Step 1: field data collection - monviso/Thermal-map-from-low-cost-UAV-camera GitHub Wiki

Let's go to the field!

We report here data from the Semenyih River, Malaysia (https://goo.gl/maps/ECGuaAT6Ahbt5dm17). Data collection if field with UAV has been extensively covered in many publications (see references). Few overall tips:

  • Let the camera heat for at least 20 min
  • flight slowly (max 4 m/s) with images frontal overlap of 80% and lateral overlap of 70%
  • If you need to choose a colour palette for your sensor that will result in the output thermal image (as for DJI M2ED) do some test to chose one that maximize contrast in the thermal images
  • place regular Ground Control Points to cover all the interested area if to maximize the accuracy of the absolute position of your map (ideally you need at least 1 GCP visible in every image and 2 or more in the central area).
  • Visible GCP from thermal images are squared carton boards (40x40 cm, but size depending on flight altitude and camera resolution)covered with aluminium foils (see following image)

Figure6 (2)

To correctly apply the method, water temperature loggers must be placed to heavenly cover the area of study in a sufficient number to be split into two subset (see step 5.). Following the setting we developed for the Semenyih river.

map