Masked Stretch script

[Dimitris Platis]

Has anyone noticed in the Masked Stretch script that :
1. Stretching produces abnormally enhanced circular star cores, instead of smoothed out halos? Any way to fix it?
2. Stretching sometimes pruduces abnormal histograms with overstretched/flattened color channels, while other channels are fine?

[pfile]

i've seen #1, especially after deconvolution. if there was ringing around bright stars that you did not detect in the linear image, the masked stretch can really enhance the ringing.

the only way i've found to fix that is to go back and do a more careful deconvolution. sometimes the script still does this, so doing a smaller stretch with fewer iterations might also help.

on #2, i've not noticed that, but it may be happening...

[kwiechen]

yes, I have seen this especially for bright blue stars. I have used (an ugly) workaround:

Here, I reduced the blue luminance coefficient to prevent artefacts at bright blue stars.


with (rgbws) {
            channels = [  // Y, x, y
                [1.000000, 0.648431, 0.330856],
                [1.000000, 0.321152, 0.597871],
                [0.100000, 0.155886, 0.066044]
            ];

Kai