This study aims to enhance pavement crack detection methods by integrating unmanned aerial vehicles (UAVs) with deep learning techniques. Current methods encounter challenges such as low accuracy, limited efficiency, and constrained application scenarios. We introduce an innovative approach that employs a UAV equipped with a binocular camera for identifying pavement surface cracks. This method is augmented by a binocular ranging algorithm combined with edge detection and skeleton extraction algorithms, enabling the quantification of crack widths without necessitating a preset shooting distance—a notable limitation in existing UAV crack detection applications. We developed an optimized model to enhance detection accuracy, incorporating the YOLOv5s network with an Efficient Channel Attention (ECA) mechanism. This model features a decoupled head structure, replacing the original coupled head structure to optimize detection performance, and utilizes a Generalized Intersection over Union (GIoU) loss function for refined bounding box predictions. Post identification, images within the bounding boxes are segmented by the Unet++ network to accurately quantify cracks. The efficacy of the proposed method was validated on roads in complex environments, achieving a mean Average Precision (mAP) of 86.32% for crack identification and localization with the improved model. This represents a 5.30% increase in the mAP and a 6.25% increase in recall compared to the baseline network. Quantitative results indicate that the measurement error margin for crack widths was 10%, fulfilling the practical requirements for pavement crack quantification.

Due to its ability to improve mechanical properties when incorporated into cement, graphene oxide (GO) has received extensive attention from scholars. Graphene oxide is also a filler that improves the self-sensing properties of cement composites (CCs). However, existing studies have not focused sufficient attention on the electric conductivity of cement composites filled with graphene oxide (GO/CCs) and their mechanisms, especially polarization. This study examines the effects of water content and temperature on the electrical conductivity of GO/CCs. GO/CC polarization phenomena are analyzed to reveal the conductive mechanism. The results show that water has a significant influence on the electrical conductivity of GO/CCs. With increasing water loss, the electrical resistivity of GO/CCs increases by four orders of magnitude. For the same water content, a 0.1% GO concentration significantly decreases the resistivity of GO/CCs. Temperature can significantly enhance the current intensity of GO/CCs; furthermore, there is a quadratic relationship between current intensity and temperature. The conductive mechanism of GO/CCs is attributed to the interaction between ionic conductivity and electronic conductivity.

The structural damage repair of composite material is an important issue that needs to be addressed during the service life of composite materials. To investigate the effects of a scarf structure and scarf angle on the repair quality of composite material, this paper proposes a mixed-scarf (MS) repair structure that combines ramped-scarf (RS) and stepped-scarf (SS) repair structures. The effect of the repair structure on the mechanical properties was analyzed, as well as the quality of the adhesive interface. The results show that at a scarf angle of 10°, the repair efficiency and the quality of adhesive interface are better than that of scarf angles of 20° and 30°. At a scarf angle of 10°, the recovery degree of the flexural strength of the MS repair structure is 79.72%, which is 6.77% and 38.24% higher than that of the RS and SS repair structures, respectively. However, in terms of flexural modulus, regardless of repair structure, the flexural modulus is highest at a scarf angle of 20°. Furthermore, the impact strength of the MS repair structure is approximately 87.60% that of the RS repair structure; additionally, it exhibits an increase of 45.83% compared to the SS repair structure. Overall, the quality of the adhesive interface for the RS and MS repair structures is similar and better than that of the SS repair structure. In conclusion, the MS repair structure is well suited for small-angle scarf repairs, whereas the RS repair structure is more appropriate for large-angle repairs; in contrast, the SS repair structure demonstrates the least effective performance in terms of repair outcomes.

The safe and reliable operation of aero-engines critically depends on the structural integrity of their key components. To address the challenges in aero-engine defect detection, including diverse morphologies, complex backgrounds, small targets, and large-scale variations, a geometry-adaptive and context-optimized fusion network (GC-Net) was proposed. To enhance feature extraction for geometrically complex defects, an adaptive geometric convolution module dynamically aligns convolution kernels with irregular contours of defects via deformable sampling and an edge-aware mechanism. A cascaded context optimization module was designed to integrate dual-path channel aggregation and dual-axis spatial attention to suppress background interference and improve the feature representation of small targets. A dual-direction detail-enhanced fusion network was constructed to facilitate deep interactions across multi-scale features and improve detection consistency for defects at different scales. A cross-scale fusion detection head was developed to achieve dynamic multi-scale feature fusion and accurate bounding box localization. Furthermore, a composite regression loss function combining scale‐dynamic loss and shape‐aware IoU was designed to further optimize the model’s performance in detecting multi‐scale and irregularly shaped defects. Experimental results on a self-built aero-engine borescope image dataset show that GC-Net surpasses the baseline model by 4.8% in mAP ₅₀ and 3.8% in mAP _50:95 , with notable relative improvements on small and medium-sized defects despite severe class imbalance inherent in real-world inspection data. The model achieves a recall of 81.5% and a precision of 93.3%, while maintaining 5.1 M parameters and 135.5 FPS. Cross-domain experiments on the public industrial surface defect datasets GC10-DET and NEU-DET further validate the robustness and transferability of the proposed method, demonstrating its potential applicability beyond the specific aero-engine domain. GC-Net achieves an excellent balance between accuracy, speed, and model complexity, offering a reliable and efficient visual inspection solution for the intelligent maintenance of aero-engines.