|
Yep I can see lots of hot pixels there, the "purple" ones are actually blue pixels and there are also green hot pixels too throughout. the colour of the image is off and the green hot pixels look like sharp stars in your image.
take more blacks and use bias frames too. dont rush things, set outside one night and just shoot any part of the night sky really, take a minimum of 20 light subs, 10 darks, 10 bias and 10 flats. That will give you a basic set of subs to learn the processing, forget about trying to capture a target in particular. That will distract and shape your processing. But combining the captured subs is identical regardless of the starscape, also in how you post process after combining really. the rgb tinge you refer to is sensor noise and some thermal probably. its always in subs really manifesting as an even amount of random noise in the three colour channels. this is in practice the main "Noise" in Signal to Noise ratio and to reduce it you take more subs as the strong signal (ie the picture you want) is present in all frames so when averaged it result in close to 100% of the actual value while the noise which is in three colour channels separately at much lower levels gets averaged down to a lower amount making it much less noticable and thus increasing the Signal to Noise ratio. Its part of why you should never push your background in processing to pure black values as it shouldn't be. The other reason is its not actually pure black ever. Every empty patch of space a telescope has peered closely at contains stars and galaxies no matter how faint. so the blackness of space is really the very very very dark grey of space. Plus photoshop only gives you 8 bits of control and that single bit gap between 0,0,0 (pure black) and 1,1,1 (almost black) often contains a heap of signal which you lose forever in pursuit of eradicating noise (which you cant really do technically).
A common misunderstanding of imaging learners is that more subs means a better brighter picture. This is not the case. It gives you a more accurate signal and increases SNR. Increasing SNR gives you more room to stretch the combined data to then make a better brighter image. You still have to put in effort to to make it happen. But taking those subs (lights darks bias flats) all play their part in giving you a better end SNR but once combined and display on the screen you usually only see a black image because thats whats there UNTIL you start stretching. As you stretch you are amplifying both the Signal as well as the Noise! So combining everything just lets you be able to stretch further before the noise becomes too noticable again. Yes you can get away with making images from only lights but as you progress and encounter limitations you start to appreciate adding darks, bias and flats much more and the more you take of each during your session the better the potential results of your final image will be. Just taking a couple of the others wont do much for your image that you can see initially but later on you will. Especially when you are taking the time to set up and hunt down a target I suggest you also dedicate the time to take about two dozen each of dark, bias and flat frames too. So in a years time you can return to the set you took and be able to make better use of all the data you captured. Start off by considering it an essential part of the capture process instead of only capturing your lights as its a good habit that will help in the long run.
right now your on a good start, more subs will even out that rgb noise background and taking more darks and bias will help those hot pixels to be killed off.
|