English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Deep learning-based object detection is a powerful approach for detectingfaulty insulators in power lines. This involves training an object detectionmodel from scratch, or fine tuning a model that is pre-trained on benchmarkcomputer vision datasets. This approach works well with a large number ofinsulator images, but can result in unreliable models in the low data regime.The current literature mainly focuses on detecting the presence or absence ofinsulator caps, which is a relatively easy detection task, and does notconsider detection of finer faults such as flashed and broken disks. In thisarticle, we formulate three object detection tasks for insulator and assetinspection from aerial images, focusing on incipient faults in disks. We curatea large reference dataset of insulator images that can be used to learn robustfeatures for detecting healthy and faulty insulators. We study the advantage ofusing this dataset in the low target data regime by pre-training on thereference dataset followed by fine-tuning on the target dataset. The resultssuggest that object detection models can be used to detect faults in insulatorsat a much incipient stage, and that transfer learning adds value depending onthe type of object detection model. We identify key factors that dictateperformance in the low data-regime and outline potential approaches to improvethe state-of-the-art.

Object detection-based inspection of power line insulators: Incipient fault detection in the low data-regime