Welcome to the NEW Infrared Training Forum. To view the old message board, please click here.
Analyzing thermal images of water in motion - Printable Version

+- Welcome to the NEW Infrared Training Forum. To view the old message board, please click here. (http://forum.infraredtraining.com)
+-- Forum: Thermography (/forumdisplay.php?fid=3)
+--- Forum: IR Study Hall (/forumdisplay.php?fid=12)
+--- Thread: Analyzing thermal images of water in motion (/showthread.php?tid=388)



Analyzing thermal images of water in motion - SuneN - 02-20-2016 11:09 AM

Hi ITC


I am doing a project with the Danish Enviromental Authority. We are looking for temperature signatures from polluted groundwater, entering the riverbed. The water in the river is 0-1C, and the entering polluted ground water is 8-10C.


We are trying to determine if thermography can be used as a tool to located the sections of the riverbed, where the infiltration is occurring.


We are using a FLIR Tau2 336 sensor, attached to a DJI S1000 for the field work. When gathering data, we fly 3 m/s at different heights, following the current.


My question is this:

Why does the area of the water surface, in the top of the image, appear warmer than the water surface near the bottom of the image? The problem persists when moving along the flow in the stream.

Here is a short video showing the problem when moving along the current:

https://www.dropbox.com/s/ypi1cbu8anmcq8i/2016.02.12_21.07.56.471_cut.avi?dl=0

(Could not uploade the .avi file)

The second time we flew, we tried hovering above certain locations of the river. It then seems as if the water closest to the center of the image is hotter than the water close to the perimeter of the image.

Here is a video of the hover state:

https://www.dropbox.com/s/nqrhrekwudlrw2e/2016.01.18_06.54.31.282_cut.avi?dl=0


Any input would be appreciated. If please request all settings that could be relevant for solving the issue.





RE: Analyzing thermal images of water in motion - Gary Orlove - 02-22-2016 11:29 AM

Can you provide the weather conditions at the time of the survey:
- Time of day
- Cloudy or Clear skies
- Altitude above water
- Camera direction (straight down, 45 deg angle, etc.)
- Approximate location of the survey
- Time of year

Can you upload black and white imagery instead of color.

Thanks


RE: Analyzing thermal images of water in motion - SuneN - 02-23-2016 04:18 AM

Hi Gary

Thank you for your reply.

For the first video, where we move along the current:

- Time of day: 21.07 (evening)
- Cloudy or Clear skies: Partly clouded, -2C
- Altitude above water: 50m
- Camera direction (straight down, 45 deg angle, etc.): Perpendicular to water surface
- Approximate location of the survey: Grindsted, Denmark
- Time of year: 2016.01.18

The second video, hover above stream:

- Time of day: 06.57 (morning)
- Cloudy or Clear skies: Clear sky,-4C
- Altitude above water: 50m
- Camera direction (straight down, 45 deg angle, etc.): Perpendicular to water surface
- Approximate location of the survey: Grindsted, Denmark
- Time of year: 2016.01.18

The .seq files can be downloaded here:

Along stream:
https://www.dropbox.com/s/fqigri6vtjr67li/2016.02.12_21.07.56.471_cut.seq?dl=0

Hover state:
https://www.dropbox.com/s/xm7khziatq4z2d9/2016.01.18_06.54.31.282_cut.seq?dl=0

If the links for some reason should not work, here is a link to the folder:
https://www.dropbox.com/sh/8u41dvh99j2hjcc/AADaTgxozXG8TiF-f9DznC32a?dl=0

If you need aditional information, please let me know Smile

Thx

Regards
Sune


RE: Analyzing thermal images of water in motion - Gary Orlove - 02-23-2016 12:44 PM

After looking at your .seq files; the variations you are seeing appear to be consistent with fixed pattern noise and possibly some vignetting. I measured an average temperature discrepancy of 0.6 C between the top and the bottom of your image.

Please perform a NUC (non uniformity correction) immediately prior to collecting data to minimize these effects.


RE: Analyzing thermal images of water in motion - SuneN - 02-24-2016 02:35 PM

Hi Gary

Again, thank you for your time, and reply.

When you say NUC, are you referring to FFC? We cant find NUC listed in the FLIR TAU2/QUARK2 SOFTWARE IDD. (it is actually listed 1 place, but not described in any way)

The camera does the FFC every 20 seconds, or when the temperature range changes in the picture. What delta T value actually triggers the FFC, im not sure of.

Kind Regards

Sune


RE: Analyzing thermal images of water in motion - Gary Orlove - 02-24-2016 02:46 PM

FFC is the same thing as NUC. There is a shutter between the camera sensor and the lens. This shutter is used to perform a flat-field correction, or FFC. During FFC, the shutter presents a uniform temperature source to each detector element in the array. While imaging the flat-field source, the camera updates the offset correction coefficients, resulting in a more uniform image after the process is complete. More information below:

http://www.flir.com/cvs/cores/knowledgebase/index.cfm?CFTREEITEMKEY=327&view=35774


RE: Analyzing thermal images of water in motion - SuneN - 02-24-2016 03:05 PM

I see.

We are binding a button on the remote to execute a manual FFC. I did not realize the importance of the FFC prior to each shot.

Thanks Gary.