation distortion (NRD) and image noise in multi-modal remote sensing images (MRSI) lead to abrupt changes in feature directions, resulting in sensitivity to rotational variation, sparse correct matches, and high false match rates. To address these challenges, this paper proposes a second-order tensor orientation feature transformation (SOFT) method to improve the rotational invariance of MRSI matching and increase the number of correct matches. The SOFT method has two main contributions: (1) A novel second-order tensor orientation descriptor is constructed by generating a tensor orientation feature map using a designed second-order tensor function, which is then combined with a GLOH-like descriptor framework to achieve robust rotational invariance in multi-modal image matching. (2) An error-removal global-local iterative optimization is introduced, employing a skewness of mixed pixel intensity (SMPI) function to automatically select matching seed points, followed by an iterative partition optimization strategy for refining corresponding points. Experiments on 744 groups of typical MRSIs demonstrate that the SOFT method significantly outperforms nine state-of-the-art methods, achieving an average 97% improvement in the number of correct matches, an average 25.51% improvement in the rate of correct matches, and an average reduction in RMSE of 2.69 pixels. The proposed SOFT method thus offers robust MRSI matching, with strong rotational invariance and precise identification of corresponding points, proving its effectiveness for complex remote sensing scenarios. Access to experiment-related data and codes will be provided at https://skyearth.org/research/.

Change detection (CD) aims to identify and locate changes in the same observed surface coverage area across bitemporal images. This technique has widespread applications in urban planning, land use, and disaster damage extraction. Deep learning-based CD methods typically use learnable encoders to map bitemporal images to a common domain distribution space, allowing for the discrimination and localization of change and invariant features. However, due to differences in imaging mechanisms, seasons, and shooting angles, a large number of pseudo-changes may easily appear, affecting the accurate recognition of the domain distribution space. Additionally, binary CD focuses solely on whether scene targets have changed, resulting in change labels that encompass a variety of different objects, thus increasing the significance of intra-class differences. To address the aforementioned issues, we propose a domain distribution reconstruction learning (DDRL) framework for binary CD, which effectively mitigates the problem of pseudo-changes by detecting abnormal feature domain distributions. Specifically, DDRL first extracts multi-scale features from bitemporal images using a Siamese cross-window self-attention module, achieving feature domain transformation from the original space. Subsequently, it employs a graph attention enhanced module to improve the low-level domain distribution, enabling it to focus on change regions. Additionally, DDRL utilizes a cross-domain feature contrastive learning module for reconstructive learning of high-level fused features. This process ensures that intra-class features are compact while inter-class features are dispersed within the high-level domain distribution, thereby significantly improving the domain distribution representation to discriminate pseudo-changes. Experimental results show that the proposed DDRL performs excellently across multiple public datasets, surpassing mainstream methods and significantly improving CD performance. The source code will be made available at https://github.com/yzygit1230/DDRL.

Detecting small objects from UAV-captured aerial imagery is a critical yet challenging task, hindered by factors such as small object size, complex backgrounds, and subtle inter-class differences. Single-modal methods lack the robustness for all-weather operation, while existing multimodal solutions are often too computationally expensive for deployment on resource-constrained UAVs. To this end, we propose TAF-YOLO, a lightweight and efficient multimodal detection framework designed to balance accuracy and efficiency. First, an early fusion module, the Two-branch Adaptive Fusion Network (TAFNet), which adaptively integrates visible and infrared information at both pixel and channel levels before the feature extractor, maximizing complementary data while minimizing redundancy. Second, a Large Adaptive Selective Kernel (LASK) module that dynamically expands the receptive field using multi-scale convolutions and spatial attention, preserving crucial details of small objects during downsampling. Finally, an optimized feature neck architecture that replaces PANet's bidirectional path with a more efficient top-down pathway. This is enhanced by a Dual-Stream Attention Bridge (DSAB) that injects high-level semantics into low-level features, improving localization without significant computational overhead. On the VEDAI benchmark, TAF-YOLO achieves 67.2% mAP50, outperforming the CFT model by 2.7% and demonstrating superior performance against seven other YOLO variants. Our work presents a practical and powerful solution that enables real-time, all-weather object detection on resource-constrained UAVs.

Feature matching between high-resolution satellite stereos plays an important role in satellite image orientation. However, images of changed regions, weak-textured regions and occluded regions may generate low-quality matches or even mismatches. Furthermore, matching throughout the entire satellite images often has extremely high time cost. To compute good matching results at low time cost, this paper proposes an image block selection method for high-resolution satellite stereos, which processes feature matching in several optimal blocks instead of the entire images. The core of the method is to formulate the block selection into the optimization of an energy function, and a greedy strategy is designed to compute an approximate solution. The experimental comparisons on various satellite stereos show that the proposed method could achieve similar matching accuracy and much lower time cost when compared with some state-of-the-art satellite image matching methods. Thus, the proposed method is a good compromise between matching accuracy and matching time, which has great potential in large-scale satellite applications.