Insulators are essential components in power transmission lines. Due to the harsh variations in bad environments, insulators may experience faults. Detecting these insulator faults promptly and effectively is an urgent issue. To rapidly and accurately locate insulators and their faulty regions in aerial images of insulators with complex backgrounds and varying fault sizes, this paper proposes an improved YOLOv8 algorithm for the detection of multiple insulator fault types (MFI-YOLO). This algorithm achieved target feature extraction in complex background images by replacing the C2F network constructed by fusing the GhostNet and multi-scale asymmetric convolution (MSA-GhostBlock). Furthermore, in the feature fusion stage, a multi-scale feature fusion structure called ResPANet, based on residual skip connections, was constructed to replace the PANet. This enhancement aims to improve the network detection accuracy in multi-target scenarios. Finally, to evaluate the algorithm's performance, this study constructed a target detection dataset containing four types of insulators: normal, self-explosive, damaged, and flashover. Experimental results indicate that, compared to the original model, the improved model has increased mean accuracy from 89.2% to 93.9%. The designed model exhibits high detection accuracy in the insulator and its three fault categories, especially for some hard-to-detect fitting.

The accurate and efficient detection of power lines and towers in aerial drone images with complex backgrounds is crucial for the safety of power grid operations and low-altitude drone flights. In this paper, we propose a new method that enhances the deep learning segmentation model UNet algorithm called TLSUNet. We enhance the UNet algorithm by using a lightweight backbone structure to extract the features and then reconstructing them with contextual information features. In this network model, to reduce its parameters and computational complexity, we adopt DFC-GhostNet (Dubbed Full Connected) as the backbone feature extraction network, which is composed of the DFC-GhostBottleneck structure and uses asymmetric convolution to capture long-distance targets in transmission lines, thus enhancing the model’s extraction capability. Additionally, we design a hybrid feature extraction module based on convolution and a transformer to refine deep semantic features and improve the model’s ability to locate towers and transmission lines in complex environments. Finally, we adopt the up-sampling operator CARAFE (Content-Aware Re-Assembly of FEature) to improve segmentation accuracy by enhancing target restoration using contextual neighborhood pixel information correlation under feature decoding. Our experiments on public aerial photography datasets demonstrate that the improved model requires only 8.3% of the original model’s computational effort and has only 21.4% of the original model’s parameters, while achieving a reduction in inference speed delay by 0.012 s. The segmentation metrics also showed significant improvements, with the mIOU improving from 79.75% to 86.46% and the mDice improving from 87.83% to 92.40%. These results confirm the effectiveness of our proposed method.

In transmission lines, bolt fittings are critical components that connect towers and insulators. These fittings are prone to defects, such as loosening or missing bolts, due to natural environmental factors, posing significant risks to the system's operation. Drone inspection images are increasingly used for the intelligent detection of bolt fittings, yet the small size of bolt components and the uneven distribution of sample quantities limit detection accuracy. To address these challenges, this paper proposes a multi-type bolt fitting defect detection algorithm aimed at improving detection accuracy and robustness. To tackle the issue of small target detection, we design a novel down-sampling method that incorporates both global and local feature perception, reducing the loss of small target information. This method is integrated into the Res-PANet, a multi-scale feature fusion structure based on residual skip connections, which compensates for the loss of small target details in high-level semantic features. Additionally, to address the sample imbalance problem, we introduce adaptive class weighting and a Minimum Peripheral Point Distance (MPPD) bounding box similarity constraint to better locate imbalanced sample regions from the loss function perspective. Experimental results, using a detection dataset containing four types of targets—normal bolts, missing pins, loose pins, and missing nuts—demonstrate the effectiveness of the proposed algorithm. Compared to the original YOLOv8 model, the improved model shows a 1.9% accuracy improvement for detecting missing pins, a 3.2% improvement for loose pins, and a 4.2% improvement for missing nuts. The overall mean accuracy increased from 69.5% to 72.1%. These results provide a solid technical foundation for detecting bolt fitting defects in transmission lines, offering significant practical value for future inspection tasks.

In natural environments, the connecting bolts of overhead lines and power towers are prone to loosening and missing, posing potential risks to the safe and stable operation of the power system. This paper reviews the challenges in bolt defect detection using power vision technology, with a particular focus on unmanned aerial vehicle (UAV) images. These UAV images offer a cost-effective and flexible solution for detecting bolt defects. However, challenges remain, including missed detection due to the small size of bolts, false detection caused by dense and occluded bolts, and underfitting resulting from imbalanced bolt defect datasets. To address these issues, this paper summarizes solutions that leverage deep learning algorithms. An experimental analysis is conducted on a dataset derived from UAV inspections, comparing the detection characteristics and visualizing the results of various algorithms. The paper also discusses future trends in the application of UAV-based power vision technology for bolt defect detection, providing insights for the advancement of intelligent power inspection.

To address the issue of low detection accuracy caused by the similarity between inspection targets and backgrounds in power transmission line inspection images, this paper designs a visual defect inspection framework for electrical components based on weakly supervised contrastive learning pretraining. First, two power transmission line inspection datasets are constructed: one containing coarse labels and another with fine-grained labels. The coarse label dataset is used for weakly supervised contrastive learning pretraining, where the coarse label information distinguishing the target from the background is utilized to perform feature metric learning on one-dimensional (global) and two-dimensional (local) features. This significantly improves the ability to distinguish between background and target features in local regions. The fine-grained label dataset is used for supervised target detection model transfer learning, which provides precise target location and category information, further enhancing the model’s accuracy and robustness in detecting targets against complex backgrounds. Additionally, this paper introduces a deformable convolution operation at the global image level to learn the boundaries between the background and the target. Furthermore, a novel multi-path coordinated attention mechanism is designed to dynamically optimize the boundary offset, further enhancing the global feature distinction and target detection performance. Experimental results show that the proposed method improves the detection accuracy from 76.0% to 79.4% compared to the original algorithm. When transferred to similar target detection algorithms, the method achieves an improvement ranging from 0.7% to 2.8%, effectively enhancing the performance of existing algorithms in detecting targets in complex background environments. The dataset and code are available at https://github.com/19971483525/IEEE-TIM.