> The line that’s blacked out of such an image is filled in by averaging the readouts from pixels in the columns adjacent to it

I am not sure this would be a feature for a precision instrument. I don’t care about pretty pictures, but I do want that all data I receive is from actual measurements I can trust. A researcher can average columns of pixels later for the press release, but the paper should only count the actual measured ones.