• Ultraviolet Photography
  •  

Going Thermal...

20 replies to this topic

#1 Cadmium

    Member

  • Members(+)
  • 1,002 posts

Posted 01 December 2017 - 20:46

Playing around with a FLIR® (forward looking infrared), thermal imager.
Low resolution, 320 x 240, but interesting and fun.
I don't know much about using it yet, but a couple first samples:
Attached Image: FLIR0002b.jpg

#2 Cadmium

    Member

  • Members(+)
  • 1,002 posts

Posted 01 December 2017 - 23:21

Warm pipe
Attached Image: FLIR0006.jpg

Warm turkeys
Attached Image: FLIR0007.jpg

Grass
Attached Image: FLIR0008.jpg

Warm barn
Attached Image: FLIR0010.jpg

Warm ducks
Attached Image: FLIR0011.jpg

Warm truck engine (sitting 2 hours)
Attached Image: FLIR0013.jpg

Warm transformer
Attached Image: FLIR0014.jpg

#3 Andy Perrin

    Member

  • Members(+)
  • 907 posts
  • Location: United States

Posted 02 December 2017 - 00:12

Heh, I have a couple of these. 320x240 is actually state-of-the-art for consumer versions in thermal imagers. Mine is 160x120! FLIR uses a visible light camera to enhance the edges in those. Keep in mind that the temperature it reads is correct only if the emissivity is near 1, so you can't directly read the temp off a shiny metal surface. (There are tricks you can use, like putting a piece of black electrical tape on the metal and reading the temp of the tape.)

Edited by Andy Perrin, 02 December 2017 - 00:15.


#4 Cadmium

    Member

  • Members(+)
  • 1,002 posts

Posted 02 December 2017 - 01:07

Andy, Thanks! I think you can upgrade the E4 to an E8 with firmware patch/hack. I think the E4 has a native IR resolution of 80 x 60. The E6 is 160 x 120 I think? I don't know the upgrade story for that model, if any.
The MSX mode is based on the separate built in visual camera, which is for delineating the outlines of things, and is the same resolution for all those E series models. MSX can be turned off/on in the menu.
(the photos I show above have the MSX mode turned on.
The IR resolution is different for each of those models.
http://www.flir.com/...ents/ex-series/

Edited by Cadmium, 02 December 2017 - 01:08.


#5 Andy Perrin

    Member

  • Members(+)
  • 907 posts
  • Location: United States

Posted 02 December 2017 - 04:46

Mine is actually none of the above -- it's the discontinued i7 model! It doesn't even have the wifi connectivity, just USB. Also no MSX on mine, although the 160x120 seems to look about as good as the 320x240 you are showing above, which surprises me a little.

A fun thing to do is run some bathwater and then change the temperature when it's half full. I won't post the photo -- wouldn't want to spoil the fun!

Also, try a panorama.

My apartment and street:
Attached Image: My apartment and street in IR and visible UVP.jpg

Kenmore Square, Boston:
Attached Image: Kenmore bus terminal UVP.jpg

Edited by Andy Perrin, 02 December 2017 - 04:56.


#6 Cadmium

    Member

  • Members(+)
  • 1,002 posts

Posted 02 December 2017 - 08:20

Yes, I remember you posted a thermal panorama before, and I want to try that sometime too.

#7 Mark

    Member

  • Members(+)
  • 323 posts
  • Location: Massachusetts, USA

Posted 02 December 2017 - 14:46

Just FYI: The temperature reading for metallic (and other) surfaces can be compensated by adjusting the emissivity value in the software settings (the software with my FLIR has this option). It doesn't help much with mixed surfaces (as is certainly the case in these panoramas), but spot readings can be corrected.

#8 Andy Perrin

    Member

  • Members(+)
  • 907 posts
  • Location: United States

Posted 02 December 2017 - 16:15

Mark, it's not worth doing that. You can't really correct it for shiny metallic surfaces. They have an emissivity of like 0.2 or worse, which means the reflectivity is 80%. So what you measure is a mix of 80% reflected light and 20% emitted light. The emissivity correction assumes zero reflected light and it just scales up whatever value it reads by dividing by the emissivity setting essentially. Keep in mind that everything in the surroundings is glowing, so there is plenty of light to be reflected -- you can't just black it out very easily the way you can with a regular camera.

If you hold your camera up to a piece of aluminum foil, you'll see what I mean -- you measure the temp of whatever is reflected in the foil (e.g. your hand, or the camera itself or whatever is behind you).

Edited by Andy Perrin, 02 December 2017 - 16:31.


#9 Cadmium

    Member

  • Members(+)
  • 1,002 posts

Posted 03 December 2017 - 03:16

Wow you guys are way ahead of me here... but I just figured out how to turn off all the numbers/info/graphics on the screen so that when I snap a pic it is just the camera view.

Also, I noticed the same bright wheels on cars last night, as bright as the engine heat, and I suppose that is from the friction from driving?

Edited by Cadmium, 03 December 2017 - 03:19.


#10 Andy Perrin

    Member

  • Members(+)
  • 907 posts
  • Location: United States

Posted 03 December 2017 - 04:01

Yeah, brake pads heating up.

#11 OlDoinyo

    Member

  • Members(+)
  • 477 posts
  • Location: North Carolina

Posted 03 December 2017 - 18:45

Tires heat up from flexure and rolling friction. That is why one is advised to check tire pressure before driving (i.e. when cold.) Doing otherwise results in misleading readings. Since tires are dark, they radiate very efficiently, and warm tires appear bright in thermal-incandescence images.

#12 Mark

    Member

  • Members(+)
  • 323 posts
  • Location: Massachusetts, USA

Posted 03 December 2017 - 20:42

Just sharing some more thermal-IR fun. A good reason to install ceiling fans (particularly in places with 12' high ceilings like mine!...).

Looking up, here's a view of my heating/AC ducts (shown here in good 'ol VIS):
Attached Image: IMG_20171203_151719.jpg

A thermal view of the same scene just as the heat was starting coming on:
Attached Image: 20171203_145139_504.jpg

And then another shot, after the heat had been running for a while. Notice how the the upper walls and ceiling are warming up, but down near the floor it hasn't hardly changed at all! What's the sense in heating the top half of a room, when gravity is keeping me stuck down here on the floor!?
Attached Image: 20171203_151347_1143.jpg

Yes - I will be getting a fan sometime :)

P.S. - Its also cool (no pun intended) to be able to see where the cold beams are inside the walls.

#13 Cadmium

    Member

  • Members(+)
  • 1,002 posts

Posted 03 December 2017 - 21:43

Aquarium piping and meters (top left: pH, top right: ORP, bottom left: Temp F, bottom right: Salinity/Conductivity).
Piping shows warmth from water flowing through it at slightly higher temperature than surroundings.
Meters show warmth from internal electronics. Note how temperature meter (bottom left) has less heat, due to less complicated electronics inside (built those, so I know whats inside them).
Attached Image: FLIR0015.jpg

See the two black Maglite flashlights standing up on end?
Attached Image: FLIR0020.jpg

Edited by Cadmium, 03 December 2017 - 21:50.


#14 Cadmium

    Member

  • Members(+)
  • 1,002 posts

Posted 04 December 2017 - 03:07

Mark, I was noticing that also, wall studs and ceiling joists are visible to the thermal camera, like you say as cooler areas.

Now if I understand this correctly, our usual UV/IR converted 'full spectrum' cameras see from about 300nmm to 1100/1200nm max.
Our IR is near IR, AKA NIR, which is about 720nm to 1100nm.
Then there is SWIR, short wave IR which is about 1400nm to 3000nm,
and then there is MWIR which is 3000nm to 8000nm,
and then there is LWIR which is Longwave IR, which is the IR range that is used for thermal cameras.

https://en.wikipedia...in_the_infrared

#15 Andy Perrin

    Member

  • Members(+)
  • 907 posts
  • Location: United States

Posted 04 December 2017 - 05:01

Yeah, although the cutoffs are really fuzzy, especially for SWIR. LWIR is around 8000-13000nm. Beyond that, there is terahertz imaging, and microwave radiometry! I've seen images down to 90Ghz, which corresponds to 3mm. In microwaves, the ground turns into a giant mirror!

Attached Image: Screen Shot 2017-12-04 at 12.03.03 AM.jpg
from "Microwave Radiometry - Imaging Technologies and Applications" by M. Peichl et al.

Edited by Andy Perrin, 04 December 2017 - 05:05.


#16 Cadmium

    Member

  • Members(+)
  • 1,002 posts

Posted 04 December 2017 - 23:25

I wanted to show a comparison of the pure thermal image and the MSX image.
Here is what FLIR says about MSX:

"MSX®Thermal Image Enhancement.
MSX adds key details from the onboard visible light camera to the entire infrared image in real time.
The result: an all-in-one thermal picture with numbers, labels, and other structural features intact so you'll instantly recognize where the heat issue is."

One thing to note is that the MSX images don't always have the outlines in exactly the correct place, you can notice this between these two images, in some places.

Pure thermal IR image:
Attached Image: FLIR0047.jpg

MSX image:
Attached Image: FLIR0045.jpg

#17 Cadmium

    Member

  • Members(+)
  • 1,002 posts

Posted 06 December 2017 - 03:50

Spock :D
Attached Image: spock_1_FLIR0119.jpg

#18 Andy Perrin

    Member

  • Members(+)
  • 907 posts
  • Location: United States

Posted 06 December 2017 - 06:15

Have you taken a photo of your footprints on the rug yet?

#19 Cadmium

    Member

  • Members(+)
  • 1,002 posts

Posted 06 December 2017 - 06:28

Andy, On a hard wood floor, yes. ;-)

#20 Andy Perrin

    Member

  • Members(+)
  • 907 posts
  • Location: United States

Posted 06 December 2017 - 06:36

Also fun: garbage bags are transparent in LWIR. Even the black ones. (Not just because thermal conduction heats the bag or something; no contact is necessary. The plastic is just transparent in that spectrum.)