Weather-Domain Transfer-Based Attention YOLO for Multi-Domain Insulator Defect Detection and Classification in UAV Images

Abstract

Insulator defect detection of transmission line insulators is an important task for unmanned aerial vehicle (UAV) inspection, which is of immense importance in ensuring the stable operation of transmission lines. Transmission line insulators exist in complex weather scenarios, with small and inconsistent shapes. These insulators under various weather conditions could result in low-quality images captured, limited data numbers, and imbalanced sample problems. Traditional detection methods often struggle to accurately identify defect information, resulting in missed or false detections in real-world scenarios. In this paper, we propose a weather domain synthesis network for extracting cross-modality discriminative information on multi-domain insulator defect detection and classification tasks. Firstly, we design a novel weather domain synthesis (WDSt) module to convert various weather-conditioned insulator images to the uniform weather domain to decrease the existing domain gap. To further improve the detection performance, we leverage the attention mechanism to construct the Cross-modality Information Attention YOLO (CIA-YOLO) model to improve the detection capability for insulator defects. Here, we fuse both shallow and deep feature maps by adding the extra object detection layer, increasing the accuracy for detecting small targets. The experimental results prove the proposed Cross-modality Information Attention YOLO with the weather domain synthesis algorithm can achieve superior performance in multi-domain insulator datasets (MD-Insulator). Moreover, the proposed algorithm also gives a new perspective for decreasing the multi-domain insulator modality gap with weather-domain transfer, which can inspire more researchers to focus on the field.

Early Drought Detection in Maize Using UAV Images and YOLOv8+

“…Moreover, mAP@50:95 represents the mean average precision calculated over IOU thresholds from 0.5 to 95%, incremented at 5% intervals, resulting in a total of 10 thresholds. By averaging the AP values at these various thresholds, mAP@50:95 serves as a comprehensive evaluation metric that offers a holistic assessment of the model's object detection performance [39].…”

Detection Transformer with Multi-Scale Fusion Attention Mechanism for Aero-Engine Turbine Blade Cast Defect Detection Considering Comprehensive Features

“…Compared to the initial baseline (only using random data augmentation) where AP 50 = 86.3, the obtained AP 50 due to the optimal selection of Mixup and Mosaic image ratios was improved by 1.5 points, demonstrating the effectiveness of the Mixup and Mosaic data augmentation methods in enhancing the model's detection precision. In the second experiments, to further improve the model's detection precision, a relatively small range around the optimal value of AP 50 was set as (a, b) = [ (30,20), (30,30), (40, 20)] and the value of c was computed for subsequent testing. Referring to Table 4, the model showed the best performance when (a, b, c) = (30, 30,5).…”

“…In the second experiments, to further improve the model's detection precision, a relatively small range around the optimal value of AP 50 was set as (a, b) = [ (30,20), (30,30), (40, 20)] and the value of c was computed for subsequent testing. Referring to Table 4, the model showed the best performance when (a, b, c) = (30, 30,5). Notably, AP 50 shows a downward trend while c increases, verifying that a larger c will lead to the model overfitting which may hinder the improvement of model defect precision.…”

“…In contrast, one-stage methods can effectively address the issue of the speed of detection. For example, Liu et al [ 30 ] proposed a weather domain-integrated network aimed at the task of multi-domain insulator defect detection and classification. This technique extracts cross-modal discriminant information, thereby improving the accuracy of detecting small targets.…”