Rainbow trapping is a significant strategy for capturing waves of different frequencies at different spatial positions. It offers significant potential for multi-frequency on-chip nanophotonic devices in the era of big data and has attracted much interest recently, and topological rainbow trapping that is immune to disorder has been observed in optics and acoustics. Reported here is the topological rainbow trapping of elastic waves using gradient valley phononic crystal plates (PCPs) with a hexagonal lattice, based on valley edge states that are localized at the boundary between two PCPs with opposite valley topological phases. The gradient valley PCPs are achieved by gradually adjusting the height of the scatterers or the thickness of the PCPs. The topological rainbow trapping of elastic waves is shown by applying vertical vibration to the studied gradient PCPs, which are simple in structure and easily fabricated, providing a good solution for manipulating and trapping elastic waves.
The performance of adhesive-bonded joints, which are being increasingly used in the aerospace and automotive industries, under hot-humid conditions remains largely unknown. The effect of surface treatments on the strength durability of aluminum-lithium (Al-Li) alloy adhesive-bonded joints under hot-humid exposure was investigated. Varied surface characteristics were achieved by different mechanical and chemical treatments; combined mechanical and phosphoric acid anodization (PAA) treatments were also used. The adhesive-bonded joints were then exposed to a hot-humid environment. The surface morphology, strength degradation, failure modes, and corrosion resistance of the substrates were analyzed. The results indicated that the combined PAA/mechanical method improved the joint durability under hot-humid exposure. The treated surface, with a roughness of 2.85 μm, exhibited the best durability, losing only 6% of strength after exposure; the failure mode remained cohesive-failure dominant. The durability enhancement was attributed to the porous and moderate rough surface characteristics, which protected the substrate from corrosion and decreased the rate of moisture diffusion from the bonding interface, thereby reducing the degradation of the bonding interface and improving the durability of the joint.
Small targets in drone imagery are often difficult to accurately locate and identify due to scale imbalance and limitations, such as pixel representation and dynamic environmental interference, and the balance between detection accuracy and resource consumption of the model also poses challenges. Therefore, we propose an interpretable computer vision framework based on YOLOv12m, called LCW-YOLO. First, we adopt multi-scale heterogeneous convolutional kernels to improve the lightweight channel-level and spatial attention combined context (LA2C2f) structure, enhancing spatial perception capabilities while reducing model computational load. Second, to enhance feature fusion capabilities, we propose the Convolutional Attention Integration Module (CAIM), enabling the fusion of original features across channels, spatial dimensions, and layers, thereby strengthening contextual attention. Finally, the model incorporates Wise-IoU (WIoU) v3, which dynamically allocates loss weights for detected objects. This allows the model to adjust its focus on samples of average quality during training based on object difficulty, thereby improving the model’s generalization capabilities. According to experimental results, LCW-YOLO eliminates 0.4 M parameters and improves mAP@0.5 by 3.3% on the VisDrone2019 dataset when compared to YOLOv12m. And the model improves mAP@0.5 by 1.9% on the UAVVaste dataset. In the task of identifying small objects with drones, LCW-YOLO, as an explainable AI (XAI) model, provides visual detection results and effectively balances accuracy, lightweight design, and generalization capabilities.
Rainbow trapping and the frequency separation of elastic waves in topological phononic metamaterials hold significant promise for information processing. However, a shared acoustic channel for frequency separation complicates acoustics signal processing. In this paper, we constructed a unit cell by etching curved slots into an aluminum plate and analyzed its topological properties using two-dimensional polarization techniques. By adjusting the structural parameters and creating three gradient structures, we achieved rainbow trapping and frequency separation across a dual frequency band. This allows waves of different frequencies to be split and directed to various locations, while other frequencies can be output from different ports. This study offers an effective approach for separating and capturing acoustic signals, with potential applications in engineering and technology.
The autonomous navigation of inspection robots in complex forest environments heavily relies on accurate trunk detection. However, existing detection models struggle to achieve both high accuracy and real-time performance on resource-constrained edge devices. To address this challenge, this study proposes a lightweight algorithm named YOLOv11-TrunkLight. The core innovations of the algorithm include (1) a novel StarNet_Trunk backbone network, which replaces traditional residual connections with element-wise multiplication and incorporates depthwise separable convolutions, significantly reducing computational complexity while maintaining a large receptive field; (2) the C2DA deformable attention module, which effectively handles the geometric deformation of tree trunks through dynamic relative position bias encoding; and (3) the EffiDet detection head, which improves detection speed and reduces the number of parameters through dual-path feature decoupling and a dynamic anchor mechanism. Experimental results demonstrate that compared to the baseline YOLOv11 model, our method improves detection speed by 13.5%, reduces the number of parameters by 34.6%, and decreases computational load (FLOPs) by 39.7%, while the average precision (mAP) is only marginally reduced by 0.1%. These advancements make the algorithm particularly suitable for deployment on resource-constrained edge devices of inspection robots, providing reliable technical support for intelligent forestry management.