When a severe natural disaster occurs, the extraction of post-disaster building damage information is one of the methods to quickly obtain disaster information. The increasingly mature high-resolution remote sensing technology provides a solid foundation for obtaining information about building damage. To address the issues with inaccurate building positioning in existing building damage assessment methods, as well as poor classification due to similar minor and major damage characteristics in building damage classification. Based on U-Net, we designed a two-stage building damage assessment network. The first stage is an independent U-Net focused on building segmentation, followed by a Siamese U-Net focused on building damage classification. The Extra Skip Connection and Asymmetric Convolution Block were used for enhancing the network's ability to segment buildings on different scales; Shuffle Attention directed the network's attention to the correlation of buildings before and after the disaster. The xBD dataset was used for training and testing in the study, and the overall performance was evaluated using a balanced F-score (F1). The improved network had an F1 of 0.8741 for localization and F1 of 0.7536 for classification. When compared to other methods, it achieved better overall performance for building damage assessment and was able to generalize to multiple disasters.

Wind power generation is a widely used power generation technology. Among these, the wind turbine blade is an important part of a wind turbine. If the wind turbine blade is damaged, it will cause serious consequences. The traditional methods of defect detection for wind turbine blades are mainly manual detection and acoustic nondestructive detection, which are unsafe and time-consuming, and have low accuracy. In order to detect the defects on wind turbine blades more safely, conveniently, and accurately, this paper studied a defect detection method for wind turbine blades based on digital image processing. Because the log-Gabor filter used needed to extract features through multiple filter templates, the number of output images was large. Firstly, this paper used the Lévy flight strategy to improve the PSO algorithm to create the LPSO algorithm. The improved LPSO algorithm could successfully solve the PSO algorithm’s problem of falling into the local optimal solution. Then, the LPSO algorithm and log-Gabor filter were used to generate an adaptive filter, which could directly output the optimal results in multiple feature extraction images. Finally, a classifier based on HOG + SVM was used to identify and classify the defect types. The method extracted and identified the scratch-type, crack-type, sand-hole-type, and spot-type defects, and the recognition rate was more than 92%.

As technology advances and society evolves, deep learning is becoming easier to operate. Many unscrupulous people are using deep learning technology to create fake pictures and fake videos that seriously endanger the stability of the country and society. Examples include faking politicians to make inappropriate statements, using face-swapping technology to spread false information, and creating fake videos to obtain money. In view of this social problem, based on the original fake face detection system, this paper proposes using a new network of EfficientNet-V2 to distinguish the authenticity of pictures and videos. Moreover, our method was used to deal with two current mainstream large-scale fake face datasets, and EfficientNet-V2 highlighted the superior performance of the new network by comparing the existing detection network with the actual training and testing results. Finally, based on improving the accuracy of the detection system in distinguishing real and fake faces, the actual pictures and videos are detected, and an excellent visualization effect is achieved.

The water shoreline is essential for unmanned surface vessels (USVs) to navigate autonomously. Many existing traditional water shoreline detections approaches not only fail to overcome the effects of water reflections, image inversions, and other factors but are also unsuitable for water shoreline detection in a variety of weather conditions and in complex inland river scenarios. Therefore, we propose a water shoreline detection approach based on an enhanced Pyramid Scene Parsing Network (PSPNet). We introduce a migration learning approach to the PSPNet feature backbone extraction network Resnet50 to improve training efficiency and add a Convolutional Block Attention Module (CBAM) attention mechanism module to improve the robustness of training. In addition, the pyramid pooling module adds the branch of the atrous convolution module. Finally, the waterfront segmentation map is processed by the Canny edge detection method, which detects the water shorelines. For the network's training and validation, we use the USVInland dataset, the world's first urban inland driverless dataset. The experimental results show that the segmentation accuracy MIou of this paper is 96.87% and Accuracy is 98.41, which are higher than some mainstream algorithms. It is capable of detecting water shorelines accurately in a variety of interior river situations.

The insulation performance of the core transformer is determined by the winding quality of the insulation layer between the coils. It is necessary to assure uniform winding tension in order to ensure the insulation performance of the transformer and improve production efficiency. Thus, in this article, the dynamic characteristics of the transformer insulation winding process are analyzed, and the insulation winding tension control system is designed to decrease the tension fluctuation caused by the alternating change of the elliptical core radius in the winding process. Firstly, the time-varying characteristics of roll radius and inertia are analyzed, the mechanical structure and dynamic models of unwinding and rewinding are established, and the effects of upstream and downstream span length and dancer inertia on the resonant frequency and response performance of the system are revealed. Later, the influence factors of tension fluctuation and disturbance propagation frequency are investigated, and the tension control model of Hybrid dancer position feedback is established. The variable universe fuzzy PI controller is designed by setting fuzzy rules and universe adjustment factors. Finally, the stability of the controller and the weakening effect on the tension disturbance are verified by simulation, and the performance and application value of the automatic winding tension control system are tested by a series of experiments.