Parrot thermal radiometric jpeg file

[cpiga]

Hello,
The bebop parrot thermal is a drone UAV equipped with a flir one thermocamera that record the radiometric data and the palette data (I suppose) inside jpg file.
Exiftool is able to extract this data for Flir file format but doesn't work with the Parrot Bepop Thermal one.
I attach a file as example.
Can you help me to extract the thermal radiometric data with exiftool?

[Phil Harvey]

I can extract the attached thermal image from this file.  Is this what you are looking for?

If so, I will add the ability to extract this as "RawThermalImage".

BTW, is this drone called the "Parrot Bebop-Pro Thermal"?  And is there an option to store a higher-resolution thermal image?  (I would like to see if it is possible for the image to span multiple JPEG segments, which it would need to do if its size were greater than 64 kB.)

- Phil

[cpiga]

Great! Please explain to me how to do this miracle
Yes this drone is the Parrot Bebop-Pro Thermal.
This is the link to the parrot website:
https://www.parrot.com/business-solutions-us/parrot-professional/parrot-bebop-pro-thermal

Carlo

[StarGeek]

Update to the newest release.

Try this command
exiftool -b -W %d%f.%s -RawThermalImage <FileOrDir>

If you're on Windows, make sure to use CMD and not Powershell.

[JimM]

First time posting.
Thanks to Phil Harvey and Tomas123 for providing very useful information on thermal imaging metadata.

Regarding the Parrot radiometric jpeg file posted above, I notice that Flir calibration data is contained in the XMP data:

Code Select Expand


---- XMP-Parrot ----
FLIR Calib R                    : 1479680.000000000
FLIR Calib B                    : 1435.000000000
FLIR Calib F                    : 1.000000000
FLIR Calib O                    : 1740.220000000
FLIR Calib Tau Win              : 0.800000000
FLIR Calib T Win                : 25.000000000
FLIR Calib T Bg                 : 22.000000000
FLIR Calib Emissivity           : 0.980000000
FLIR Optical Center Diff Yaw    : -1.441612005
FLIR Optical Center Diff Pitch  : -1.260097980
FLIR Optical Center Diff Roll   : 89.359703064
FLIR Scaling Factor             : 0.992968976


Some of these parameters are the same as Tomas123 has previously posted for Flir cameras, but some are different.
Does anyone know how to use these parameters to calculate temperatures ?
I looked on the Parrot website forum but couldn't find anything.
TIA.

[Phil Harvey]

Hi Jim,

I suggest you ask this question in this thermal imaging forum.

- Phil

[JimM]

I wasn't able to find anything on the thermal imaging eevblog, however, I think I figured out how to use the Calibration parameters shown above. Previously, tomas123 posted a spreadsheet  that used 15 parameters  to calculate object temperature. That calculation included the affect of atmospheric loss.

I'm guessing here a little bit, but first of all, in the tomas123 calculations, the terms R1 and R2 (or PlanckR1 and PlanckR2) always appear as R1/R2. So CalibR above is just R1/R2. Secondly, instead of atmospheric loss we have the loss due to the window (in front of the camera) this is CalibTauWin. I guess this loss is much larger than atmospheric loss. Similarly, instead of atmospheric temperature we have CalibTWin. CalibTBg is reflected temperature. The temperatures need to be in Kelvin. If we substitute these terms in the tomas123 spreadsheet equations, we can calculate temperatures. If anyone sees anything wrong with this, please let me know.

Edit to add that CalibO (offset) is defined as positive. In previous Flir metadata it was negative. So need to change the sign in equations where PlanckO is used.