Drone object detection serves as a fundamental task for more advanced applications. However, drone images typically exhibit challenges such as small object sizes, dense distributions, and high levels of overlap. Traditional object detection networks struggle to achieve the required accuracy and efficiency under these conditions. In this paper, we propose DAU-YOLO, a novel object detection method tailored for drone imagery, built upon the YOLOv11 framework. To enhance feature extraction, a Receptive-Field Attention (RFA) module is introduced in the backbone, allowing adaptive convolution kernel adjustments across different local regions, thereby addressing the challenge of dense object distributions. In the neck, we propose a Dynamic Attention and Upsampling (DAU) module, which incorporates additional low-level features rich in small-object information. Furthermore, Scale-Diffusion Attention and Task-Aware Attention are employed to refine these features, significantly improving the network’s ability to detect small objects. To maintain an extremely lightweight architecture, the bottom-most Bottom–Up layer is removed, reducing model complexity without compromising detection accuracy. In the experiments, the proposed method achieves state-of-the-art (SOTA) performance on the VisDrone2019 dataset. On the validation set, DAU-YOLO(l) attains an $mA{P}_{50}$ of 56.1%, surpassing the baseline YOLOv11(l) by 9.1% and the latest similar-structure method Drone-YOLO(l) by 4.8%, while maintaining only 28.9M parameters, almost half those of Drone-YOLO(l). In the discussion, we provide a detailed analysis of the improvements in small object detection as well as the trade-off between detection accuracy and inference speed. These results demonstrate the effectiveness of DAU-YOLO in addressing the challenges of drone object detection, offering a highly accurate and lightweight solution for real-time applications in complex aerial scenes.

The degree of tree curvature exerts a significant influence on the utilization of forestry resources. This study proposes an enhanced quantitative structural modeling (QSM) method, founded upon terrestrial laser scanning (TLS) point cloud data, for the precise extraction of 3D curvature characteristics of tree trunks. The conventional approach operates under the assumption that the tree trunk constitutes an upright rotating body, thereby disregarding the tree trunk’s true curvature morphology. The proposed method is founded on the classical QSM algorithm and introduces two zoom factors that can dynamically adjust the fitting parameters. This improvement leads to enhanced accuracy in the representation of tree trunk curvature and reduced computational complexity. The study utilized 146 sample trees from 13 plots in Jixi, Anhui Province, which were collected and pre-processed by TLS. The study combines point cloud segmentation, manual labeling of actual curvature and dual-factor experiments, and uses quadratic polynomials and simulated annealing algorithms to determine the optimal model factors. The validation results demonstrate that the enhanced method exhibits a greater degree of concordance between the predicted and actual curvature values within the validation set. In the regression equation, the coefficient of the two-factor method for fitting a straight line is 0.95, which is substantially higher than the 0.75 of the one-factor method. Furthermore, the two-factor model has an R² of 0.21, indicating that the two-factor optimization method generates a significantly smaller error compared to the one-factor model (with an R² of 0.12). In addition, this study discusses the possible reasons for the error in the results, as well as the shortcomings and outlook. The experimental results demonstrate the augmented method’s capacity to accurately reconstruct the 3D curvature of tree trunks in most cases. This study provides an efficient and accurate method for conducting fine-grained forest resource measurements and tree bending studies.

Remote sensing image segmentation is particularly difficult due to the coexistence of large-scale variations and fine-grained structures in very high-resolution imagery. Conventional CNN-based or transformer-based networks often struggle to capture global context while preserving boundary details, leading to degraded performance on small or thin objects. To address these challenges, we propose a self-prompt calibration network based on segment anything model 2 (SC-SAM). The SC-SAM achieves self-prompt by feeding mask prompts from a lightweight decoder into frozen prompt encoder. Output calibration is achieved through the proposed cross-probability guided calibration (CPGC) module, which employs cross-probability uncertainty as complementary guidance to refine final predictions via self-prompted outputs. Furthermore, to better preserve contextual and structural information across multiple scales, a scale-decoupled kernel mixture (SDKM) module is designed. Experimental results on the ISPRS Vaihingen and Potsdam dataset demonstrate that the proposed approach surpasses the state-of-the-art methods by 1.02% and 1.34% in mIoU, highlighting its effectiveness. This study provides new insights into adapting SAM for domain-specific remote sensing segmentation tasks.

The semantic segmentation of remote sensing images is vital for accurate surface monitoring and environmental assessment. Multimodal remote sensing images (RSIs) provide a more comprehensive dimension of information, enabling faster and more scientific decision-making. However, existing methods primarily focus on modality and spectral channels when utilizing spectral features, with limited consideration of their association to ground object types. This association, commonly referred to as the spectral characteristics of ground objects (SCGO), results in distinct spectral responses across different modalities and holds significant potential for improving the segmentation accuracy of multimodal RSIs. Meanwhile, the inclusion of redundant features in the fusion process can also interfere with model performance. To address these problems, a ground object spectral feature awareness network (SFANet) specifically designed for RSIs that effectively leverages spectral features by incorporating the SCGO is proposed. SFANet includes two innovative modules: (1) the Spectral Aware Feature Fusion module, which integrates multimodal features in the encoder based on SCGO, and (2) the Adaptive Spectral Enhancement module, which reduces the confusion from redundant information in the decoder. SFANet significantly improves the mIoU by 5.66% and 4.76% compared to the baseline on two datasets, outperforming existing multimodal RSIs segmentation networks by adaptively enhanced spectral feature awareness. SFANet demonstrates significant advancements over other multimodal RSIs segmentation networks and provides new perspectives for RSI-specific network design by incorporating spectral characteristics. This work offers new perspectives for the design of segmentation networks for RSIs.