Search the Community

[since a "generally formal presentation" is required, any help from the admins is appreciated] In a TriColour image we want to construct an image in which each RGB channel represents a certain wavelength band. This is analog to how our eyes and camera work in the visible spectrum, having red light in the red channel, green light in the green channel and blue light in the blue channel. In a TriColour image we do the same, except we don't use red, green and blue light, but other bands, often in the UV or IR spectrum. This is how I do it. I used an image I already posted here as my example. To build this particular image I took three photos of the same subject at 730, 850 and 940 nm. If you use different light sources, it is important to place them in the exact same spot. If you use the same light sources and filter the bands with banpass filters, you have to be careful not to move your camera when changing them. As Bernard already said, this technique is only suitable for static objects. I suggest reading his topic too, where he describes his method. Here are my images, converted to black and white (I took them directly in monochrome in-camera, and I only have the .jpgs). If your images have colors, it is very important to convert them to black and white. 730 nm: 850 nm: 940 nm: You can already see differences between the images. Now I would transform these images into "channels": I open them in IrfanView, and go to "images": and then go to "Color corrections..." (I couldn't take a screenshot of the dropdown window). Here you will find RGB sliders in the lower left: If you want to make a "red channel", drop the G and B sliders to zero. if you want to make a "green channel", drop the R and B sliders to zero, and for a "blue channel" drop R and G to zero. I converted the images as follows: Red: 940 nm; Green: 850 nm; Blue: 730 nm; This is how they should look like after the procedure: 730 nm: 850 nm: 940 nm: The last step is the stacking. I use a software called "Image Stacker" for that. If you are going to use the same software, remember to select "Stack": and this is the final result: Placing a neutral target in the images (such as PTFE) can help to balance colors. One advantage of doing the white balance in IrfanView is that it just re-weights the channels, without creating anything that wasn't there. Other members use other softwares, and you can use your own.

An interesting paper in the current issue of PLOS Biology: https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.3002444 I haven't had time to read it fully yet, but the illustrations look good!

Just had a storm pass through recently that cleared the air and left some snow on the mountains so I thought I'd do a UV-VIS-IR comparison series today. The hills are about 12 miles away and the mountains are about 40 miles out. All shots were taken with a Sony A6000 FS-converted with Sirchie 60mm quartz lens. UV Filter- Baader-U VIS Filter- Kolari UV-IR Block IR Filter- Zomei 950nm The lack of edge sharpness is really obvious in the VIS shot because of the much wider spectrum captured. I should have maybe used a green filter to limit the spectrum enough for this all-quartz lens to get a sharper image. The Sirchie lens was designed for narrowband forensic work and not for this kind of photography. These are reasonably lined up well enough to do a blink comparison if you download them. The reduction of haze in each step due to declining Rayleigh scattering of the longer wavelengths is easy to see here.

Isn't this beautiful.

After spending a year or so experimenting with IR photography I obtained a Tiffen 18A filter for UV photography. It had a similar pass spectrum to the modern day ZWB1 filter with maybe a later cut-on above 300nm. They both have a good UV-A passband but also an IR passband that extends into the deep red. I decided to use a non-panchromatic film (little or no red sensitivity) that was just sensitive to blue & green to get a 'pure' UV image. I had a roll of Kodak SO-410 film which was used for photomicrography and imaging phosphor screens like oscilloscopes and radar screens that were blue or green. I did several VIS/UV comparison shots on an old Agfa rangefinder camera to see what the differences were. To keep the bandwidths about the same I used a #47 blue filter for the VIS photos. Unfortunately I was unable to locate the matching UV & VIS negatives for most of them. Here is one set from my photo album (remember those?) that I can't find the negs for. Left pic is with the #47 blue filter and right pic is with the 18A. These show a mountain about 2 miles away that gets very hazy in the UV pic. Apparently evidence of the extra Rayleigh scattering of UV. The only set of matching VIS and UV negs I could find are the recent scans I just made below. First one is blue VIS and second one is UV-A. If you download these last two pics you can do a 'blink' test and see the different haze effect between Blue and UV-A more clearly. Aside from a few indoor tests with spectrum tubes shortly afterwards I lost interest in UV photography. Compared to the general darkening and increased haze in UV, IR photography offered haze penetration and a magical brightened rendition of vegetation. The adventure of color IR film was also a huge attraction. Only in recent years did I get interested in UV photography again thanks largely to this site and the many contributors here.

The following experiments are inspired by the findings of @Christoph Here is a series of pictures with my full spectrum canon 1200D and a stack of these three filters : Midopt Triple bandpass 550/660/850nm Lee "Flesh pink" gel GRB3 from Tangsinuo The transmission curve of the three filters combined should look something like this : Both the green and the IR spike are heavily filtered. When the TB is used alone, it seems to suffer from a very weak red transmission : without the GRB3 the red channel records as much (if not more) IR than red, requiring IR subtraction of 100%. The green spike transmission to the contrary is very powerful. To the eye (at least mine) the filter is green. I'm not sure how bright humans are able to percieve red at 660nm, its pretty deep already. I then decided to reduce the transmission of IR and green equally to give room for the red transmission. The GRB3 is used to minimise IR contamination in the RED channel and the Lee gel is used minimise green contamination in the blue/IR channel and to make it more or less match the level of output of the red channel. The blue channel stays very underexposed compared to the two others, and is brought back in white balance (pretty extreme : lower than 1900Kelvin). The exposure value with these filters ranges from 1/50s to 1/100s in sunlight at f5.6, 200iso. It's a very dark combo. All the pictures are channel swapped in Resolve from the camera Jpegs, the process is video ready. No saturation was added and zero IR substraction is needed. The fact that IR transmission is cut by 90% by the GRB3 causes some green leaks to be percieved in the Blue/IR channel. This leads to the need to apply a hue correction to the sky in order to make it look properly blue and not purple-ish. This process is simple and non-destructive. The original color of the sky after channel swap : Important notice : The channel swap causes a significant decrease in contrast and micro-contrast. For a moment I thought this was due to the Jpeg compression but it is not. Do you remember this law ? Y = 0.1 Blue + 0.6 Green + 0.3 Red It is actually crucial in understanding why the images look better unswapped. It describes how the human eye is sensitive to luminosity. The green value is far more decisive in the percieved brightness of an object than the red and blue values. A quick exemple Unswapped image : Swapped image : it looks less detailed and dynamic. The contrast beetween the bright grass and the dark trees is lessened Swapped image but with the "preserve luminance" box ticked : the image is sharper and more alive. (Click on the image and use the viewer to compare both instantaneously.) Explanation : The grass in the unswapped image appears turquoise. Let's make it cyan for the sake of clarity. The trees appear blue. Cyan RGB values are (0; 255; 255) Blue RGB values are (0; 0; 255) The minimum luminance(Y) is 0 and the maximum luminance is 1. Y(cyan grass) = 0.1x Blue(255/255) + 0.6x Green(255/255) + 0.3x Red(0/255) = 0.7 Y(blue trees) = 0.1 x B(255/255) + 0.6 x G(0/255) + 0.3 x R(0/255) = 0.1 in the unswapped image the contrast beetween the brightness of the trees and the grass is important : Y=0.1 versus Y=0.7. Now, let's do the same for the swapped image where the grass is magenta and the trees are red. Y(magenta grass) = 0.1 x B(255/255) + 0.6 x G(0/255) + 0.3 x R(255/255) = 0.4 Y(red trees) = 0.1 x B(0/255) + 0.6 x G(0/255) 0.3 x R (255/255) = 0.3 in the swapped image the contrast beetween the brightness of the trees and grass is less important than in the original image (Y=0.3 versus Y=0.4) In this chart the colors are ranked by luminance. white Y=1, yellow Y=0.9, cyan Y=0.7, green Y=0.6, magenta Y=0.4, red Y=0.3, blue Y=0.1, black Y=0 As you can see here the brightness of Magenta and Red is very close as opposed to cyan and blue that are very far apart. So that's why channel swapping sometimes makes the image lose quality. A solution to this is to tick the box "preserve luminance" in the channel mixer. It doesn't work in every situation since it makes the yellow objects turn way darker, leading to an unnatural look. Ticking the box also makes the sky way brighter. It's a tradeoff that has to be made for each individual picture. In the selection I posted above, a few have the box ticked and most don't.

I have been taking a lot of photos lately. This time I decided to investigate a pair of cucumbers. I illuminated them with a halogen spotlight, which emits enough light for IR, visible and UV. IR tri color is using my GRB3 method UV is taken with a ZWB2+QB39 stack (which surprisingly does not leak significantly, even with halogen) I will also be including the individual pictures if anyone else wanted to take a shot at processing them (please, do post). I'd especially appreciate if someone managed to stack all 7 channels continually. I only could stack them by binning the visible and the IR part of the spectrum together. The pictures are in full hd so if you still have a 1080p display, you might want to enlarge. visible IR tri color 850nm+720nm+red Aerochrome simulation GBUV full spectrum individual color channels: 950nm longpass ~850nm band ~720nm band red band green band blue band UV band (400-350nm) I also decided to stack the images in Photoshop and pick the range stack mode, I got interesting results. All of the bands stacked: All of the bands except for UV stacked: IR only stacked: IR only stacked (normalized): Bonus: IR stacked, normalized and processed with Topaz Denoise AI: Here's the IR range mapped on the visible image: Here's the range between 720 and 850 bands mapped on the visible image:

As you know, all cameras react differently to IR light once they are converted to full spectrum. I don't know if this is really well understood so I wanted to start a topic and have your opinion about it. I just got the Sony f828 (the camera that can be converted to full spectrum with just a magnet) and I can now compare its fullspectrum colors to my Canon. Canon 1200D, white balanced full spectrum colors, no additional filters : Sony f828, white balanced full spectrum colors, no additional filter : The pictures were white balanced and saturated from the RAW files in Lightroom. The two cameras are pretty far apart as we can see from these pictures. They are also pretty far apart in terms of technology : the first is an 18Mpx CMOS and the second is a 20 years old 8Mpx CCD one of a kind RGBE sensor. Sony camera are known to not perform as good as Canon with the IRchrome filter. The images above indeed show that this Canon camera has a predisposition to record IR in the red channel compared to the Sony. Now do more recent Sony cameras produce blue SOOC like this one does ? I don't know... Maybe the sony users here can help me from their experience.

Today I took a few UV/IR photos of some people in my university, including the arm of one girl who is covered in freckles (or similar). Also I used a Helios lens I have, it passes UV to about 340-350 nm and so it should behave similarly to a Canon 40 mm f/2.8 pancake. I had a PTFE square, but I didn't use it for these photos. Photos rendered in black and white. Full-spectrum Canon EOS M, Helios-44M-5 (stopped down, maybe at f/8). - UV: ZWB2 (2 mm) + Chinese BG39 (2 mm); - VIS: Chinese BG39 (2 mm). There should be a negligible amount of UV contamination; - IR: Hoya R72. UV (ISO 800, 1/2 s): VIS (ISO 100, 1/500 s): IR (ISO 100 1/500 s): Tricolour/composite (blue: UV; green: VIS; red: IR):

Multispectral photography was one of the things I was really excited to try when I finally got a UV-pass filter, and the other week I finally had some free time to go out and try it Editing took took a pretty long time (definitely a much different workflow than I'm used to), but I'm very happy with the results! I love the colors of the brush in the foreground, I think they're more nuanced than regular false-color IR with a Red25 (or similar) filter. I want to try this again on a clearer day, since the haze is definitely not being kind to the shorter wavelengths IR -> R Red -> G UV -> B Raw frames processed in Darktable, then mostly aligned with the Align_Image_Stack tool from Hugin (but the left part of the panorama needed some extra hand-alignment in Gimp since Hugin couldn't do it perfectly). Then composed in Gimp, into Hugin to stitch the two together, and finally processed further in Darktable Taken on a FS Sony A5000 with a Sigma 19mm f/2.8 DN And here are the individual IR, Vis, and UV shots: Vis (TSN575): IR (850nm): (it always amazes me how well IR is able to punch through haze!) UV (ZWB2 + TSN575):

Hello, my name is Fedia, I'm 31 and I spend lot of time creating images through digital photography. I'm stoked to be able to join this forum since it has been an incredible ressource without which I couldn't have developed the techniques I use today. I have talked to some of the people of this forum on other platforms, used their advice and want to thank them also. I've been doing photography since 2014 when I entered film school. I got my first full spectrum camera in the fall of 2020. Even before I got into full spectrum photography, the main thread I explored was color, that I achieved through filters or extreme white balance settings on regular cameras. Even today with full spcetrum photography I don't really use softwares. I try to do everything in camera. Therefore I exclusively shoot jpegs and only work contrast in Lightroom. I never opened Photoshop. I am a big technical nerd I guess you could say, but the truth is my relation to technicity has a lot to do with my artistic sensitivity, both are entangled. I have a lot of ideas of subjects to discuss on this forum so I will post about them in the near-future. But for now I figured I will post some of my photos to introduce myself. Every one of the following photos are the result of a different filter combination with no color processing outside of the body of the camera. I will precise the camera and the filters each time. Canon 1000D full spectrum + Lee "CID to Tungsten" Canon 1000D FS + Lee "loving amber" + db850 Canon 1200D FS + DB850 + Lee "Liberty green" Canon 1000D + unknown dark pink gel Canon 1000D + Lee "steel blue" Canon 1200D FS + Lee "peacock blue" + GRB3 Canon 1200D FS + hoya 80A + hoya x(0) + cokin 089 (warm diffuser) Sigma DP1s FS Sigma DP1s FS + Hoya 80C Sigma DP1s FS + GRB3 Bonus : Sigma DP1s FS no filter. If you want to see more I have an instagram : https://www.instagram.com/fedialebarboc/?hl=fr And a tumblr : https://fedialegrill.tumblr.com/ Thanks again, Fedia.

I recently aquired a planetary CMOS camera, the ZWO ASI678MC. Here is my planetary rig with 300mm f/4 PF and stacked 1.4 and 2x converters to provide an 840mm f/11 lens. This has an IMX678c 7.7mm wide back illuminated chip in 4K resolution and 2um pixel pitch. It has an excellent IR response. Example with an Antlia 685nm IR-pass filter and Nikon AF 20mm f.2.8 lens: but the ZWO chart is a little less clear on what happens on the the UV side: The sensor has an AR coated protective window. The curve above is with this window in place. A chart of this window's transmission spectrum in the specs of another ZWO camera model gives a hint that deep UV responses cannot be expected: However I happened to have a 20mm diameter 2mm thick ZWB1 filter, which by itself of course leaks IR, but found a QB39 IR cut filter, also in 20mm diameter 2mm thickness in ebay seller tangsinuo's store. The transmission is only given for the 1mm thickness: Then I found a 1,25" "moon filter" from Agena Asto that had a donor cell with a wide rim. The 20mm filters had a loose fit in it, but worked with the ZBW1 closest to the sensor and retaining ring that the slightly narrower QB39 would slip into; scotch tape was used on the edge as spacer between filters to avoid newton rings. Then I have a Cassarit 50mm f/2.8 lens that way back was given to me as part of a corroded camera in parts in a plastic bag by late Professor Krog in Oslo. As the lens did not show signs of water damage, just needed cleaning, my guess is that the camera succumbed to a humid tropical climate, possibly when he participated in the 1967 Alpha Helix Expedition in Amazonas. While it is not listed among the better classes of UV lenses, it is a triplet with old coatings. Visual response with a ZWO UVIR cut filter, all of these were captured as small videos with FireCapture and stacked in Autostakkert, Small versions are with unadjusted colors before using a gray control point on the upper right qwhite patch in the left panel. I try to keep aperture of the Cassarit at f/5.6. The UV-IR cut filter is designed for astro work and has an enhanced red respons up to above 656 nm. Light souce is my nightrider Lumen 900 bike light: Then with ZWB1+QB39 filters only with my Convoy S2+ 365nm and Tank 007TK-566, both with what is likely ZBW2 filters on the front at 4m distance from the target. 200ms exposure and Gain 320. Some preliminary tests indicate that a Gain of 100 corresponds to ISO 100 sensitivity and from there approximately a dobling of ISO for every increase in gain of 50. So this would be about ISO2000: Then the same combination in sunlight, 200ms gain 228: It is notably different color responses, with yellow-green dominating with 365nm source and sunlight colder, which I assume is due to dominance of longer wavelengths in sunlight that records more in the blue channel. To test effectiveness of the filtering here is IR pass 685 nm 1ms exposure, Gain 26: then ZWB1+QB39, 192ms Gain 26: Finally ZWB1+QB39+ IRpass 685nm, 1s Gain 405 - This would be about 10 EVs below the UV exposure if my calculations are correct: [There is a light leak at lower right as the IR pass filter was mounted to a snout that was just held onto the front of the Cassarit.] So I will conclude that for my limited practical purpose, IR leak is not a problem with this combination. A couple of days ago to my delight the first dandelions showed up outside our department at the university to allow a practical test. The initial try was handheld as my tripod was at home and it is recorded on my phone with the ZWO ASICap app, so here is just a low res overview, ZWB1+QB39 : I picked up one dandelion and brought home for a more proper test. The best result appeared with a cloud diffusing the light. ZWB1+QB39, f/5.6, 517.5ms exposure, Gain 200 (corresponding to ISO400), 12 bit mode, 32 frame .ser raw video in Firecapture, stacked the 8 best, WB adjusted according to another capture on a Teflon tape wrapped plastic case where histograms were equalized. Gamma adjusted to 2.2 in irfanview (as the video is linear), with following increase in contrast and saturation (click for larger version): Not bad for a lens that is possibly 60-70 years old and never designed for UV captures! That is all for tonight, more to follow...

Hi All, I'm considering jumping ship from my Nikon D3200 to a Sony mirrorless. I want to use more of my vintage lenses in IR, UV and Vis, and the 46.5mm FFD is restricting me from using some of them (for example, my ISCO Ultra Star HD 55mm). That's the primary driver for moving to mirrorless. Primary subjects would be landscapes, architecture, macro, and people (in that order) in IR/UV/Vis. It looks like the A7 (in terms of cost vs features vs weather sealing) fits the bill. Not really interested in the A7R at 36 megapixels, since none of my lenses resolve to that (I'm looking to keep my Nikkor DX lenses, apart from the kit lens, and use them with an adapter). I'll probably get it full spectrum converted by Kolari Vision. So, from an IR and UV perspective, does anyone have any input for: The ability to easily set a CWB in both spectrum bands (my D3200 is lousy at it. I read the methods at http://www.ultraviol...um-nikon-dslrs/ but the D3200 does not have CWB "slots" - you can either take a CWB sample, or use a photo, but using a photo usually doesn't have any apparent effect. No further tinkering is permitted within the camera). The quality of IR and UV video - how does it stack up against, say, a Lumix G camera? Is there a real advantage to having a full frame sensor for IR or UV photography? From ISO 800 upwards, my D3200's crop sensor has an (IMO) unacceptable amount of noise. I have read that full frame sensors generate less noise at higher ISOs than crop sensors, but does that also work under the same principle for non-visible light? I'm not totally sold on the A7 at this point - perhaps a Panasonic or Olympus might be a good fit, too. I have a Pentax Q10 that I've been playing around with, but even with a super short FFD of 9mm or so, the small mount flange diameter, and tiny 12MP 5.4x crop sensor, it isn't conducive to being converted. One last question on Kolari Vision: I got my D3200 converted by them about 3 years ago, before they offered the AR coating. This time around, from a UV perspective, is it worth having that done? Thanks, Andy

My interest in light didn't end when I studied laser physics in Hungary and didn't start in the UK while developing high power lasers. I'm now in Switzerland focused on the infrared part of the spectrum, NIR, SWIR, hyperspectral, multispectral. Continously collecting light: line and area scan cameras, optical simulations, spectrometers, slit goggles.. I'm DIY-optics (dot com) himself. ;) You have a very nice community here, I appreciate being a new member. Hello Everyone

Casswell, T. (2022) Gazania rigens (L.) Gaertn. (Asteraceae) Gazania. Flowers photographed in ultraviolet, infrared and visible light. Also with multispectral stack. LINK Location: 17 June 2022 Australia Collected from a street garden adjacent to the beachfront at Coolum Synonyms: several -- see Reference Other Common Names: Treasure flower Comment: This species is recognized as a weed in some areas of Australia, however it is widely used for a groundcover in suburban street gardens, median strips, traffic islands, etc in south eastern Queensland due to its hardy nature and prolific flowers in a variety of colours blooming nearly all year round. Two samples were taken and placed in a water filled vase approx 20 min after collection. Each was imaged using Baader U, Kolari U, Kolari Hot Mirror V2 & a generic 650nm infrared filter. All eight image runs were focus stacked. An attempt was made to combine infrared (R), visible (G) & UV (B) into a multispectral image (mantis shrimp vision?), which proved difficult due to the dynamic nature of the flower given its circumstances. Reference: 1. Wikipedia (18 June 2022) Gazania rigens. Wikimedia Foundation, San Francisco, CA. Equipment: Converted Canon EOS R5 EL-Nikkor 105/5.6. Baader U, Kolari U, Kolari Hot Mirror V2 & a generic 650nm infrared filters Technique: UV, visible & infrared colour balanced using a white PTFE sheet with exposure dialled down to avoid any RGB channel clipping. Focus stacked using Helicon Focus 8. ISO100, f/11 for all shots. Multispectral - aligned using PT Gui with manually placed control points Yellowish White Sample: Visible light, t = 1/100s UV (Baader U), t = 5s UV (Kolari U), t = 5s (some highlight clipping) 650nm Infrared, t = 1/80s Multispectral using the Kolari U for UV (blue channel) Yellow Sample: Visible light, t = 1/80s (slight overexposure) UV (Baader U), t = 8s UV (Kolari U), t = 8s 650nm Infrared, t = 1/80s

SAFETY WARNING: UV-C is dangerous to your eyes and your skin. UVP DOES NOT SUPPORT USING UV-C ILLUMINATION. [UV SAFETY] UV-C Light Dangers A short YouTube video that includes a 254 nm clip: https://youtu.be/CaXzgumCn34. The gear they use is interesting.

I know this is primarily a UV forum, but having put together my ZBW2 UV filter stack with some cheaper 3rd party alternatives, I've come to realize that I am sorely lacking in overall knowledge regarding light modification techniques and filters. After seeing a post on here showing the Kolari IR Chrome filter I use could be recreated by stacking 3rd party filters together, I tried looking around the forum to see if there were any other interesting combinations or filters. I noticed there isn't a pinned thread or compendium dedicated to comparing various IR and other filters. A dedicated thread/page would be extremely useful for new members such as myself!

Several days ago I have taken images of the moon in Ultraviolet Light using a Baader U 2" filter in front of a Beroflex 400mm lens mounted on a full spectrum converted Olympus OM-D EM10 Mark II camera. I have used a Nano Tracker to minimize blurring due to earth rotation. Unfortunately the sharpness of the Beroflex lens is not perfect in Ultraviolet and Infrared range, nevertheless I can see at least in two marked areas differences in brightness for UV image compared to the images taken in the Visible and Infrared range. The image taken with Olympus 75-300 mm zoom lens mounted on a standard Olympus OM-D EM10 Mark II camera is for comparison, because the sharpness of this zoom lens at 300 mm is better that the sharpness of the old Beroflex 400 mm lens. Did anyone else tried to take UV images of the moon with some success? Ultraviolet Light: Baader U 2" Visible Light: Hoya UV IR Cut Infrared Light: Zomei IR720 Visible Light: Olympus 75-300mm at 300mm on Olympus OM-D EM10 Mark II without additional filter

Same settings and equipment as the other flowers in this series, with exceptions noted below. UV 2mm UG11 + 1.75mm S8612 (Convoy S2+) Saturation strongly increased Visible BG38 2mm and DB850 filter (645-405nm) NIR-Red-Green (550-645nm + 835-875nm) Tiffen#12 + DB850 filter, and my usual algorithm for making the IRG, described in the DB850 filter thread NIR Hoya R72 + DB850 filter (835-875nm) SWIR (1500-1600nm, made by the panorama method and reduced for higher resolution) UVIVF under Convoy S2+ and using BG38 2mm and DB850 filters --- Comments: The SWIR looks a lot like the visible but with a lighter disk in this one, enough that I got scared and decided to check if my filter was leaking. However when I stacked it with my hard-coated premium Thorlabs 1200nm long pass (guaranteed >OD5), it looked the same, so I guess it just looks similar by coincidence. If you look at the bottom left of the flower, there is a tiny splash of water on a petal, so you can see that the water, although dark, isn't inky when it's just a thin layer like that. This morning the disk florets had started to open up, and they are white in the SWIR 1500-1600nm band:

I did this yesterday impulsively and it has been one of the most fun and rewarding projects I've done in a while. That's why I decided to share it with you today. I took three total images of the same subject, from the same perspective, with different parts of the spectrum filtered. The subject is a random plant my mother owns, a PTFE sheet, and my RGB LED strip controller. I wanted to use it as an improvised color checker, but it didn't work out so well since the actual dye containing layer is covered with UV- absorbing plastic from what I see. (lightsource for each image was just sunlight) Here are the images: visible IR (720nm) UV (ZWB2 + QB39) - this was shot with only the light coming through the window, it is winter so opening all the windows was not an option, but it seems to have produced an ok quality image regardless, only downside being that the exposure time needed was long, around three minutes. The images at this point are perfectly aligned, you can download them and stack them yourself, they should fit perfectly. And it was a lot of work to align them I admit, photoshop failed at doing it properly so I had to do it fully manually. Worth it. Prior to aligning, all the images were rendered using darktable, with light balance set on the PTFE sheet, most processing options turned off with the exception of highlight reconstruction in the UV image. That is why they are so flat, nothing should be blown out here. Upcoming images are more edited. I used AMaZE + VNG4 as my demosaicing method. In this next portion, I will showcase two trichromes I produced using this data and their color corrected variants. IR-R Visible-G UV-B trichrome As you can see, this image looks a little weird, I think it is because the illumination in UV was not the best, causing extreme contrast, though I'm not entirely sure if the yellow cast would go away if I brought the whole setup on open sunlight, maybe I am just unlucky and almost all the materials visualized absorb significantly more in UV, including my grey wall paint. I took the liberty of color correcting the photo to bring out some color details. Looks a bit less washed out, not perfect but sufficient. Next I have a GBUV stack. I extracted the green and blue channels from the visible image, mapped them to red and green, then I mapped the UV channel to blue. Once again, the image is washed out, so here's the color corrected version. Up next I have a few things that might not be of more "scientific" significance, but I think they're really fun. For this I used the photoshop stack mode function for smart objects. I used the maximum, minimum, median and mean functions. https://helpx.adobe.com/photoshop/using/image-stacks.html On this website, you can find a full list of possible stack modes that I could use plus their practical and mathematical explanation. I ended up not using entropy, kurtosis, range, skewness, standard deviation, summation and variance. If anyone would like to see these please do tell me so I can add them. They're mostly just a jumbled mess of seemingly random colors or completely white though. Keep in mind I normalized and edited the images a bit, so for example the upcoming minimum isn't extremely dark. If you want to see the unedited versions, again please do tell. stack mode minimum stack mode maximum stack mode mean stack mode median And the last series of images which I think are the most interesting. They are the IR and the UV images layered with the visible image in "color" mode. So that the image keeps the texture of the UV/IR image but the colors are normal. IR And my favorite, UV. I think the UV really is an interesting aesthetic that could perhaps be used in product photography or something, I dunno, just find the image really interesting to look at. That's all, I'm about to go watch a movie, I hope you liked my post, see you in the comments!

It's always amazing what some filters and a little patience can do, once the camera is put on a sturdy tripod. For this capture, I combined captures in UV, visible light, and IR, using the Nikon D200 and a very old 55mm f/3.5 Micro-Nikkor lens. I think the result was nice, but your mileage might vary. The old Micro-Nikkor does not transmit deep into UV, but that hardly matters in this setup. The scene should be familiar to Nordic people; comprising a tidal pool (Norwegian 'hellkar') on a glacially-scoured coastal rock face (Norwegian 'svaberg'). I believe the designations are quite similar in both Norwegian and Swedish languages, those two countries (+ Finland) where such rocks are common on the coastline.

Based on a discussion here I was really intrigued by the graph shown for silver with the strong dip in reflectance around 320nm. I ordered a sheet of silver leaf from ebay (edible silver leaf and it was about 2GBP delivered) and decided to test it. Silver leaf was mounted on cardboard using double sided tape. Images done with my Monochrome converted Nikon d850m and Rayfact 105mm lens. f11 and ISO400 apart from the 254nm image (which was f11 and ISO6400). Light source was a Hamamatsu LC8 200W xenon lamp, but for the 254nm image I used a 4W UVP filtered lamp. I included a 99% Spectralon diffuse reflectance standard in each image and exposure time was adjusted to keep that as constant as possible for all the images. Images cropped and resized, but no further modifications. Visible - LC8 light and room light, Baader UV/IR cut filter 390nm to 340nm - Thorlabs 10nm bandpass filter, Hamamatsu LC8 Xenon lamp 330nm to 310nm - Edmund Optics 10nm bandpass filter, Hamamatsu LC8 Xenon lamp 300nm - Edmund optics 10nm bandpass filter and Hoya U-340 4mm as the EO filter leaks light above about 600nm, Hamamatsu LC8 Xenon lamp 254nm - Sirchie 253.7nm filter, UVP 4W 254nm lamp Here's the images. The reflectance of the silver leaf does indeed drop sharply at around 320nm. I recorded the RAW files too, and extracted the channel responses for the Spectralon and the silver leaf. Ratioing the silver leaf against the spectralon gave an interesting graph which closely matched the behaviour shared in the original thread (down to 300nm at least). Simple experiment, but took a while to setup. One of the images (the 360nm one) is a bit blurry, which I didn't noticed until I was processing the images. Also some of the dichroic filters show some evidence of light bouncing around between the layers (see the highlights on the 340nm one especially). Am I 100% confident with the 254nm image being 254nm and not leaks? Yes, pretty much as I've tested this filter before, but I suppose being ultra-picky that is the image and data I am least confident with as it doesn't match the published graph.

I've just had a very geeky few days evaluating a Phase One IQ4 Achromatic camera. Medium format BSI sensor (53.4 mm x 40.0 mm), 151Mp, black and white. Thanks to Teamwork Digital Ltd in the UK for making this possible and sending it to me along with an adapter to use my Hasselblad lenses and a really solid tripod. This was one that really interested me, as it should be good for UV as well as visible and IR, and I could try out my Zeiss UV Sonnar on it, as that was made for the Hasselblad 6x6 cameras. First impressions, it is very solidly built, and very well made. Here's a couple of pictures of the camera with some of the lenses I was trying out (El Nikkor 80mm f5.6, and the Zeiss UV Sonnar). Was able to take some images in UV, visible and IR and thought I'd share a few here. Landscape - Chobham Common in the UK in the IR and visible. Natural light in the evening. IR (Zeiss UV Sonnar, Hoya R72) Visible (Zeiss UV Sonnar, Schott S8612 1.5mm plus 420nm long pass) IR (Zeiss UV Sonnar, Hoya R72) Product shot in visible light using a big softbox. Single Bowens GM500 flash. Flower shots (from local flower shop - Tangerine and Green, Englefield Green, UK), in visible and UV. Single Bowens GM500 flash with quartz tube. Sunflower in UV (El Nikkor 80mm f5.6, Baader U) Not sure on this one - Dahlia perhaps - anyway, a white flower in UV and visible. Visible (Schott S8612 1.5mm plus 420nm long pass) UV (Baader U) These images have obviously all been reduced in resolution for sharing. As an example of the resolution of the original images, below is the image of the Sunflower in the UV along with a region marked in red. That little red square is 1000 pixels by 1000 pixels, and this is what it looks like in the original image. Working with the files is certainly challenging for the computer - a full size image in high quality jpeg is around 60Mb and the raw files are pushing 200Mb. You need a lot of storage with a camera like this. Not seen much UV imaging done with medium format, so thought it would be interesting to share. Unfortunately with only 3 days with it, I barely even learned how to use it, but it certainly impressed me in that short time. It has now sadly gone back to the dealer, and if I want to buy one I'll need to get buying those lottery tickets.....

I have a longstanding interest in local history, and lately this has crossed over a number of times with my interest in multispectral imaging because of the possibilities of revealing seemingly lost information. These possibilities have been well-investigated by the historical community, so I'm not doing anything new and exciting by their standards here, at least from a technical standpoint. One of the better-known examples is Christina Duffy's multispectral imaging of a burnt Magna Carta. This post will be on a commonly used method of combining multiple images from different (sometimes overlapping) spectral bands to extract the text of this advertisement. The method is called Independent Component Analysis (ICA) or Blind Signal Separation (BSS). The main text of the ad is reasonably clear, but there is smaller text that is nearly illegible and it would be nice to recover it. Independent Component Analysis originated in the audio community. The original problem it was meant to solve is known as the "cocktail party problem" — you are at a party and two people are talking at once: how do you figure out what each person is saying? Each ear hears something slightly different (because it is facing a different direction, at a different distance from each speaker, etc.) so your brain can untangle the resulting mess somehow, but what if you want a computer to do it? In the computer version, you have two recordings from different mics (representing your ears) and the computer's task is to spit out the two original audio streams. The way it was solved was to imagine that each recording is a weighted average (linear combination in math-speak) of the original sources, but you don't know the weights. The problem becomes to recover the weights. Different ICA methods take different approaches to finding the weights. The method I used is called fastICA. In the context of image analysis, we imagine that each channel of our multispectral image (not just the R,G, and B, also additional channels for UV and IR) contains some information about the hidden letters, but different colored letters might reflect in different parts of the spectrum. The orange text, for example, is not visible in UV. This means that the ICA algorithm (which is just adding cleverly-chosen weighted sums and differences of the original channels) would be able to subtract off the brick background in principle, making the text easier to read. Could you do this by hand? Technically yes. It is just adding and subtracting channels, after all. But in the current example there are 12 channels coming from 4 images, and determining the correct weights — all 144 of them — by trial and error would be very laborious indeed. The images used here were taken with the following filters/stacks using the Novoflex Noflexar 3.5/35mm: UV- 2mm UG11 and 1.5mm S8612 Visible- Hoya UV/IR Cut IR/vis- TIffen #12 IR- Hoya R72 UV: Visible (for reference): IR/vis (Tiffen #12). This has had the Aerochrome treatment described in the other post on my helicopter flight: IR: The IR does the best by itself in revealing the smaller text, but it does not make it fully readable. Now we run the ICA. If you give the ICA 12 channels (4 photos x 3 channels/photo) then it will give back 12 "independent components" - images that are statistically independent and therefore should hopefully reveal unique information. Here is what you actually get back (with some contrast adjustment): The ICA process is not 100% unique. The ICA components that are revealed will come out randomly inverted (because statistically, changing the sign from + to - does not affect whether a channel is correlated to any of the others). So it is permissible to invert them back to normal. What we see above is that (as predicted) the ICA managed to wipe the text off the wall altogether in some cases, and in others bits of the text remain. The images fall into 4 groups: (1) images showing the main text ("Royal Crown Cola"/"Mansfield Market"), (2) images of plain brick wall, (3) images with bits of the smaller orange text that we are interested in, and (4) one entirely blank image (noise). That last one is because both the Tiffen and the Hoya R72 image contain duplicate infrared info, so it subtracts IR - IR and gets noise. I took an average of each of the first three groups and then put the results in the channels of an Lab file. This was the final result: The smaller text is now readable (barely)! The ads read, MANSFIELD MARKET FRESH KILLED POULTRY MEAT GROCERY FRUITS VEGETABLE (something, probably FRESH) EVERYDAY and Drink ROYAL CROWN COLA (unreadable) References I learned a lot from this review article, and if you are interested in writing your own ICA routine, I recommend it highly, especially for its comments on the pros and cons of different methods: Choi, S., Cichocki, A., Park, H.M. and Lee, S.Y., 2004. Blind Source Separation and Independent Component Analysis: A Review. The wikipedia article on fastICA has a nice overview of that particular method. https://en.wikipedia.org/wiki/FastICA A fairly advanced book on the topic (read the review above before diving into this). Cichocki, A. and Amari, S.I., 2002. Adaptive blind signal and image processing: learning algorithms and applications (Vol. 1). John Wiley & Sons. Christina Duffy's piece on the burned Magna Carta is fun reading. https://www.bl.uk/ma...rnt-magna-carta