As a special insulation control, insulators protect transmission lines. In the process of detecting insulators by UAV, locating the insulators is a prerequisite for defect detection. However, In complex transmission line contexts, heavily occluded insulators are abundant. This kind of insulator is difficult to locate accurately due to its small size and extreme aspect ratio. To address this issue, an occlusion insulator detection system based on YOLOX is proposed in this work. Firstly, an improved SPP module is used to extract the semantic information of the backbone's P5 layer. Secondly, the multi-feature fusion module of attention mechanism (AFF-BiFPN) fully uses deep, shallow, and original feature information, extracting effective information about occluded insulators and small defects. To enhance the accuracy of insulator localization and the detection performance of the network, an adaptive anchor frame coarse extraction method is proposed. In addition, a data augmentation is proposed to simulate the occluded insulators. Experiments show that the method can detect defective insulators in the foreground while locating the occluded insulators in the background. On the test set, the insulator identification precision is 90.71%, and the recall is 88.25%.

Aerial image detection of transmission tower and power line in transmission lines is the key technology for unmanned aerial vehicles (UAVs) intelligent inspection, path planning, and obstacle avoidance. Different from previous transmission lines detection methods, this work uses an instance segmentation algorithm to detect transmission towers and power lines. However, the aerial survey image’s large size and high resolution result in a low signal-to-noise ratio, while the high length to diameter ratio of power lines present great challenges. To solve these issues, this paper conducts instance segmentation of aerial survey images based on the improved Solov2 network. Specifically, the neck of the model has been replaced by PaFPN to enhance the feature fusion ability and reduce feature loss. The designed MaskIou branch processes feature relationships before and after the Mask branch of Solov2 and calculates the segmentation loss of each type of mask. Transfer Learning is adopted to pre-train the network and learn data from a single type of lightning conductor so that the network can better identify linear features with high length to diameter ratio that is difficult to recognize. Comparison tests with multiple instance segmentation networks show that the proposed improved network can achieve higher precision in the segmentation.

As the main cause of foreign body intrusion, bird’s nests pose a significant potential risk to high-voltage transmission towers. Currently, most research uses convolutional neural networks (CNNs) to detect partial images of towers captured by manually operated unmanned aerial vehicles (UAVs). However, these methods have difficulties in learning the locations and potential risk information of bird’s nests and generating structured data results. To address these challenges, this paper presents a quantitative assessment framework for non-structural bird’s nest risk information of transmission towers using high-resolution UAV panoramic images. The integrity of the tower’s structure information is maintained by using panoramic UAV images. The proposed method involves estimating the tower’s pose and determining whether each key point of the tower has a bird’s nest, transforming the non-structural bird’s nest locations and risk information into structured data. A risk score is used to classify the level of potential risk caused by bird’s nests. The experiments demonstrate that the proposed framework can accurately detect the key points of five types of towers, with an F1-score of 83.09% for bird’s nest detection and an accuracy of 91.59% for risk assessment.

We propose a symmetric method of accurately estimating the number of metro passengers from an individual image. To this end, we developed a network for metro-passenger counting called MPCNet, which provides a data-driven and deep learning method of understanding highly congested scenes and accurately estimating crowds, as well as presenting high-quality density maps. The proposed MPCNet is composed of two major components: A deep convolutional neural network (CNN) as the front end, for deep feature extraction; and a multi-column atrous CNN as the back-end, with atrous spatial pyramid pooling (ASPP) to deliver multi-scale reception fields. Existing crowd-counting datasets do not adequately cover all the challenging situations considered in our work. Therefore, we collected specific subway passenger video to compile and label a large new dataset that includes 346 images with 3475 annotated heads. We conducted extensive experiments with this and other datasets to verify the effectiveness of the proposed model. Our results demonstrate the excellent performance of the proposed MPCNet.

Defect detection for non-standard industrial parts like micro induction motors has been important issue in industrial control systems. The constructional imperfections of armatures hold a large proportion of the reported induction motor failures, for which the detection methods are mostly manual inspection or image classification neural networks. However, the end-to-end networks mentioned above cannot accurately distinguish the defect types, locations, and specific state simultaneously. In this paper, the Logic-guided Intelligent Defect Detection System (LIDDS) has been proposed to address the issue above. The system consists of the Modified Feature Extraction (MFE) module, Logic-guided Classification (LC) unit, and Defect Extent Estimation (DEE) unit. The MFE module deployed attention mechanisms to enhance the ability of learning multi-scale features over background noise. The LC unit is achieved under cross-referencing multiple conditions, to jointly detect the defects with strict size constraints. Furthermore, the DEE unit estimates the extent of the complicated defects to facilitate future severity research. For evaluation, the microarmature surface feature dataset containing defects and auxiliary features is established. Experimental results indicate that the proposed system has an overall high F1 score and accuracy for the armature defect detection and classification. Meanwhile, it shows robust precision under motion and optical perturbations, supporting severity assessment in industrial environments. The dataset of this article is open source and available https://github.com/ashEsto475/microarmature_surface_feature_dataset.