Digital sensor performance curves

[DaveO]

I recently asked one of my contacts in the RPS Imaging Science group if he knew of any published work on UV absorption of the RGB dyes used in Bayer arrays and he pointed me to the website of Truesense Imaging Inc.

 

Here's a link to a document about a modern 6576 x 4384 pixel CCD

http://www.truesense...hy/47-KAI-29050

 

pg 14 shows response of a monochrome array and on pg 15 with a Bayer array, but alas only 400 - 1100 nm

pg 37 shows cover glass transmission which does a dive at about 325 nm

 

 

and one to some reference material

 

http://www.truesense...rence-documents

 

Then choose

 

Image Sensor Color Correction

 

pg 3 shows the spectral response of a CMOS sensor without a Bayer arrray over the region 350 - 1050 nm then on pg 4 is the same sensor with Bayer array so you can clearly see the miniscule response that we use below 400 nm with the red being the largest.

[colinbm]

Interesting, thanks Dave

For the small bit I understand, I am surprised the cover glass got to 325nm at 50%, yet the CFA with microlenses only got to 400nm at 50% with the blue dye ??

Yet the panchromatic was able to get to 340nm at 50%, with microlenses.

I haven't seen any results from anyone using one of MaxMax's, sans CFA & microlens, UV cameras ?

Cheers

Col

[Alex H]

I haven't seen any results from anyone using one of MaxMax's, sans CFA & microlens, UV cameras ?

Cheers

Col

 

What differences do you expect between MaxMax's and anyone else's monochrome-converted (sans CFA & microlenses) cameras?

[colinbm]

Thanks Alex

What I guess I met was the MaxMax UV dSLR's conversions rather then a CCD machine vision camera, with the increase in UV depth.

Is mono not better for the extra UV transmittance or is the preference to false colour UV, as against depth of UV ?

Col

[DaveO]

Here's another interesting paper

 

http://www.aphesa.com/downloads/download2.php?id=1

 

Dave

[JCDowdy]

... then on pg 4 is the same sensor with Bayer array so you can clearly see the miniscule response that we use below 400 nm with the red being the largest.

 

This is consistent with what I have seen elsewhere but the blue channel, nearest the UV-A has the greatest response.

 

For example - https://www.maxmax.c...al_response.htm

I don't know if they ever formally published this, the methodology is not fully described on the website.

 

and here: http://forum.mflenses.com/userpix/20114/455_KODAK_KLI14403_spectral_response_color_1.jpg

 

and here: http://i.stack.imgur.com/T9Np9.jpg

 

- John

[Andrea B.]

The max-max write-up says that UV was not tested for because the monochromator could not output below 375nm.

And it's not clear whether they were testing with or without the internal blocking filters?

Those charts kinda make you think that your UV photos should record a lot in the blue channel, but they don't.

 

Shane did a nice test of the D70 a long time ago: http://www.beyondvis...UVNikonD70.html

[Shane]

When I corresponded with Dan (Maxmax) several years ago it was determined that his light source was the limiting factor.

 

Response in the blue channel is dependent on many factors related to both semiconductor construction (materials and structural design) as well as overlying modifications e.g. sensor cover glass, internal cut/blocking filters and Bayer dyes etc. While the same sensor (eg Sony) may be utilized by different companies (Nikon, Sony, Pentax etc), the overlying modifications are DSLR manufacturer specific. At that point the application of different Bayer dyes can have different effects on the blue channel response. The Nikon D70 exhibits blue channel response to just below 400nm but quickly switches to red response by ~385nm. The Fuji S3Pro predominantly exhibits a strong R channel response below ~385nm but has better B channel than the Nikon D70.

[JCDowdy]

Alas, the methods are nowhere close to being summarized much less well described and it is not always clear if responses are for modified or unmodified cameras. The MAXMAX curves were for modified cameras, I assume without lenses, but I failed to see the light source or type of monochromator indicated.

 

Shane, I have seen your earlier work and would have certianly cited it also. I simply did not recall it at the time and just pointed out a few quick examples. May I ask what type of monochromator you used?

Thanks,

- John

[DaveO]

As you may have guessed by now, my original contact used to work for Kodak in the UK so the first refs were from the company that was spun off from the big yellow box. I'm happy to try to follow up any queries which you may have

 

Dave

[Shane]

May I ask what type of monochromator you used?

 

I used the excitation "side" an Aminco AB2 Spectrofluorometer.

[JCDowdy]

I used the excitation "side" an Aminco AB2 Spectrofluorometer.

That's a nice fully featured spectrofluorometer, although a bit long in the tooth by now. But hey, I have some instruments in my lab dating from the '80s! :) The most challenging part of keeping them going is the instrument control. Correct me if I am wrong but that spectrofluorometer has a single grating monochromator doesn't it?

[Shane]

although a bit long in the tooth by now

 

Although still one of the best performing spectrofluorometers and still produced and supported today by Thermofisher. Thermo was going to drop this system when they purchased American Instrument Company but due to outcry from the industry users (myself included at the time) because there was no other system with performance that could beat the Aminco, they decided to keep production going.

 

It's been 6+ years since I used one but I think it was single grating.