Skip To Content

Introduction to the model

Banner image for the model

Wildfire and smoke classification is crucial for early detection and response. With increasing wildfire incidents exacerbated by climate change and urban expansion, timely identification of potential fires and smoke can significantly improve emergency response, reduce damage, and save lives. Deploying an automated wildfire and smoke classification system can help wildfire management personnel continuously monitor vast and remote forest areas prone to wildfires.

The Wildfire And Smoke Classification pretrained model allows you to identify the presence of wildfires and the early signs of smoke that may lead to a wildfire. The model uses imagery from aerial drones or ground-based camera systems and can classify wildfires and smoke across various settings and lighting conditions. Identifying wildfire threats early enables quicker intervention and more effective resource allocation, leading to improved wildfire management and response, and helping to mitigate the impact of wildfires on communities and ecosystems.

License requirements

To complete this workflow, the following are the license requirements:

  • ArcGIS DesktopArcGIS Image Analyst extension for ArcGIS Pro
  • ArcGIS EnterpriseArcGIS Image Server
  • ArcGIS OnlineArcGIS Pro or Professional Plus user type.

Model details

This model has the following characteristics:

  • Input— High-resolution three-band drone images or an orthomosaic is used as input in a compatible image format pursuant to the Classify Objects Using Deep Learning tool.
  • Output—Classified image of the scene labeled as either fire or nofire.
  • Compute—This workflow is computationally intensive and a GPU with a compute capability of 6.0 or higher is recommended.
  • Applicable geographies—This model is expected to work well in all regions globally. However, results can vary for images that are statistically dissimilar from the training data.
  • Architecture—This model uses the ResNet50 model architecture implemented in ArcGIS API for Python.
  • Accuracy metrics—This model has an overall accuracy of 96.28 percent.

Access and download the model

To access and download the model, complete the following steps:

  1. Download the Wildfire And Smoke Classification 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.

  2. Browse to ArcGIS Living Atlas of the World.
  3. Sign in with your ArcGIS Online credentials.
  4. Search for Wildfire And Smoke Classification and open the item page from the search results.
  5. Click the Download button to download the model.

    You can use the downloaded .dlpk file directly in ArcGIS Pro.

Release notes

The following are the release notes:

DateDescription

August 2024

First release of Wildfire And Smoke Classification