Skip to content
Loading…
This layer is editable and shared with the public. To prevent unwanted editing, unshare this item or approve it for public data collection.
Finish setting up your layer
Describe your item below. Add fields on the Data tab. Configure editing on the Settings tab. Configure drawing and pop-ups through Map Viewer or Visualization tab.

Deep learning model to perform damage assessment on drone and aerial imagery. A brief summary of the item is not available. Add a brief summary about the item.

‎Deep learning package by

New notebook runtime available. You can update the runtime from the settings tab of the item details page.

Item created: Sep 30, 2024 Item updated: Jan 2, 2025 Number of downloads: 753

Snapshot last refreshed:

1971 characters left.

Description

This model performs damage assessment on drone and aerial imagery. It is trained to identify features of interest to disaster response managers from aerial images. The model is fine-tuned on the LADI v2 dataset, which contains 10,000 aerial images labeled by volunteers from the Civil Air Patrol. The Low Altitude Disaster Imagery (LADI) dataset was created to address the relative lack of annotated post-disaster aerial imagery in the computer vision community. Low altitude post-disaster aerial imagery from small planes and UAVs can provide high-resolution imagery to emergency management agencies to help them prioritize response efforts and perform damage assessments. In order to accelerate their workflow, computer vision can be used to automatically identify images that contain features of interest, including infrastructure such as buildings and roads, damage to such infrastructure, and hazards such as floods or debris.

For LADI v2, the authors used CAP volunteers who were trained in the FEMA damage assessment process, and collected damage labels using the defined FEMA Preliminary Damage Assessment scale: unaffected, affected, minor, major, destroyed. These damage levels have specific criteria, helping reduce the subjectivity of identifying whether a structure is damaged. This model is a pretrained classifier trained on the LADI v2 dataset and serves as a basis for fine-tuning and potential deployments.

The model performs multi-label classification and categorizes images as belonging to one or more of the following classes:

  • bridges_any
  • buildings_any
  • buildings_affected_or_greater
  • buildings_minor_or_greater
  • debris_any
  • flooding_any
  • flooding_structures
  • roads_any
  • roads_damage
  • trees_any
  • trees_damage
  • water_any


Using the model

Follow the guide to use the model. Before using this model, ensure that the supported deep learning libraries are installed. For more details, check Deep Learning Libraries Installer for ArcGIS.

Fine-tuning the model
This model cannot be fine-tuned using ArcGIS tools.

Input

High resolution individual drone imagery or an orthomosaic.

Output

Feature class containing classified disaster.

Applicable geographies

The model is expected to work well in US or similar kind of geographies .

Model architecture

This model uses the google/bit-50 model architecture for classification. Refer to https://github.com/LADI-Dataset/ladi-overview for model training code

Accuracy metrics

This model has metrics for each class is mentioned here.

Training data
The model has been trained on the LADI v2 dataset. 

Sample results
Here are a few results from the model.

Predicted classes: buildings_any;flooding_any;water_any;flooding_structures;buildings_affected_or_greater 



Predicted classes: debris_any,buildings_any;water_any;trees_damage;flooding_any


Predicted classeswater_any;flooding_any;buildings_any;trees_damage;road_damage


Citation
Scheele, S., Picchione, K., & Liu, J. (2024, June 4). LADI v2: Multi-label Dataset and Classifiers for Low-Altitude Disaster Imagery.
  • Developed by: Jeff Liu, Sam Scheele. 

  • Funded by: Department of the Air Force under Air Force Contract No. FA8702-15-D-0001

  • License: MIT for code, CC-BY-4.0 for LADI datset

An in-depth description of the item is not available.

Layers

Ground Layers

Tools

Tables

Basemap

Project Contents:

Solution Contents

Contents

Layers

Screenshots

Terms of Use

esri logoThis work is licensed under the Esri Master License Agreement.

View Summary | View Terms of Use

No special restrictions or limitations on using the item's content have been provided.

Comments (0)

Sign in to add a comment.
Item Information

LowHigh

Item Information

LowHigh

Make your item easy to find, understand, and use by providing this information.

    Details

    Dashboard views: Desktop

    Creating data in:

    Published as:

      Other Views:

        Dependent items in the recycle bin

          Applicable: 2d

          Size: 89.777 MB

          Attachments size: 0 KB

          ID: 72748653112b47498764b1918c6a628d

          Image Count: 0

          Image Properties

          Layer Drawing

          Using tiles from a cache

          Dynamically from data

          Share
          Owner

          esri_analytics

          Folder

          Categories

          This item has not been categorized.

          Assign Category
          Edit Tags
          Credits (Attribution)
          No acknowledgements.

          ESRI, MIT Lincoln Laboratory, United States Department of the Air Force

          URL View
          WMTS View
          Your tile layer is ready to use
          This tile layer will automatically create tiles as needed and cache them for future use. No further configuration is required. View the Settings tab to change the default options. Build tiles manually for specific scales and extents to improve display performance for the first person to view the tile layer at that scale and extent. Tiles must exist if the layer will be used offline.
          All items were exported successfully
          ${numberOfItems} item(s) were exported successfully. Some item(s) skipped or failed to export.
          See description for more information
          Cannot import
          Export packages from newer portal versions cannot be imported to older versions.
          Loading…