Open navigation

General Purpose Analytics

These delair.ai analytics serve a variety of use cases found among different industries:

  1. Contour Lines
  2. Vehicle Detection
  3. Object Detection
  4. Thermal Mapping

1. Contour Lines

1.1. Definition


Contour lines are defined as lines joining all points of same elevation within a map.

delair.ai computes and displays these lines from the generated Digital Surface Model (DSM).

Contour lines are generated under one of these supported geospatial vector formats :

  • Shapefile
  • kml
  • GeoJSON

1.2. Generation

Once a photogrammetry process has been completed with an orthomosaic and a DSM , it is then possible to launch the Contour Lines from the Analytics panel. Check the presence of the DSM from the Layers tab :

Select Analytics then Contour Lines :

Then select settings. Choose the unit for the Elevation Interval between available Metric and Imperial System units, define the interval, as well as the output file format. Then press Launch for computation.

Computation status is shown as a tag on Contour Lines analytics (successive steps are pending, then processing, then available)

1.3. Display & Export

Once available contour lines are managed for display as a layer (Lines color - white as default - can be changed from the right panel).

Go to the Download tab if you need to export your Contour Lines for further use with a third-party software.

2. Vehicle Detection

Vehicle Detection Analytic enables an inventory of the presence of wanted or unwanted vehicles in an area of interest that was surveyed by drones. This can notably be used for counting / statistics as an inventory or to detect encroachment by a vehicle in an unauthorized zone.

This analytic runs an automated detection and classification of land vehicles from a RGB dataset, using machine learning algorithms. It is therefore possible to run it from georeferenced pictures by processing them first in an orthomap with one of the photogrammetry engines built in delair.ai, or to run the analytic from an already existing orthomap.

Currently, 7 (seven) different classes of vehicles can be recognized :

  1. Excavator
  2. Car
  3. Construction Truck
  4. Van
  5. Truck
  6. Construction Machinery
  7. Tractor

Workflow:

The analytic is launched from the Analytics side panel :

Once the process is complete, the available outputs are:

  • Visualisation : Tagged Vehicles as Vectors visible in the 2D Layers
  • Export : .geojson files for the 7 classes

Select Vehicle Detection in Layers panel to display vehicles, as in the example below :

Select the " Vehicle_detection-VehicleDetection.geojson " file from the Export section :

The .geojson files contains the coordinates of each detection in each Vehicle class.

"properties": {
        "class": "car"
      }
    },
    {
      "type": "Feature",
      "id": "166.446805--22.250718-166.446831--22.250675",
      "geometry": {
        "type": "Polygon",
        "coordinates": [
          [
            [
              166.44680544639476,
              -22.250718075291516
            ],

2. Object Detection


It is possible to order a custom object detection analytic from delair.ai in a manner similar to the Vehicle Detection. Contact us in order to analyze your requirements in terms of classes, service level agreements, availability of base datasets for algorithm training to your specific use case, and pricing.

3. Thermal Mapping

This process generates a thermal map from a thermal LWIR or MWIR pictures dataset. Sensors currently supported are the FLIR Tau 2 & Vue Pro series (.rjpeg or TIFF Format).

The generated layer can be browsed for temperature reading in inspection use cases, or serve as a basis for specific analytics such as automated Solar Plant Thermal Inspection.

3.1. Supported Sensors

Supported thermal sensors list is here.

3.2. Map Display

A thermal map is a 2D layer in delair.ai, it can be displayed or hidden from the left hand Layers panel:

3.3. Temperature Reading

To read the temperature information recorded in any point of the map, Select the Value At Point tool available from the right hand side panel of the open thermal map, hover over the map and click at the desired location to display the temperature reading. The temperature display is always Celsius.

Did you find it helpful? Yes No

Send feedback
Sorry we couldn't be helpful. Help us improve this article with your feedback.