Concrete surface crack detection is a critical research area for ensuring the safety of infrastructure, such as bridges, tunnels and nuclear power plants, and facilitating timely structural damage repair. Addressing issues in existing methods, such as high cost, lengthy processing times, low efficiency, poor effectiveness and difficulty in application on mobile terminals, this paper proposes an improved lightweight concrete surface crack detection algorithm, YOLOv8-Crack Detection (YOLOv8-CD), based on an improved YOLOv8. The algorithm integrates the strengths of visual attention networks (VANs) and Large Convolutional Attention (LCA) modules, introducing a Large Separable Kernel Attention (LSKA) module for extracting concrete surface crack and local feature information, adapted for features such as fracture susceptibility, large spans and slender shapes, thereby effectively emphasizing crack shapes. The Ghost module in the YOLOv8 backbone efficiently extracts essential information from original features at a minimal cost, enhancing feature extraction capability. Moreover, replacing the original convolution structure with GSConv in the neck network and employing the VoV-GSCSP module adapted for the YOLOv8 framework reduces floating-point operations during feature channel fusion, thereby lowering computational complexity whilst maintaining model accuracy. Experimental results on the RDD2022 and Wall Crack datasets demonstrate the improved algorithm increases in mAP50 by 15.2% and 12.3%, respectively, and in mAP50-95 by 22.7% and 17.2%, respectively, whilst achieving a reduced model computational load of only 7.9 × 10⁹, a decrease of 3.6%. The algorithm achieves a detection speed of 88 FPS, enabling real-time and accurate detection of concrete surface crack targets. Comparison with other mainstream object detection algorithms validates the effectiveness and superiority of the proposed approach.

Bridge crack detection is a key factor in ensuring the safety and extending the lifespan of bridges. Traditional detection methods often suffer from low efficiency and insufficient accuracy. The development of computer vision has gradually made bridge crack detection methods based on deep learning to become a research hotspot. In this study, a lightweight bridge crack detection algorithm, YOLO11-Bridge Detection (YOLO11-BD), is proposed based on the optimization of the YOLO11 model. This algorithm uses an efficient multiscale conv all (EMSCA) module to enhance channel and spatial attention, thereby strengthening its ability to extract crack features. Additionally, the algorithm improves detection accuracy without increasing the model size. Furthermore, a lightweight detection head (LDH) is introduced to process feature information from different channels using efficient grouped convolutions. It reduces the model’s parameters and computations whilst preserving accuracy, thereby achieving a lightweight model. Experimental results show that compared with the original YOLO11, the YOLO11-BD algorithm improves mAP50 and mAP50-95 on the bridge crack dataset by 3.1% and 4.8%, respectively, whilst significantly reducing GFLOPs by 19.05%. Its frame per second remains higher than 500, demonstrating excellent real-time detection capability and high computational efficiency. The algorithm proposed in this study provides an efficient and flexible solution for the monitoring of bridge cracks using remote sensing devices such as drones, and it has significant practical application value. Its lightweight design ensures strong cross-platform adaptability and provides reliable technical support for intelligent bridge management and maintenance.

Compressive strength of concrete is an important parameter in the design of concrete structures and the prediction of their durability. Therefore, it is of great significance to predict the compressive strength of concrete. In this study, a fully connected neural network model is developed using the PyTorch framework to predict the compressive strength of concrete and compared with six other machine learning models. These models are multiple linear regression, K‐nearest neighbor regression, support vector machine, decision tree, random forest, light gradient boosting machine, and artificial neural network. The model is trained using 4,253 data with seven input parameters, including cement (C), fly ash (F), mineral powder (K), fine aggregate (FA), coarse aggregate (CA), water reducer admixture (WRA), and water (W). Three thousand six hundred twenty‐one data in the datasets are used to train the prediction model after data cleaning, and 632 data are used to validate the model. The results show that the fully connected neural network model based on PyTorch frame can predict the compressive strength of concrete with higher accuracy. Therefore, it is a reliable and useful method to optimize the artificial network model. So, it has important application value in practice. The strength of concrete can be predicted in advance, making the project more efficient and reducing costs. Besides, by adjusting the mix ratio, combining the strength prediction results in different environments and industries to ensure the quality of construction.

In order to quantitatively analyze the electromagnetic exposure dose of an inverter in a pure electric vehicle to the driver’s body and assess the safety of the electromagnetic exposure, based on a real human anatomy model in the virtual home project, a real human model with several organs and tissues, including muscles, bones, a heart, lungs, a liver, kidneys, a bladder, a skull, a scalp, white matter, and a cerebellum, was constructed. The inverter of a pure electric vehicle is considered to be the electromagnetic exposure source; for this study, an equivalent electromagnetic environment model composed of a real human body, an inverter, and a vehicle body was built. The distribution of induced fields in the driver’s tissues and organs was calculated and analyzed using the finite element method. The results show that the distribution of the magnetic flux density, induced electric field, and induced current density in the driver’s body was affected by the spatial distance of the inverter. The farther the distance was, the weaker the value was. Specifically, due to the different dielectric properties of the different tissues, the induced field in the different tissues was significantly different. However, the maximum magnetic flux density over the space occupied by the driver’s body and induced electric field in the driver’s trunk and central nervous system satisfied the exposure limits of the International Commission on Non-Ionization Radiation Protection, indicating that the electromagnetic environments generated by the inverter proposed in this paper are safe for the vehicle driver’s health. The numerical results of this study could also effectively supplement the study of the electromagnetic environments of pure electric vehicles and provide some references for protecting the drivers of pure electric vehicles from electromagnetic radiation.

On the basis of the basic principle of electromagnetic dosimetry, the paper studies the electromagnetic exposure of a passenger’s body to a compound electromagnetic field caused by the combined action of DC and AC bus currents of a pure electric vehicle inverter. By building an electromagnetic model of a pure electric vehicle body, adult human body and inverter, the finite element method is used to calculate the magnetic induction intensity(|B|), induced electric field intensity(|E|) and induced current density(|J|) of the compound electromagnetic field to the human trunk and central nervous system in the driver and front seat passenger. The numerical results are compared with the exposure limits defined by the International Commission on Non-Ionizing Radiation Protection (ICNIRP). The fields are well below the safe exposure limits defined by ICNIRP.

Concrete surface crack detection is a critical problem in the health monitoring and maintenance of engineering structures. The existence and development of cracks may lead to the deterioration of structural performance, potentially causing serious safety accidents. However, detecting cracks accurately remains challenging due to various factors such as uneven lighting, noise interference, and complex backgrounds, which often lead to incomplete or false detections. Traditional manual inspection methods are subjective, inefficient, and costly, while existing deep learning-based approaches still have the problem of insufficient precision and completeness. Therefore, this paper proposes a new crack detection model based on an improved TransUNet: AG-TransUNet, an adaptive multi-head self-attention mechanism, and a gated mechanism-based decoding module (GRU-T) is introduced to improve the accuracy and completeness of crack detection. Experimental results show that the AG-TransUNet outperforms the original TransUNet with a 4.05% increase in precision, a 2.59% improvement in F1-score, and a 0.36% enhancement in IoU on the CFD dataset. The AG-TransUNet achieves a 2.21% increase in precision, a 5.63% improvement in F1-score, and a 9.07% enhancement in IoU on the concrete crack dataset. In addition, in order to further quantitatively analyze the crack width, the orthogonal skeleton method is used to calculate the maximum width of a single crack to provide a reference for engineering maintenance. Experiments show that the maximum error between the real values and detection results is about 5%. Therefore, the proposed method better meets the needs of crack detection in practical engineering applications and provides a solution for improving the efficiency of crack detection.