0

I am going to take a few photos of hardware for documentary purposes; unfortunately the only light available is monochromatic (think of low pressure sodium-vapor lamp). This is not really a big problem as the images should be in black/white anyways for easy printout in a manual.

This got me thinking about the DNG => monochrome image process.

In a normal develop process (on the technical level) you would take the DNG, demosaic it by interpolating the two missing colors per pixel. Then you filter/desaturate these pixels by doing some calculations with all three color values.

This inevitably mixes some information of all the sensor readings from surrounding pixels into one, which I think is not necessary to reconstruct a black/white image from the readings when taken under (almost) monochromatic light.

What I imagine can be done instead is: Take each sensor reading, multiply it with a factor (or maybe something logarithmic must be done) that corresponds to its sensitivity for the given monochromatic light source. With this you should be able to reconstruct a black/white image with more details.

Is there a program available that can do that (windows or linux is fine)? Or can I simulate that somehow with lightroom/adobe raw?

(please also tell me if this idea is totally nonsense and this would not at all preserve a little bit more detail)

2

You don't need to go to the extremes of attempting your own demosaicing, doing so would be a lot of work for no real benefit — all good raw converters use an algorithm that attempts to identify and exploit regions where the hue doesn't change in order to maximise the amount of detail recovered.

In any case unless you are reproducing images at 1:1 magnification, the quality of demosaicing will have very little effect on the result.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.