This document explains how to use the Insulator Defect Detection pretrained model available on ArcGIS Living Atlas of the World. The model is used to detect insulators and classify defects from high-resolution oriented imagery of insulator strings on transmission towers.
Electric transmission towers use insulators to prevent the leakage of current from the conductors to the ground. These transmission lines carry electricity at very high voltages, which sometimes damages the insulators. Flashover damage happens when the current passes through the air gaps between the insulators. The insulators also break when the high-voltage current passes through the insulator body or sometimes due to mechanical load. Identifying such defects in these insulators can help in prioritizing maintenance and repairs. This can prevent loss of energy and further damage to the transmission infrastructure.
Power corporations perform regular inspections of transmission and distribution infrastructure. To perform these inspections images of transmission assets are collected using helicopters, drones, or sometimes from vehicles or people on the ground. The images collected are then manually checked to identify any defects in these assets. An inspection flight can generate hundreds of images in just one mile. There are thousands of miles of transmission lines that run across the length and breadth of a country. Manually checking each image can be a tedious task. This model can be used to automate the task of detecting defects in insulators.
License requirements
To complete this workflow, the following are the license requirements:
- ArcGIS Desktop—ArcGIS Image Analyst extension for ArcGIS Pro
- ArcGIS Enterprise—ArcGIS Image Server
- ArcGIS Online—ArcGIS Image for ArcGIS Online
Model details
This model has the following characteristics:
- Input—8-bit, 3-band, high-resolution oriented imagery.
- Output—Feature layer representing detected and classified insulators.
- Compute—This workflow is compute intensive and a GPU with compute capability of 6.0 or higher is recommended.
- Applicable geographies—This model is expected to work well in the United States.
- Architecture—This model uses the MMDetection-reppoints model architecture implemented in ArcGIS API for Python.
- Accuracy metrics—The table below summarizes the average precision of the model on the validation dataset.
Class Average precision score Broken Insulators
0.92
Insulators with flashover damage
0.86
Insulators with no issues
0.72
Access and download the model
Download the Insulator Defect Detection pretrained model from ArcGIS Living Atlas of the World. Alternatively, access the model directly from ArcGIS Pro, or consume it in ArcGIS Image for ArcGIS Online.
- Browse to ArcGIS Living Atlas of the World.
- Sign in with your ArcGIS Online credentials.
- Search for Insulator Defect Detection and open the item page from the search results.
- Click the Download button to download the model.
You can use the downloaded .dlpk file directly in ArcGIS Pro, or upload and use it in ArcGIS Enterprise. Additionally, you can fine-tune the pretrained model if necessary.
Release notes
The following are the release notes:
| Date | Description |
|---|---|
| July 2023 |
First release of Insulator Defect Detection |