Unmanned aerial vehicle (UAV) vision systems have been widely deployed for aerial monitoring applications, yet small-target detection in UAV imagery remains a significant challenge due to minimal pixel representation, substantial scale variations, complex background interference, and varying illumination conditions. Existing object detection algorithms struggle to maintain high accuracy when processing small targets with fewer than 32 × 32 pixels in UAV-captured scenes, particularly in complex environments where target-background confusion is prevalent. To address these limitations, this study proposes RMH-YOLO, a refined multi-scale architecture. The model incorporates four key innovations: a Refined Feature Module (RFM) that fuses channel and spatial attention mechanisms to enhance weak feature representation of small targets while maintaining contextual integrity; a Multi-scale Focus-and-Diffuse (MFFD) network that employs a focus-diffuse transmission pathway to preserve fine-grained spatial details from high-resolution layers and propagate them to semantic features; an efficient CS-Head detection architecture that utilizes parameter-sharing convolution to enable efficient processing on embedded platforms; and an optimized loss function combining Normalized Wasserstein Distance (NWD) with InnerCIoU to improve localization accuracy for small targets. Experimental validation on the VisDrone2019 dataset demonstrates that RMH-YOLO achieves a precision and recall of 53.0% and 40.4%, representing improvements of 8.8% and 7.4% over the YOLOv8n baseline. The proposed method attains mAP50 and mAP50:95 of 42.4% and 25.7%, corresponding to enhancements of 9.2% and 6.4%, respectively, while maintaining computational efficiency with only 1.3 M parameters and 16.7 G FLOPs. Experimental results confirm that RMH-YOLO effectively improves small-target detection accuracy while maintaining computational efficiency, demonstrating its broad application potential in diverse UAV aerial monitoring scenarios.

Structural health monitoring (SHM) relies heavily on the real-time extraction of valid information from raw streaming vibration data. However, the widely used traditional variational mode decomposition (VMD) method has two critical technical bottlenecks, which include key decomposition parameters that rely on manual empirical setting with unavoidable subjectivity and batch processing mechanisms that cannot support online real-time denoising of continuous monitoring data. To overcome these challenges, this paper develops a novel Recursive Sliding Bandwidth-Aware Variational Mode Decomposition (RSBAVMD) method and carries out a full set of scientific research. Firstly, we introduce the weighted spectrum trend (WST) method to adaptively divide the signal spectrum into sub-bands and automatically solves the optimal number of decomposition modes and the matched penalty factor for each mode, eliminating the dependence on manual parameter tuning. Then, to enable online processing of streaming SHM data, we introduce an online iterative update method based on recursive sliding Fourier transform, thus breaking the offline application limitation of conventional VMD. We conduct validation experiments using numerical simulated signals and measured structural vibration data. The results indicate that the proposed RSBAVMD method achieves adaptive determination of components and penalty factors. Compared with traditional VMD and mainstream online denoising methods, this method has the best denoising effect. Compared with the VMD method, the proposed method reduces analysis time by about 80% and is less affected by the amount of analyzed data. The proposed RSBAVMD method can satisfy the real-time denoising requirements of streaming vibration data while ensuring accuracy, providing reliable support for online SHM and structural state assessment.