Closed RFSH closed 3 years ago
After discussing the changes with @karlcz I realized what I'm doing for the filters is completely wrong. Each filter is supposed to change only part of the color value. I was applying all the function to each R G B value, but
So I made the proper changes and also
10 ** (e/100)
formula and used in the existing formula (So it would produce 0.1 to 10 values.)We had another discussion about this and decided that I should generate a intensity (value) histogram as a debugging tool so it's easier to figure out how the filters are working (for brightfield maybe RGB histogram might be better) and Ideally we might want both the before and after histograms.
Eventually we might want to change our filter logic to be based on the displayed image, and therefore we would need to know the min/max/medium intensity of the image. Ideally this should be generated once on load based on the highest resolution of image. But since that's not feasible we might want to do this based on the top of the pyramid image.
Other requested changes are:
We realized that the contrast/brightness are difficult for users. So instead we decided to implement a intensity range. Users can pick the black_level
and white_level
values that will internally map to a slope and offset.
This PR will add tooltips to osd viewer. Summary of changes: