Видео

Thanks Hermann. Glad you found it helpful.

I asked Mike for the best way to uninstall v1. I'll post again when I find out myself - I think I simply deleted it.

I absolutely agree :-) No intuitive way to remove it, no info on that nowhere.

Check out the SJAA videos Parts A ( ruclips.net/video/dGXcWVvHKf0/видео.html) and B (ruclips.net/video/5t9h3VfpWR4/видео.html) available via the ghsastro.co.uk site. They are the most extensive instructions that have been created, and using the latest process/module Pixinsight version. They probably contain more information that you even want. At the same time, I encourage you to share any learnings/insights you might have gained in trying the tool, via one of the astrobin.com or other forum series. Thanks, Dave

My apologies - this was my first video... I will replace this one when time permits

Yes, my apologies. I will replace and update when time permits. Please visit ghsastro.co.uk for more recent ones with better sound quality.

HI Jane, and thanks for your suggestion. We have been going a different route with GHS - image blending was only included here to temper the colour stretch - although as you point out, it can be use to blend any two images. I often do what you suggest, but I use Pixelmath instead - where you can say new image = a * image1 + (1-a) * image 2. With masking or if statements you can also make it apply wherever you like. I do this for sharpening, HDRMT, etc to temper these processes too. Hope this helps and thanks for your comment.

Thanks Jane.

Hi Boswix. Glad you like. Your idea is a good one, but I think the script does put in a generic name if you leave the input blank. If the name exists already, it puts in a number at the end.

HI Clifford. Funny, but I never actually finished processing this one...and I am not sure what I had done to calibrate the colours. My normal practice is a three steps. 1) Linear Stage: My normal practice (now) is to use GHS to line up the histogram peaks - in a SHO image that generally means stretching red/blue channel - this still leaves the image quite green, and essentially still linear. The Green channel likely has a larger tail so I also try and smear out the red/blue by stretching out with a substantial b, and bring it them back to the green peak using HP. (Even though technically this makes the image non-linear - I don't worry about it) - If I am using an extracted luminance, after this step is when I extract it. 2) I remove the stars typically and then do an initial stretch to non-linear with GHS using RGB, but you can also leave the stars in. - the resulting image will still be green. However, after the initial stretch I will then stretch the red and blue more, individually. This will bring more colours into the image reds, yellows, aquas, and blues - but still leave some green. It takes a little practice, but you don't want to spoil the colour balances, so I employ HP and LP along with D, SP, and B at this stage. 3) Finally I use the method of Cuiv the lazy Geek ruclips.net/video/UzL_UwcqkfI/видео.html to remove as much remaining green as I want via LRGB combination and reducing green contribution to the colour, and the stretched L I extracted before. I also remove any magentas by invert-->SCNR-->invert method as CUIV outlines. The other method, is of course, just to use SCNR to remove green, but this is far less satisfactory. Hope this helps and sorry I can't remember what I did with the NA nebula. Dave

Glad it helps. In my opinion, it takes a little time investment to master GHS, but in the end I think it really helps the images. Linear image just means that the pixel value/intensity is linearly proportional to the photon capture or signal. Unfortunately, our eyes don't perceive light linearly - it's an interesting exercise to view an image with various amounts of linear stretch only just to see how your perception of the image is hand-cuffed if you restricted yourself to linear transformations.

Thanks Robert. Mike has added many shortcuts in v2.1 that minimize numerical typing too, that I failed to show. Mike and I are also scheduled for TAIC youtube/webcast on May 29, 2022 if you are interested.

Thanks Robert

Thank you! Cheers!

Hi Brian, video 3 has your galaxy(ies). M51 is the image. Hope this helps.

Thanks Rick. While Mike and I work together on the design and how it should work, the actual making it a reality is Mike's work. (I help if the code calls for Fortran IV).

Thanks Rick

Thanks Mark

Thanks again, Des

HI Calzune, Try to determine the noise level versus the faint nebulosity. Between these levels is where you want to place the most contrast - ie separate the nebulosity from the noise. ie. put SP between signal and noise. If there is no or little brightness separation between these, then stretching alone won't be able to distinguish between them. Note that stretching only considers the brightness of the pixel itself, and does not consider the value of surrounding pixels. Stretching with/without a mask will only help if the mask somehow masks the noise and not the signal. To really help with this, you may have to apply Mure noise reduction, deconvolution, or a wavelet (Fourier transform space) - something that takes both the pixel value itself as well as surrounding pixels into account to try and distinguish between noise and weak signal. ie reduce the noise while trying to leave/emphasize the signal. The other option is more integration time, or course - to average out random noise, and average in consistent signal. Hope this helps

Thanks for the feedback, Dale. A couple more to make.

Yes, that is indeed what is going on. My apologies. (you might be able to tell that creating you-tube videos is a little bit out of my box) Should be fixed goint forward, and I may come back and re-record. Glad you enjoyed and it was helpful though.

@@davidpayne8646 If you want, you can send me the video file and I can clean the buzz sound and send it back to you. There is no need to re-record it again. Second video is great and I think you are already getting the hang of it. This GHS is a game changer for me so your videos are really helpful.

@ Sounds like a great idea. Tell me how, and I'll take you up on it.

Thanks Des, Mike's work on v2 makes it seemlessly integrate with mine too.

I am Stan?

Thanks Bill. I am so glad you like it. The pixelmath option was my first attempt, but I find the script much easier. Mike (my collaborator) did the heavy lifting to turn it into the script.

Thanks for the interest. I hope it works for you.