IGD-YOLOv8s: insulator defect detection via iterative attention and generalized dynamic feature pyramids

Insulators are important components of transmission lines. Aiming at the problems of small detection targets and complex detection backgrounds in the images of insulators taken by drones in power inspections, this paper proposes an improved BM-YOLOv5 insulator defect detection method based on the YOLOv5 network. The multi-head self-attention structure is introduced to improve the performance of the model to capture long-distance dependencies, so that it can pay attention to global information; the original feature pyramid structure is replaced with a weighted bidirectional feature pyramid structure, and additional weights are added to features of different scales. Make full use of features between different scales. Based on the newly constructed insulator defect data set with complex background and small targets, the method is compared with the commonly used insulator defect detection models YOLOv5, YOLOv3, Faster-RCNN, and SSD. Compared with YOLOv5, the average accuracy has increased by 5.1%.

<abstract> <p>Insulators are crucial insulation components and structural supports in power grids, playing a vital role in the transmission lines. Due to temperature fluctuations, internal stress, or damage from hail, insulators are prone to injury. Automatic detection of damaged insulators faces challenges such as diverse types, small defect targets, and complex backgrounds and shapes. Most research for detecting insulator defects has focused on a single defect type or a specific material. However, the insulators in the grid's transmission lines have different colors and materials. Various insulator defects coexist, and the existing methods have difficulty meeting the practical application requirements. Current methods suffer from low detection accuracy and mAP0.5 cannot meet application requirements. This paper proposes an improved you only look once version 7 (YOLOv7) model for multi-type insulator defect detection. First, our model replaces the spatial pyramid pooling cross stage partial network (SPPCSPC) module with the receptive filed block (RFB) module to enhance the network's feature extraction capability. Second, a coordinate attention (CA) mechanism is introduced into the head part to enhance the network's feature representation ability and to improve detection accuracy. Third, a wise intersection over union (WIoU) loss function is employed to address the low-quality samples hindering model generalization during training, thereby improving the model's overall performance. The experimental results indicate that the proposed model exhibits enhancements across various performance metrics. Specifically, there is a 1.6% advancement in mAP_0.5, a corresponding 1.6% enhancement in mAP_0.5:0.95, a 1.3% elevation in precision, and a 1% increase in recall. Moreover, the model achieves parameter reduction by 3.2 million, leading to a decrease of 2.5 GFLOPS in computational cost. Notably, there is also an improvement of 2.81 milliseconds in single-image detection speed. This improved model can detect insulator defects for diverse materials, color insulators, and partial damage shapes in complex backgrounds.</p> </abstract>

Thanks to the excellent feature representation capabilities of neural networks, target detection methods based on deep learning are now widely applied in synthetic aperture radar (SAR) ship detection. However, the multi-scale variation, small targets with complex background such as islands, sea clutter, and inland facilities in SAR images increase the difficulty for SAR ship detection. To increase the detection performance, in this paper, a novel deep learning network for SAR ship detection, termed as attention-guided balanced feature pyramid network (A-BFPN), is proposed to better exploit semantic and multilevel complementary features, which consists of the following two main steps. First, in order to reduce interferences from complex backgrounds, the enhanced refinement module (ERM) is developed to enable BFPN to learn the dependency features from the channel and space dimensions, respectively, which enhances the representation of ship objects. Second, the channel attention-guided fusion network (CAFN) model is designed to obtain optimized multi-scale features and reduce serious aliasing effects in hybrid feature maps. Finally, we illustrate the effectiveness of the proposed method, adopting the existing SAR Ship Detection Dataset (SSDD) and Large-Scale SAR Ship Detection Dataset-v1.0 (LS-SSDD-v1.0). Experimental results show that the proposed method is superior to the existing algorithms, especially for multi-scale small ship targets under complex background.

UAV infrared sensor technology plays an irreplaceable role in various fields. High-altitude infrared images present significant challenges for feature extraction due to their uniform texture and color, fragile and variable edge information, numerous background interference factors, and low pixel occupancy of small targets such as humans, bicycles, and diverse vehicles. In this paper, we propose a Multi-scale Dual-Branch Dynamic Feature Aggregation Network (MDDFA-Net) specifically designed to address these challenges in UAV infrared image processing. Firstly, a multi-scale dual-branch structure is employed to extract multi-level and edge feature information, which is crucial for detecting small targets in complex backgrounds. Subsequently, features at three different scales are fed into an Adaptive Feature Fusion Module for feature attention-weighted fusion, effectively filtering out background interference. Finally, the Multi-Scale Feature Enhancement and Fusion Module integrates high-level and low-level features across three scales to eliminate redundant information and enhance target detection accuracy. We conducted comprehensive experiments using the HIT-UAV dataset, which is characterized by its diversity and complexity, particularly in capturing small targets in high-altitude infrared images. Our method outperforms various state-of-the-art (SOTA) models across multiple evaluation metrics and also demonstrates strong inference speed capabilities across different devices, thereby proving the advantages of this approach in UAV infrared sensor image processing, especially for multi-scale small target detection.

UAV flight is a highly autonomous special flight activity, it can be detected in the air, on the ground and any other environment, it has the advantages of high efficiency and safety. UAV has the characteristics of safety and flexibility, and it can replace manual power line inspection and insulator defect detection. This paper mainly realizes the image recognition of insulators through the research on the image defect recognition of aerial insulators, and carries out the insulator defect detection on the image recognition results. The traditional insulator defect detection method is to identify insulator defects through the action of mechanical force using probe emission, and then through a series of complex physical processes to form an electrical signal. But this method can no longer meet the requirements of modern society for aircraft power systems. Therefore, researchers have proposed a higher level and more accurate defect detection method, which is realized by computer technology to detect insulators, so that the possible danger points during flight can be more accurately judged, and this reduces to a certain extent the human factors that cause losses when accidents occur. This paper mainly focuses on the UAV paddle view angle, orientation and other parameters extraction algorithm to study.

Insulators are critical electric components in transmission lines. Recognizing insulators and detecting the faults timely and accurately is essential for maintaining the safety and stability of transmission lines. Traditional methods have low accuracy and poor applicability in insulator recognition and fault detection. An insulator recognition and fault detection model was proposed in the article aiming at improving the insulator recognition and fault detection accuracy. First, based on the faster region convolutional neural network (RCNN), the feature pyramid networks (FPNs) were used to improve the Faster RCNN model and locate the insulators with complex background image. Then, the target area was clipped to remove the redundant background noise, and the hue, saturation, and value (HSV) color space adaptive threshold algorithm was applied for image segmentation due to the influence of light, background noise, and shooting angle. Finally, line detection, image rotation, and vertical projection were used to finish the insulator fault detection. The experimental results show that the proposed insulator recognition and fault detection model can recognize the insulators and detect fault types with better accuracy and achieve a mean average precision (mAP) of 90.8% for glass insulators and 91.7% for composite insulators on the testing dataset. Additionally, the proposed method meets the intelligent inspection of insulator faults in transmission lines and has good engineering application value.

SnakeNet: An adaptive network for small object and complex background for insulator surface defect detection

The faults caused by insulator defects will seriously threaten the operational safety of the power grid. Therefore, insulator defect detection play a crucial role in inspecting transmission lines. Compared with traditional methods, the network such as You Only Look Once (YOLO) family based on deep learning have high accuracy and strong robustness in insulator recognition and fault detection. However, the performance of these network are usually affected by the shooting conditions as well as aerial images with diverse types of insulators and complex backgrounds, resulting in poor detection result. In addition, the relatively small insulator fault (bunch-drop) area in aerial images will also make detection difficult. To solve these problems, this paper proposes an improved insulator defect detection model based on YOLOv4 (ID-YOLO). To create our model, we design a new backbone network structure, Cross Stage Partial and Residual Split Attention Network (CSP-ResNeSt), that can solve the interference problem of complex backgrounds in aerial images to enhance the network’s feature extraction capability. In addition, we adopt a new multiscale Bidirectional Feature Pyramid Network with Simple Attention Module (Bi-SimAM-FPN), which can address the difficulty of identifying a small scale of insulator defects in an image for more efficient feature fusion. We experimentally demonstrate that the mean average precision (mAP) of the proposed model is 95.63%, which is 3.5% higher than that of the YOLOv4. Most importantly, the detection speed of this model can reach 63 FPS, which meets the requirements of real-time detection of insulator bunch-drop faults.

Insulator fault detection is one of the essential tasks for high-voltage transmission lines’ intelligent inspection. In this study, a modified model based on You Only Look Once (YOLO) is proposed for detecting insulator faults in aerial images with a complex background. Firstly, aerial images with one fault or multiple faults are collected in diverse scenes, and then a novel dataset is established. Secondly, to increase feature reuse and propagation in the low-resolution feature layers, a Cross Stage Partial Dense YOLO (CSPD-YOLO) model is proposed based on YOLO-v3 and the Cross Stage Partial Network. The feature pyramid network and improved loss function are adopted to the CSPD-YOLO model, improving the accuracy of insulator fault detection. Finally, the proposed CSPD-YOLO model and compared models are trained and tested on the established dataset. The average precision of CSPD-YOLO model is 4.9% and 1.8% higher than that of YOLO-v3 and YOLO-v4, and the running time of CSPD-YOLO (0.011 s) model is slightly longer than that of YOLO-v3 (0.01 s) and YOLO-v4 (0.01 s). Compared with the excellent object detection models YOLO-v3 and YOLO-v4, the experimental results and analysis demonstrate that the proposed CSPD-YOLO model performs better in insulator fault detection from high-voltage transmission lines with a complex background.

Infrared small target (IRST) detection focuses on segmenting small infrared targets from complex backgrounds. Recent Convolutional Neural Networks (CNNs) show strong performance on detecting infrared small targets with complex background. Existing CNNs-based methods mainly have two weaknesses. First, features of small targets are likely to lose in deep stages of networks. Second, infrared small targets are always shapeless, which will cause more false detections. To solve the above mentioned two weaknesses, we propose a saliency-transformer combined knowledge distillation guided network (ST-KDNet). In our proposed ST-KDNet, we first use transformer-based segmentation branch to extract the attention region of small targets. Then we apply saliency detection branch to filter some irrelevant similar targets, where the saliency mask is used to guide the transformer-based segmentation branch. To further enhance representation ability of small target on the low-level feature, we introduce a knowledge distillation guidance. Extensive experiments on benchmark datasets, MDFA and SIRST, prove that ST-KDNet outperforms previous state-of-the-art (SOTA) methods.KeywordsST-KDNetSaliency-transformerKnowledge distillationInfrared small target detection

Efficient and accurate insulator defect detection is essential for maintaining the safe and stable operation of transmission lines. However, the detection effectiveness is adversely impacted by complex and changeable environmental backgrounds, particularly under extreme weather that elevates accident risks. Therefore, this research proposes a high‐precision intelligent strategy based on the synthetic weather algorithm and improved YOLOv7 for detecting insulator defects under extreme weather. The proposed methodology involves augmenting the dataset with synthetic rain, snow, and fog algorithm processing. Additionally, the original dataset undergoes augmentation through affine and colour transformations to improve model's generalisation performance under complex power inspection backgrounds. To achieve higher recognition accuracy in severe weather, an improved YOLOv7 algorithm for insulator defect detection is proposed, integrating focal loss with SIoU loss function and incorporating an optimised decoupled head structure. Experimental results indicate that the synthetic weather algorithm processing significantly improves the insulator defect detection accuracy under extreme weather, increasing the mean average precision by 2.4%. Furthermore, the authors’ improved YOLOv7 model achieves 91.8% for the mean average precision, outperforming the benchmark model by 2.3%. With a detection speed of 46.5 frames per second, the model meets the requirement of real‐time detection of insulators and their defects during power inspection.

In a complex background, insulator fault is the main factor behind transmission accidents. With the wide application of unmanned aerial vehicle (UAV) photography, digital image recognition technology has been further developed to detect the position and fault of insulators. There are two mainstream methods based on deep learning: the first is the “two-stage” example for a region convolutional neural network and the second is the “one-stage” example such as a single-shot multibox detector (SSD), both of which pose many difficulties and challenges. However, due to the complex background and various types of insulators, few researchers apply the “two-stage” method for the detection of insulator faults in aerial images. Moreover, the detection performance of “one-stage” methods is poor for small targets because of the smaller scope of vision and lower accuracy in target detection. In this article, the authors propose an accurate and real-time method for small object detection, an example for insulator location, and its fault inspection based on a mixed-grouped fire single-shot multibox detector (MGFSSD). Based on SSD and deconvolutional single-shot detector (DSSD) networks, the MGFSSD algorithm solves the problems of inaccurate recognition in small objects of the SSD and complex structure and long running time of the DSSD. To resolve the problems of some target repeated detection and small-target missing detection of the original SSD, the authors describe how to design an effective and lightweight feature fusion module to improve the performance of traditional SSDs so that the classifier network can take full advantage of the relationship between the pyramid layer features without changing the base network closest to the input data. The data processing results show that the method can effectively detect insulator faults. The average detection accuracy of insulator faults is 92.4% and the average recall rate is 91.2%.

Infrared dim and small target detection based on U-Transformer

How to detect small targets accurately under complex background and low signal-to-clutter ratio is of great significance to the development of precision guided weapons and infrared early warning. The traditional local contrast method is difficult to detect small and dim targets in complex background. In this paper, in order to improve the traditional local contrast method and detect small targets effectively under complex background conditions, a novel method base on Facet-kernel filtering local contrast measure (FFLCM) is proposed for small target detection. Initially, a nest sliding window structure of the central layer and the surrounding background layer is given. Then, the Facet-kernel filter is used to enhance the target in the center layer, the gray similarity difference between the central layer and the surrounding layer is calculated to suppress the background. Finally, a threshold operation is used to extract target. Experimental results demonstrate that our proposed method could effectively enhance small targets and suppress complex background clutters simultaneously.

As the failure of power line insulators leads to the failure of power transmission systems, an insulator inspection system based on an aerial platform is widely used. Insulator defect detection is performed against complex backgrounds in aerial images, presenting an interesting but challenging problem. Traditional methods, based on handcrafted features or shallow-learning techniques, can only localize insulators and detect faults under specific detection conditions, such as when sufficient prior knowledge is available, with low background interference, at certain object scales, or under specific illumination conditions. This paper discusses the automatic detection of insulator defects using aerial images, accurately localizing insulator defects appearing in input images captured from real inspection environments. We propose a novel deep convolutional neural network (CNN) cascading architecture for performing localization and detecting defects in insulators. The cascading network uses a CNN based on a region proposal network to transform defect inspection into a two-level object detection problem. To address the scarcity of defect images in a real inspection environment, a data augmentation method is also proposed that includes four operations: 1) affine transformation; 2) insulator segmentation and background fusion; 3) Gaussian blur; and 4) brightness transformation. Defect detection precision and recall of the proposed method are 0.91 and 0.96 using a standard insulator dataset, and insulator defects under various conditions can be successfully detected. Experimental results demonstrate that this method meets the robustness and accuracy requirements for insulator defect detection.