I like it, I think that more data could certainly help, but the other consideration that you could try is that 20x5min Ha is plenty of Ha, 19x5min OIII is probably enough too.
You could attempt doing a HOO image, and not utilizing the SII data incase that is in fact the cause of the spottyness...
Trying to pull insufficient data up to meet sufficient data will always result in some artifacts...
So a HOO blend, a synthetic SHO using (0.6*Ha)+(0.4*OIII) as S, or even taking (0.7*SII) + (0.3*Ha) as the SII channel may well clear it up.
|