How do I create and apply a vignetting correction mask?

I am trying to correct for vignetting that appears in High Dynamic Range Images. In order to do so, I need to create some sort of vignetting correction mask that can be applied to the image in order to counteract the light falloff at the periphery of the image. I have an image of a uniform luminance surface that shows pixel vignetting, and the idea is to divide the entire image by the max pixel value to create an image where each pixel value is a ratio of the max value. In theory, if I divide the vignetted image by the newly created "ratio image", then the vignetting would be corrected. From reading other posts and answers, I am assuming I would have to convert the image to grayscale so that I am dealing with only one pixel value instead of the RGB values. If there is a way to do it without converting it to grayscale then that would be preferred. I know basic commands in matlab so any help would be greatly appreciated.

[Josh]

I know that vignetting is also depends on aperture size, and I have all of the .jpeg and .RAW files used to compile the HDRI.

Answer 1

[Image Analyst]

No you do not have to convert to grayscale, but you do get a grayscale image because you do it one one color channel at a time. Essentially this:

% Divide the actual image by the modeled image.
correctedColorChannel = single(oneColorChannel) ./ single(noiselessBackgroundImage);

which is in the attached file. But first you have to create the model background image, which is created from your "blank shot" by the other attached file.

[Josh]

Thank you for your quick response. I will try this and let you know how it works.