Monochrome Camera / Techniques?

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: Pau, rjlittlefield, ChrisR, Chris S.

ray_parkhurst
Posts: 2752
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Monochrome Camera / Techniques?

Post by ray_parkhurst »

I'm seriously considering going with a monochrome camera to help improve my coin variety detail photography. I have done a search of available cameras, but did not find one that I found immediately suitable. I see the options as:

Bespoke DSLR/MILC
Converted DSLR MILC
Industrial/Scientific

Would like to hear from folks who have monochrome cameras or folks who have checked into the options as to what is available and pros/cons.

Also would like to hear and discuss about other techniques for monochrome workflows.

Lou Jost
Posts: 4592
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Re: Monochrome Camera / Techniques?

Post by Lou Jost »

I have a monochrome Micro Four Thirds camera with cooled sensor. I run it with the sensor below zero centigrade, and this makes a big difference. If you get a monochrome camera, I recommend you get a cooled one. I bought mine several years ago so there are surely better ones out there now.

The huge advantage of monochrome cameras is that you can use telecentric photolithography lenses (typically 5x or 1ox) with extremely high NA and enormous image circles. You would use a monochrome light source matched to the lens, as you know from your background). This is not too hard, and you get an extra resolution boost because you use short wavelength light. As you know, these lenses have always been the sharpest lenses in the world at the time they were built, and though current versions weigh several tons and take up a whole room and cost as much as a house, the 1980's versions are affordable and usable.

You may have trouble with lighting though, as working distance is small. You might be able to solve this though.

Macro_Cosmos
Posts: 958
Joined: Mon Jan 15, 2018 9:23 pm
Location: Sydney

Re: Monochrome Camera / Techniques?

Post by Macro_Cosmos »

My theoretical setup is simple.

Detector/camera
tube lens
filter wheel
objective

This sits on a programmed stackshot, designed to return to origin, switch filters once, then continue the stack.

As Pau correctly pointed out:
Pau wrote:
Thu Sep 03, 2020 1:21 pm
If you are able to place the filters of the wheel into the infinite space between objective and tube lens it will avoid parfocality issues even if their thickness or RI are mismatched
However, I don't think you're able to exploit this. So you're left with rather expensive 52mm lrgb. Parfocal will multiply the cost, as RGB light focuses on a different spot on the optical plane.
Solution: Just do a deeper stack

I think you should consider QHYCCD, their stuff is now really good. The CEO himself is an avid imager, he knows what he's doing and what people want.
https://www.astroungu.com.au/products/qhy163
Something like this, built in filter wheel!

All left is to program the stackshot 3x to somehow trigger the filter wheel for a rotation. I think their product support team can help. They've always been helpful to my requests, even if some are stupidly unreasonable and niche (like using the wemacro rail, a competitor's product with their controller).

I'll leave the rest to those who have done astro work. Is LRGB necessary? Can we get away with just RGB?
Theory tells me that we shouldn't. Monochrome > full spectrum. RGB bandpass filters aren't perfect, they may let in IR and UV light. I use tungsten halogen illuminators, vast amounts of IR. Simply installing 2 UV-IR block filters and a daylight balance filter might not be sufficient. I think this will depend on your lighting ultimately. If your lights are very good with strictly VIS output, then L might be unnecessary.

(All these were thoughts in my head, I haven't found the time to materialise it. I do have an upcoming project that's similar so maybe, it's hyperspectral stuff)

jmc
Posts: 98
Joined: Fri Apr 17, 2020 3:14 am

Re: Monochrome Camera / Techniques?

Post by jmc »

I have a couple of monochrome converted DSLRs - a Nikon d850 and a Canon EOS 5DSR - which were done by MaxMax in the US.

I got these because of my research into UV photography, and removal of the Bayer filter has a significant impact on UV sensitivity. Mine have a simple quartz window over the sensor, and no other internal filtration. I use filters in front of the lens for choosing what wavelengths to image. The choice of internal filter is big one, and let me explain why.

If you get a conversion, be careful about checking with the converter about light leaks, or the effects of any internal IR LED shutter monitor systems. Both of my cameras show some evidence of picking up the internal LED, resulting in fogging of the image. This gets worse with increasing ISO. Whether it is an issue for you will depend on your application. This problem however can go away if you put a filter inside the camera over the sensor to block the IR. Whether or not this is possible for you depends on what you want to use the camera to image. If just visible light, then it is a very easy fix for the problem. Something like Schott BG39 or S8612 would be a good choice for that. I can't do that as I'm interested in imaging UV down to 300nm, and that type of filter would severely limit its usefulness.

I have been able to measure an improvement in resolution for a monochrome camera vs a normal unmodified camera. This is using RAW files and Monochrome2DNG to avoid demosaicing artifacts. It wasn't huge (perhaps a 20% improvement) but it was visible. Tonal range should also be improved, although I have not systematically checked that.

Removing the Bayer filter and microlenses can have a impact on light sensitivity, and how much of an impact depends to some extent on camera sensor design. I cannot go into too much detail on that as I am currently writing an article on monochrome cameras which goes into this in more detail. However with BSI sensors such as the d850, then removal of the Bayer filter/microlens layer improves overall sensitivity in the visible spectrum, while for my 5DSR with a more conventional sensor design, overall sensitivity was about the same as the unmodified camera.

Overall it's a complex area, and not a cheap option, but it does offer some improvements over a normal colour sensor. Diglloyd also has a monochrome d850 and has written about it, but a lot of his stuff is behind a paid firewall.
Jonathan Crowther

ray_parkhurst
Posts: 2752
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Re: Monochrome Camera / Techniques?

Post by ray_parkhurst »

Thanks for the great replies guys.

My motivation for going monochrome was to simply eliminate the Bayer filter so I could take advantage of full resolution of the sensor. I would use white light as I am not trying to get extreme resolution. My output would be traditional black and white style photos with overall FOVs of ~20mm and expected high IQ at 100% for resolutions of 1-2um using the Rayfact 3.5x for 1um and the 95mm Printing-Nikkor for 2um.

I did look into the MaxMax conversions but they are quite expensive, typically ~$6k. I'm hoping I don't have to spend that much since I doubt I'd go down that path.

I took a shot at a few simple techniques to produce a monochrome result. Using the 'monochrome' picture style on my camera was less than rewarding, as it seemed to add extra sharpening, extra contrast, and a "film grain" look to the image which I could not control. Simpy turning the saturation down of course gave the same basic image. Looking at only the G channel was actually OK, so it may be that the color channels are not well-aligned during demosaicing, or perhaps the alignment is "different" than my monitor colors. Are there useful techniques for monochrome output from Bayer cameras that might be more successful?

rjlittlefield
Site Admin
Posts: 21037
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Monochrome Camera / Techniques?

Post by rjlittlefield »

ray_parkhurst wrote:
Fri Sep 04, 2020 7:33 am
Are there useful techniques for monochrome output from Bayer cameras that might be more successful?
For your application, not likely.

dcraw has an option -d (document mode) that skips interpolation and generates output values independently for each pixel position. This option is described as "Good for photographing black-and-white documents", and that part about "black and white documents" is important.

The process relies on having an accurate white balance so that it can compensate for different responsiveness at the R, G, B pixel positions. If the white balance is off, or worse yet if color varies across the image, then you would see the Bayer filter pattern as a bright/dark mosaic. An extreme version of this problem is shown at http://www.photomacrography.net/forum/v ... 987#p84987 .

I don't see any way to hide the presence of a Bayer filter unless color is perfectly constant.

--Rik

Lou Jost
Posts: 4592
Joined: Fri Sep 04, 2015 7:03 am
Location: Ecuador
Contact:

Re: Monochrome Camera / Techniques?

Post by Lou Jost »

I don't see any way to hide the presence of a Bayer filter unless color is perfectly constant.
You can eliminate the Bayer filter's mosaicing effect by using 4-shot pixel shifting camera like the Pentax k-1. That can be treated as a monochrome camera, and the cheapest FF monochrome by far. For macro I think it is a complete substitute. For astonomy this is less clear, since the color filters reduce light (but the microlenses are still present, unlike in converted cameras, so that partially compensates).

rjlittlefield
Site Admin
Posts: 21037
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Monochrome Camera / Techniques?

Post by rjlittlefield »

Lou Jost wrote:
Fri Sep 04, 2020 11:19 am
I don't see any way to hide the presence of a Bayer filter unless color is perfectly constant.
You can eliminate the Bayer filter's mosaicing effect by using 4-shot pixel shifting camera like the Pentax k-1.
Yes, good point. My comment applied only to a single exposure.

--Rik

ray_parkhurst
Posts: 2752
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Re: Monochrome Camera / Techniques?

Post by ray_parkhurst »

I shot a full Cent at m=0.7 and f4 using a new lens I got from Robert, a Bogen 60mm WA. It produces almost a reference-quality full-coin image. The full coin image is shown below, along with a 100% crop around the die break between B and E of LIBERTY. This is known as "BIE" and in fact this particular BIE is the largest known one of the Lincoln Series. Most look really more like an "I".

I took the 100% crop and reduced it to a B/W image with much increased contrast to sort of look like a simplified line drawing. This is actually more detail than would be shown in the early attribution guides, but it gets the idea across. I was just curious if I'd be able to create something like the historical drawings from a photo and indeed it's not too difficult. Something like this may be useful as an adjunct to the higher quality images needed for attribution, and indeed this may actually be more useful for making overlays than even a monochrome image.

IMG_0001_1_1.JPG
IMG_0001_2_2.JPG
IMG_0001_2_3.JPG

rjlittlefield
Site Admin
Posts: 21037
Joined: Tue Aug 01, 2006 8:34 am
Location: Richland, Washington State, USA
Contact:

Re: Monochrome Camera / Techniques?

Post by rjlittlefield »

Is this from your HR T2i, so, 18 megapixel APS-C sensor with AA filter removed?

--Rik

ray_parkhurst
Posts: 2752
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Re: Monochrome Camera / Techniques?

Post by ray_parkhurst »

rjlittlefield wrote:
Fri Sep 04, 2020 5:57 pm
Is this from your HR T2i, so, 18 megapixel APS-C sensor with AA filter removed?

--Rik
Yes, EA6.8, no AA filter.

Macro_Cosmos
Posts: 958
Joined: Mon Jan 15, 2018 9:23 pm
Location: Sydney

Re: Monochrome Camera / Techniques?

Post by Macro_Cosmos »

Lighting is impeccable on that coin. Simply amazing. I like the gradations.

ray_parkhurst
Posts: 2752
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Re: Monochrome Camera / Techniques?

Post by ray_parkhurst »

Macro_Cosmos wrote:
Sat Sep 05, 2020 6:04 am
Lighting is impeccable on that coin. Simply amazing. I like the gradations.
Thank you! It is partly due to the lens and the specific lighting angle it allows due to its focal length and working distance, and is why I mentioned that this is an "almost reference-quality" image. Everything comes together nicely with this lens.

dolmadis
Posts: 737
Joined: Wed Dec 07, 2011 1:51 pm
Location: UK

Re: Monochrome Camera / Techniques?

Post by dolmadis »

ray_parkhurst wrote:
Sat Sep 05, 2020 8:06 am
Macro_Cosmos wrote:
Sat Sep 05, 2020 6:04 am
Lighting is impeccable on that coin. Simply amazing. I like the gradations.
Thank you! It is partly due to the lens and the specific lighting angle it allows due to its focal length and working distance, and is why I mentioned that this is an "almost reference-quality" image. Everything comes together nicely with this lens.
Hi Ray

Can i ask please how you set up the lighting for that image and what lighting and diffusion you used?

Did you mount it direct to a body? With extensions? FL and WD is also of interest.

It looks great !!

Thanks

John

ray_parkhurst
Posts: 2752
Joined: Sat Nov 20, 2010 10:40 am
Location: Santa Clara, CA, USA
Contact:

Re: Monochrome Camera / Techniques?

Post by ray_parkhurst »

dolmadis wrote:
Sat Sep 05, 2020 9:17 am
Can i ask please how you set up the lighting for that image and what lighting and diffusion you used?

Did you mount it direct to a body? With extensions? FL and WD is also of interest.
I use masked ringlights for most of my full-coin shots. The ringlights are from "Angel Eyes" and are 12V driven, intended for automotive use. In this case I used a 50mm diameter light. This places the light just outside the lens front element so it is close to the optical path as possible without causing any blocking. I mask the light to keep it from shining directly toward the lens to avoid flare issues. I also mask it to "shape" the light source. The unmasked parts of the ringlight are from 1:00-3:00 and 9:00-11:00. This spread of angles minimizes the need for diffusion, and indeed I did not use any.

The lens FL is 60mm and is mounted on a Pentax Auto Bellows. At m=0.7 the total extension from sensor to mounting plane is 115mm while WD is 120mm. The large WD and high lighting angle vs horizontal gives nearly axial lighting and creates the particular "look" of the image, with most light shining on the top surfaces of the device features, and very little on their sides. This creates a nice, graded shadow surrounding the features.

Post Reply Previous topicNext topic