• Ultraviolet Photography
  •  

Annisquam River, Gloucester, LWIR

Infrared LWIR
18 replies to this topic

#1 Andy Perrin

    Member

  • Members+G
  • 4,279 posts
  • Location: United States

Posted 24 December 2019 - 06:52

A scene I've shot many times in other wavelengths (and posted here) but never before in the longwave infrared.

This is a 38 image pano of 320x240 images shot with the FLIR E60, raw extracted with Exiftool, stitched with Panorama Stitcher, and tone-curved with Aurora HDR software. Also sharpened with Smart Deblur.

Attached Image: Gloucester pano 2_res.jpg

Incidentally, I tried straightening the horizon, but this seems to be one of those cases where it's better to "break the rules" because the image lost a lot of its drama.

Edited by Andy Perrin, 28 December 2019 - 20:42.


#2 Mark

    Member

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

Posted 28 December 2019 - 14:40

Nice pano (taken at night?). How do you take all the images for a such panos? Handheld? Do you use some sort of guide/grid?

#3 Andy Perrin

    Member

  • Members+G
  • 4,279 posts
  • Location: United States

Posted 28 December 2019 - 15:32

Taken around 1am, yeah. To avoid the sun messing with things. It was semi-handheld, on a monopod (technically a folded up tripod) you just need about 1/2 of each image overlapping the last one. One can get good at going on a side to side grid. I usually miss one or two spots per pano (usually sky) and fill them with Adobe Content Aware Fill tool and crop the ragged edges.

Edited by Andy Perrin, 28 December 2019 - 15:33.


#4 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 8,973 posts
  • Location: UVP Western Division, Santa Fe, New Mexico

Posted 28 December 2019 - 19:22

I'm curious about the wave patterns in the water. Thermally speaking, there seems to be some temperature variation in the waves closest to the camera? Or what?

You are making such good use of that FLIR camera. It is not easy working with the small 320 x 240 output.

There are now some mirrorless/dslr cameras which help make panoramas by showing part of the previously taken image to allow you to match up the next shot. It is a very useful feature.
Andrea G. Blum
Often found hanging out with flowers & bees.

#5 Andy Perrin

    Member

  • Members+G
  • 4,279 posts
  • Location: United States

Posted 28 December 2019 - 20:28

Regarding the waves: This is where you have to make a distinction between a thermal photo and any old LWIR photo. The shade of white does NOT correlate directly to temperature sometimes, and one of those times is when you have a highly reflective surface like water, and also surfaces that are nearly parallel to the camera axis, like the ground/water far away. In those cases you are seeing reflections. The water far away appears black partly because it is cold and mostly because it is reflecting the VERY cold sky. The water very near the camera is reflecting the cold sky less and is more likely dark because of its own temperature (which is still warmer than the sky).

Edited by Andy Perrin, 28 December 2019 - 20:30.


#6 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 8,973 posts
  • Location: UVP Western Division, Santa Fe, New Mexico

Posted 28 December 2019 - 22:16

I was thinking along those lines, but was not sure of course, having no real experience with this. Thank you for the details!
Andrea G. Blum
Often found hanging out with flowers & bees.

#7 Bill De Jager

    Member

  • Members
  • 311 posts
  • Location: Northern California, USA

Posted 30 December 2019 - 03:47

hotography.com/content/index.php?app=forums&module=forums&section=findpost&pid=31620'>View PostAndrea B., on 28 December 2019 - 22:16, said:

I was thinking along those lines, but was not sure of course, having no real experience with this. Thank you for the details!

I had the good fortune to see reflected LWIR in a training class. The instructors were covering various ways of surveilling wildlife at night. First they covered near-infrared photography, and took a brief video of the class in visibly dark conditions with a NIR light source. No biggie for me.

Then they demonstrated LWIR photography. They had a mug sitting on a platform and poured hot water into it. The LWIR camera trained on the mug showed the bottom part of the mug glowing bright white with LWIR emissions (this system had black and white output or had been adjusted to provide such). This was as expected.

What I found fascinating, and which is something I've never seen before or since, was the reflected LWIR. Due to the angle of the camera we could see down into the mug a little ways, and we could see light from the glowing water reflected off the inside of the mug. The water wasn't directly visible but the glare from it was. This reflected light looked just like reflected visible light (given the resolution of the camera and display system and our viewing distance) and clearly showed the texture of the inside of the mug. I have to wonder if the presenters had tuned the camera to clearly show the bright thermal light from the hot surfaces while minimizing thermal light at the longer wavelengths emitted by cooler surfaces.

Seeing reflected LWIR was a real revelation for me. Of course it exists, but it seems to usually not be evident in LWIR photos and videos, and I'd read that most materials strongly absorb LWIR..

Edited by Bill De Jager, 30 December 2019 - 03:48.

Studying the botany and plant geography of California and western North America for almost 50 years.

#8 Andy Perrin

    Member

  • Members+G
  • 4,279 posts
  • Location: United States

Posted 30 December 2019 - 04:48

Quote

Seeing reflected LWIR was a real revelation for me. Of course it exists, but it seems to usually not be evident in LWIR photos and videos, and I'd read that most materials strongly absorb LWIR..
If it's not usually evident in the LWIR photos you see online, it's mainly due to photographers deliberately minimizing reflections. Just as an example, any smooth glass or metal surface will show a strong LWIR reflection. Water somewhat less so. Also nearly all materials show reflections at shallow angles to the camera axis, which is what's happening in my photo above to the water in the distance.

Reflectivity = 1 - emissivity, (for opaque materials at equilibrium). Here are some common emissivities:

Aluminum Foil 0.04
Asphalt 0.93
Concrete 0.85
Glass smooth 0.92 - 0.94
Water 0.95 - 0.963

So, for example, aluminum foil with emissivity 0.04 reflects 1-0.04 = 0.96 = 96% of incoming LWIR light. This is why it's good for wrapping baked potatoes. However, even glass reflects 1-0.92=0.08 = 8% of incoming light, which is more than a single reflection in visible light (4%)!

As implied by the prior discussion, the emissivity is actually a function of the camera angle to the surface and wavelength, so numbers like the 0.92 for glass are an average over all angles and all wavelengths in the LWIR.

Quote

and I'd read that most materials strongly absorb LWIR..
This is sort-of true, in that absorptivity = emissivity (at thermal equilibrium), and emissivity is usually pretty high, from like 0.8 to 0.99ish for non-metals. Metals are the big exception, and almost any piece of metal makes a nice mirror in LWIR.

Me waving to UVphoto board in LWIR window reflection

Lowering my hand over a newspaper, which reflects the light from my hand when the camera is at a low angle

Edited by Andy Perrin, 30 December 2019 - 06:05.


#9 Cadmium

    Member

  • Members+G
  • 4,195 posts

Posted 30 December 2019 - 05:41

Andy, I always enjoy your thermal images. Is that (removed location) in the distance.

Edited by Cadmium, 30 December 2019 - 09:55.


#10 Andy Perrin

    Member

  • Members+G
  • 4,279 posts
  • Location: United States

Posted 30 December 2019 - 05:44

Yes, Cadmium, and can I say this is getting a bit creepy? How are you doing this?

Edited by Andy Perrin, 30 December 2019 - 05:45.


#11 Bill De Jager

    Member

  • Members
  • 311 posts
  • Location: Northern California, USA

Posted 30 December 2019 - 06:00

Thanks for the explanation, Andy!
Studying the botany and plant geography of California and western North America for almost 50 years.

#12 Cadmium

    Member

  • Members+G
  • 4,195 posts

Posted 30 December 2019 - 09:38

View PostAndy Perrin, on 30 December 2019 - 05:44, said:

Yes, Cadmium, and can I say this is getting a bit creepy? How are you doing this?

Yeah, your right, hadn't thought about it like that. Sorry.
Well, you said "Annisquam River, Gloucester"... just Google map, trying to do a visual.
I edited it. Sorry again, my apology.

Very nice thermal image.

Which lens are you using on the FLIR E60? Is it the lens that come with the E60 standard?
How may pixels wide did this stitched pan end up being?

Edited by Cadmium, 30 December 2019 - 10:24.


#13 Andy Perrin

    Member

  • Members+G
  • 4,279 posts
  • Location: United States

Posted 30 December 2019 - 15:25

Thanks Cadmium (it was indeed taken from the second floor balcony of my mom’s house).
Image is 966x521 pixels. Uncropped it was 1021px. Lens is the one that came with the (old) E60. It is 22 degrees FOV.

Edited by Andy Perrin, 30 December 2019 - 15:26.


#14 Stefano

    Member

  • Members(+)
  • 2,047 posts
  • Location: Italy

Posted 11 January 2020 - 17:51

hotography.com/content/index.php?app=forums&module=forums&section=findpost&pid=31644'>View PostAndy Perrin, on 30 December 2019 - 04:48, said:


If it's not usually evident in the LWIR photos you see online, it's mainly due to photographers deliberately minimizing reflections. Just as an example, any smooth glass or metal surface will show a strong LWIR reflection. Water somewhat less so. Also nearly all materials show reflections at shallow angles to the camera axis, which is what's happening in my photo above to the water in the distance.

Reflectivity = 1 - emissivity, (for opaque materials at equilibrium). Here are some common emissivities:

Aluminum Foil 0.04
Asphalt 0.93
Concrete 0.85
Glass smooth 0.92 - 0.94
Water 0.95 - 0.963

So, for example, aluminum foil with emissivity 0.04 reflects 1-0.04 = 0.96 = 96% of incoming LWIR light. This is why it's good for wrapping baked potatoes. However, even glass reflects 1-0.92=0.08 = 8% of incoming light, which is more than a single reflection in visible light (4%)!

As implied by the prior discussion, the emissivity is actually a function of the camera angle to the surface and wavelength, so numbers like the 0.92 for glass are an average over all angles and all wavelengths in the LWIR.


This is sort-of true, in that absorptivity = emissivity (at thermal equilibrium), and emissivity is usually pretty high, from like 0.8 to 0.99ish for non-metals. Metals are the big exception, and almost any piece of metal makes a nice mirror in LWIR.

Me waving to UVphoto board in LWIR window reflection

Lowering my hand over a newspaper, which reflects the light from my hand when the camera is at a low angle
I am a bit late, don't mind that, but I can notice two things:
1) In the only two SWIR selfies you posted you appear a little different. Yes, the wavelengths are WAY far apart, but you seem younger in LWIR to me. Why do you appear to me like that? Who knows...
2) Did you notice that paper appears smoother in longer waves? I already said that in another post, but I wanted to point that out here because you can really see it. Try to do the same with visible light, (using something else than your hand), you probably won't see reflections that sharply.

Also I never shot LWIR, but I would definitely use a B&W, white-hot setting. It is the most scientifically correct.

#15 Andy Perrin

    Member

  • Members+G
  • 4,279 posts
  • Location: United States

Posted 11 January 2020 - 20:22

Quote

1) In the only two SWIR selfies you posted you appear a little different. Yes, the wavelengths are WAY far apart, but you seem younger in LWIR to me. Why do you appear to me like that? Who knows...
2) Did you notice that paper appears smoother in longer waves? I already said that in another post, but I wanted to point that out here because you can really see it. Try to do the same with visible light, (using something else than your hand), you probably won't see reflections that sharply.
Heh. The resolution is different, the lenses are very different, and the LWIR is always soft from the big diffraction effects you get at micron wavelengths. That last is the most likely culprit for both observations.

Quote

Also I never shot LWIR, but I would definitely use a B&W, white-hot setting. It is the most scientifically correct.
That's silly, all the false colors are, y'know, FALSE. You can make them anything and it doesn't change the information content of the image. I gave you two ways to interpret the colors: one is that whiter corresponds to more emission, the other is that white corresponds to more reflection in the inverted image. If there is no additional incoming light from (e.g.) the sun, or from light bouncing within the scene, then they are equivalent and both are "correct."

Edited by Andy Perrin, 11 January 2020 - 20:23.


#16 dabateman

    Da Bateman

  • Members+G
  • 2,918 posts
  • Location: Maryland

Posted 12 January 2020 - 13:52

No you should only use "Iron", then all the buildings you image look to be on fire.

There is no correct false color. Will all your UV images be BW?

#17 Stefano

    Member

  • Members(+)
  • 2,047 posts
  • Location: Italy

Posted 12 January 2020 - 16:01

Yes, I know that. Every image outside the visible spectrum has false colors, which usually are both artistic and containing information. You should render UV images as monochrome violet, and IR images as monochrome red, because that's how we see UV and IR (the cutoffs are not strictly defined, I like to use 400-700 nm for visible light). And those violet and red hues can not be reproduced by any screen. I have a 660 nm deep red LED, and I can guarantee you that it produces a red color so strongly red that the reddest red of any screen (255, 0, 0) is orangeish in comparison. A proper rendering for "outside the visible" photos should be B&W. If you have more signal, you have a brighter pixel, because it received more light (light is "bright" by definition). Then if you want to color UV or IR images nobody can tell you that they are wrongly colored, because even if we can see UV, we cannot distinguish UV hues, ex. between 360 and 390 nm, provided that the apparent brightness is the same. I personally like the standard false colors for UV, vith blue-violet for longer waves and yellow-green for shorter ones, because the artistic appearence of the image gains a lot from this. Seeing mostly B&W, with indigo cars and pale yellow flowers is beautiful for me.



#18 dabateman

    Da Bateman

  • Members+G
  • 2,918 posts
  • Location: Maryland

Posted 12 January 2020 - 17:57

hotography.com/content/index.php?app=forums&module=forums&section=findpost&pid=32167'>View PostStefano, on 12 January 2020 - 16:01, said:

Yes, I know that. Every image outside the visible spectrum has false colors, which usually are both artistic and containing information. You should render UV images as monochrome violet, and IR images as monochrome red, because that's how we see UV and IR (the cutoffs are not strictly defined, I like to use 400-700 nm for visible light). And those violet and red hues can not be reproduced by any screen. I have a 660 nm deep red LED, and I can guarantee you that it produces a red color so strongly red that the reddest red of any screen (255, 0, 0) is orangeish in comparison. A proper rendering for "outside the visible" photos should be B&W. If you have more signal, you have a brighter pixel, because it received more light (light is "bright" by definition). Then if you want to color UV or IR images nobody can tell you that they are wrongly colored, because even if we can see UV, we cannot distinguish UV hues, ex. between 360 and 390 nm, provided that the apparent brightness is the same. I personally like the standard false colors for UV, vith blue-violet for longer waves and yellow-green for shorter ones, because the artistic appearence of the image gains a lot from this. Seeing mostly B&W, with indigo cars and pale yellow flowers is beautiful for me.

But I can see a 720nm laser and it looked brown. Why not use brown for IR then?

#19 Stefano

    Member

  • Members(+)
  • 2,047 posts
  • Location: Italy

Posted 12 January 2020 - 18:06

View Postdabateman, on 12 January 2020 - 17:57, said:



But I can see a 720nm laser and it looked brown. Why not use brown for IR then?
If you see infrared as brown, use brown. I personally see my 736, 850 and 946 nm LEDs as progressively darker red. I don't have IR lasers to test, I know LEDs are not monochromatic.