I'm new to the area but I was wondering how can one get the values needed to apply a radiometric correction to any image taken from the same sensor having some dark and flat field images.
From what I read the dark field is the bias value and the gain value is the flat field minus the bias. Considering that I want a value from these images that I can apply to any image taken by the sensor I am guessing I can do the mean for the gain and bias values and use them. Also the formula I'm using to get the radiometric correction is:
Radiance = (Bias+PixelValue)/Gain
as @christoph mentioned
Is this right?
I have been spending some time on this problem the last few days and from what I can gather, calculating the radiance of a pixel looks something like this:
radiance = (DN x gain) + bias
where DN is the pixel intensity, gain is described below, and bias is the dark frame. I pulled this conversion from Matlab: https://www.mathworks.com/help/images/ref/dn2radiance.html
According to Wikipedia for flat field you can calculate the gain by dividing the average of the flat field - dark frame by the flat field - dark frame:
m = image averaged value of (F - D)
gain = m/(F - D)
I am not an expert, this is just what I have found so far.
Some other documents to look into:
For a very in-depth look at understanding the math behind intensity mapping: https://cave.cs.columbia.edu/old/publications/pdfs/Grossberg_PAMI03.pdf
Short overview and code implementation: https://gimoonnam.github.io/imageprocessing/Debevec_HDR/
For conversion to luminance (basically the subjective version of radiance): https://www.atecorp.com/atecorp/media/pdfs/data-sheets/tektronix-j16_application.pdf
Please update if you find out anything else!