Adam, you hit the nail on the head in breaking down the mechanics of what is being done to the image with SPCC. As usual, you have IMHO one of the best presentation approaches for PI out there.
Brilliant, Adam. Just brilliant. I don’t understand why there has been so much negativity floating around about SPCC and how it works and also the changes to PCC (usually the moan about having to has a astrometric solution now. I understand why it was removed, the solution is a property of the image and not of colour correction - it is required for colour correction). This video really shows why this is a better method and how it should result in a more accurate corrected image. I think you are the first person I have seen talking about SPCC in a positive way, and it is a very positive step and the developers need all our thanks.
Thank you very much for this detailed, informal and high quality tutorial on SPCC. The background information on GAIA and the information on how SPCC uses the data from GAIA is great. I have already worked with SPCC and I am thrilled. What is not clear to me is how to calibrate images taken with a Duo-Narrowband filter (Optolong L-eXtreme). Does that make any sense at all to calibrate HOO images? Thanks again for this great work, Adam!!!! 👏👏👍
Awesome! And thanks for taking the time to explain. Question: ASI 294MC Pro, Radian Triad Ultra Quadband, do I select narrowband filters mode? I don't get good graphs when I do. thx..
My SPCC has Wavelength & Bandwidth boxes for the RGB filters, not a way to select filters. And looking forward to seeing updates on the Narrowband filters mode. Thanks and a wonderful video.
Really great stuff as usual Adam, thank you. I'm wondering... how would this be applied to a narrowband workflow, SHO for instance? It's unclear what the Narrowband check box does..?
I am going to get some clarification... but SPCC calibrates super accurately the flux ratios using GAIA as a reference and your own equipment (filters). So what you get back from SPCC are the *true* strengths of each NB channel. This isn't going to make a pretty picture though...since the standard procedure is to boost the weaker channels. But... your initial result is astrophysicaly faithful. Does this answer your question>?
Really awesome video. I’m learning to use PI and, as a scientist myself, I really appreciate the in depth discussions about how and why things work. My only minor frustration with SPCC is that right now I am using a Canon DSLR (as I develop my skills) and it is not clear what the best settings are for sensor and filters. But the capabilities of the process and database are astounding!
Adam, These 2 videos are awesome! Great work! I've learned so many things from you (here or on AdamBlock Studio), thank you! I really appreciate how you go deeply into processes, it helps a lot to understand "what am I doing". You've defenitely contributed to improve my skill on PI (I still have tons of things to learn though) Just a small question: why "pro" camera has so big pixel size (I guess instrument on JWST have 17µ pixel size; that's huge)? I would bet that the smaller they are, the better it is but that's not case! Maybe a matter of noise/chipset size constraint? (small pixels => more pixel => more circuit => more (read)noise)? Thanks again for all your work. Regards;
In general "pros" want to collect as much light as possible to improve the S/N. Resolution isn't necessary the priority and nor is worrying about saturation of non-related things (like stars). JWST is a different consideration. JSWT observes at the Infrared wavelengths of light. 17micro pixels ARE SMALL for this wavelength!
Thanks Adam for all your content. In another video, the user suggested changing the white balance reference when using filters such as Optolong L-Extreme. Instead they used the white reference as 'photon flux' and not the 'average spiral galaxy' Can you elaborate on why this would be done, when using eg: ASI294MC Pro with an Optolong L Extreme filter ?
That is a longer explanation (that I have on my site). The issue is that these isn't a white balance for narrowband. In fact, all SPCC gives you is flux calibrated answers- but this isn't color determinate.
Excellent!! Questions: I’ve been tunning DBE AFTER doing SPCC, does it matter? It does seem like when I run DBE now I’m getting better results. But I’ll change it to before SPCC if that is optimal.
Yes, DBE after SPCC can technically change the color balance (depending on how you do it). So the optimal order is DBE first followed by SPCC. At the end of the day... it should make a big difference. Just remember to use the background neutralization of SPCC correctly and you should be OK.
Thank you for an excellent and educational presentation. Appreciate the background information which really helps our understanding of SPCC and the advancement of both precision and accuracy. Having a background in Metrology, those are important, distinct and well chosen words. Well done Adam and Thank You!
Hi Adam, great video, very informative and going a bit deeper into the physics and mathematics behind it than usual “click here and then there” tutorials! I have one question which arose at 29:30 in the video. When I plug the R/G x value into eq. 1, I get the y value shown there (0.815758). But if I plug the B/G x value into eq. 2, I get y = 0.7873 and not the shown y = 0.68755. But 0.7873 is the G white balance factor… I don’t get this sorted. Am I thinking wrong? Regards, Bernd Edit: Found the problem. The R/G Whitepoint x is transformed by the R/G linear fit, which is correct. But the B/G Whitepoint is transformed by the R/G linear fit as well instead of the B/G linear fit! That leads to 0.68755, which is indeed wrong. Correct is 0.7873, which you get when using the correct B/G linear fit with the B/G Whitepoint. But the white balance factors are correct again. These are normalized by the maximum of the inverses of the R/G, 1, and B/G image factors, and there we find the correct value 0.7873 for G (which is 1/(1/B/G) = B/G. So there seems to be a bug in the log output, but not in the final result.
Correct. You will note that I discovered this bug on my own and made Edoardo aware of it last week. I did my due diligence Bernd. However... I did not want to highlight the bug since it will be fixed.
@@AdamBlock Adam, absolutely no offense! I tried to reproduce the calculations to see if I understood it or not. This is a great video (as usual), so thanks for sharing!
Adam, thank you for this video. It is the first video I've watched that makes SPCC understandable rather than just a fill in the blanks and run the programme situation. I do have a question with regard to filters. I use an Atik Horizon II OSC with Hyperstar on my C11. I also use an IDAS LPS-D2 or an Optolong L-Extreme in front of my camera. As these introduce another filter in addition to the GRBG Horizon II chip filters I'm wondering what impact this would have, if any, on SPCC results and, if so, how would this be accommodated? Kind regards, Gord
For the light pollution filter...I do not think there will be a big effect (but, I am not an expert on this). For the NarrowBand- there is some thought as to what your goal is. For NB there isn't a color blaance- SPCC will just give you the signal strengths for each emission line for your system. But usually people stretch NB data and do not maintain this relationship.
@@AdamBlock Thanks Adam, That was my thought as well. I don’t know for sure but I expect that given the LP filter transmission lines there would only be a small, if any, effect. Cheers, Gord
Adam, I really appreciate your videos. I have one question I hope you will be able to comment on. Within the first minute of the official PixInsight video, "SPCC - Working with Color Sensors" they say that when working with color sensors, we should always use the Ideal QE curve setting. In this video of yours, between 25 and 26 minutes in, you select the particular Sony sensor from the OSC camera that was used in the example image that you are working with. It seems to me that if my sensor is listed in SPCC I should select it (as you did in your example) but if that sensor is in a OSC camera, do you think I should be selecting the Ideal QE curve instead?
Yeah, one of several errors. You have to remember I made this video very shortly after the process became available... no oter content creator nor even the developers attempted to go into this detail at the time. There is risk involved in creating this stuff... e.g getting something wrong... I stepped up! So the answer is, for many OSC sensors the filter specifications are not known. Usually you specify the filter transmission ANd the sensor sensitivity at given wavelengths. (These values are just mulitplied/convolved together). But in the case of OSC, the filter values are not know. So you should specify "1" for the Ideal QE and then just specify the sensor you are using. This folds the total transmission and sensitivity together into one parameter since you do not have more numbers to work with.
@@AdamBlock Thanks so much for the quick reply. Please allow me to ask a follow on question. If I’m using a duo narrowband filter with my OSC camera, should I select the Ideal QE curve for the sensor and the specific filter used in each of the R, G, and B filter fields?
@@kennethklock7419 That is more complicated. You are effectively using two filters. In addition, you would be using SPCC in the "Narrowband mode" to give you calibrated fluxes..but not calibrated colors. For NB imagery... calibrated fluxes may or may not be critical. Many people adjust the signal strengths in such a way they are far from what physics determines... so I think you should look things up on the official forum. This is not an area of strength for me (I actually find it perplexing as to why it is done.)
Awesome video. Do you know if SPCC background neutralization (and/or balance) is local or global? I.e. does it fit whole image to curve, or for each pixel of image it looks at stars (weighted by distance, I guess) and applies that? What I'm trying to figure out is whether you need ABE/DBE at all after SPCC (if it's local, I think there should be no need?).
Your video made me think. A process like SPCC makes it possible for amateur images to be relevant in a scientific way, maybe. I'm just wondering how the PI workflow would look like to make amateur images useful.
This was a great explanation Adam..thank you!!!! After working with Pixinsight for the better part of a year now and then coming back to watch your videos, I'm having a lot of "AH-HA!!!" moments of revelation, which I suppose is rather like putting the cart before the horse, but the hands on experience makes the theory a little easier to understand. That in turn helps immensely when I start work on a new project. More specifically concerning SPCC, as someone else posted here, I think the "change" took everyone by surprise, but your explanation makes the entire process far less intimidating knowing what its potential is. Keep up the great work!!
I love your enthusiasm as well as the well-done explanation for the science behind SPCC! wow how incredible. Very inspired and thanks for these 43 minutes of my day :)
Thank you very much for watching. This video has not been vetted by PixInsight developers... I am prepared to make an errata video...but I haven't heard any complaints since publishing it...so I suspect I am close with my explanation.
Thank you so much for this information! This level of precision make me humble that this is now available to me along with the technology and teachers like you! I can't wait to process and maybe reprocess some of my data just to see the results! Would you suggest doing a DBE before or after SPCC? I am a student of your course but I could never figure that out.
I’m a little lost when it comes to calculating the Catalog ratios for every star. It seems like there is only one ratio for R/G and B/G that we can get from the catalog (which are the ideal white balance ratios). How do you get the individual catalog ratio for every star so it’s not just a vertical line on the plot?
I think you might be mixing things. When you compare your stars to the catalog you are measuring a ratio of instrument response (your sensor versus GAIA data). The ratio for each star (of whatever color) will be about the same... but there will be scatter...so you do a linear fit and get a line. The slope of the line is the ratio. With this in hand (a correction)... you can then apply a white balance to your image. I haven't been thinking about this for a while... but this is my quick answer.
@@AdamBlock Yeah, I was definitely mixing things up. I was confusing the ideal white balance flux ratio with the individual star flux ratios. Must've been a long day or something... My only question now is that once I normalize the "y" values shown in the video, I get white balance factors of .8428/.6876/1 instead of .9651/.7873/1. Is there something else happening in the background or am I failing at algebra?
@@SeanDWalker4 Hmm...When it was originally released I did discover an error in that the output was correct...but the equation values were not. I thought everything was fixed as far as I know. I would need to see your particular example to know and check it. Run the values through the equations and make certain they come up with the output ratio- then you can normalize however you would like. I not certain how SPCC chooses a particular channel to be "1" ... but it doesn't matter.
@@AdamBlock I was just looking at the values on this video since I was curious. Once I get to processing more of my own images I'll have to see if the numbers makes sense. Thanks!
Oh my...indeed! I think you found the same error! Yes, it should be correct now. Good job. I thought I was special by identifying the issue when it was released. I couldn't wait for a updated version before making this video..so the error is recorded here...though I doubted anyone would ever notice. I guess it took a year... :)
Great explanation of this great development which brings science closer to the amateur photographer! It’s astonishing to see the possibilities of thus development. And it brings the discussion of what is the real color of the color you see to a next level
Since we know the true spectrum of most stars now, can't we just take very short exposures of the stars in the target, and rely on SPCC to get the colors right. Then we can take longer exposure of nebulae, galaxies, etc. without going back for several hours of integration of RGB stars? Thanks, John
@@AdamBlock If we are now following the workflow of removing stars, processing the starless target, and later adding back minimalized or morphed down RGB stars, then time required to get a good spectral representation of the stars used should only be a short time, right? It is no longer necessary to get an hour each through RGB filters for good stars to PixelMath back in. Is SPCC now filling in the full spectrum that might be missed with longer integrations? Is my thinking correct here?
@@johnadastra1754 I think there may be a misunderstanding at what SPCC is doing. Stars are removed from the same image from which the nebula (or whatever) exists. It sounds like you are asking if you can create a starless image from a long exposure (which HAS LONG DURATION EXPOSED STARS) and also take a short exposure RGB and add that into the starless version. #1 SPCC will give you the same color calibration/white balance for the long exposure star image as the short exposure one. #2 You never did have to spend an hour through each RGB filter.... other than getting as much information as necessary. SPCC doesn't affect this - there will be the same answer. We are likely not understanding each other.
@@AdamBlock Thanks Adam. Still working on exploring the full potential of this new tool. I was talking about plugging RGB stars back into a NB image. I guess I was trying to see if SPCC would lessen the amount of integration time on the stars I would use for this. Will keep trying to see how these processes are benefiting my workflow. Seems like helpful addition to PI.
@@johnadastra1754 I don't think SPCC directly affects lessening the exposure time for RGB stars that you might insert elsewhere. It will just make certain your stars..whatever exposure time, are colored accurately.
![[Live] ศึกมหึมันส์มวยไทยสังเวียนเดือด เวทีมวยชั่วคราวสวนสยาม | จันทร์ 30 กันยายน 2567](http://i.ytimg.com/vi/pM1k6f7JXUc/mqdefault.jpg)
![[UNCUT] “บอสณวัฒน์” ลากไส้!! ใครหนุนหลัง "แม่ตั๊ก ป๋าเบียร์" I คนดังนั่งเคลียร์ I 30 ก.ย. 67](http://i.ytimg.com/vi/QlD7mU2KKRk/mqdefault.jpg)
Adam, you hit the nail on the head in breaking down the mechanics of what is being done to the image with SPCC. As usual, you have IMHO one of the best presentation approaches for PI out there.
Thank you Rob... I appreciate that!
Brilliant, Adam. Just brilliant. I don’t understand why there has been so much negativity floating around about SPCC and how it works and also the changes to PCC (usually the moan about having to has a astrometric solution now. I understand why it was removed, the solution is a property of the image and not of colour correction - it is required for colour correction). This video really shows why this is a better method and how it should result in a more accurate corrected image. I think you are the first person I have seen talking about SPCC in a positive way, and it is a very positive step and the developers need all our thanks.
Thank you very much.Yes...I wanted to reframe the way people were thinking about this process,
Thank you very much for this detailed, informal and high quality tutorial on SPCC. The background information on GAIA and the information on how SPCC uses the data from GAIA is great. I have already worked with SPCC and I am thrilled. What is not clear to me is how to calibrate images taken with a Duo-Narrowband filter (Optolong L-eXtreme). Does that make any sense at all to calibrate HOO images?
Thanks again for this great work, Adam!!!! 👏👏👍
I have the same question although using the similar Antlia ALP-T. Don't want to bother Adam though so will just have to wait !
Adam, your videos are very well done and educational! Thank you for going into the process behind color calibration. It is very illuminating.
Thank you
Thank you very much, Adam.
Is there a difference on how SPCC works with OSC cameras?
Regards,
Alfredo
Not really- whether the filters cover the pixels or the entire chip really doesn't change the analysis.
@@AdamBlock Thank you
Awesome! And thanks for taking the time to explain. Question: ASI 294MC Pro, Radian Triad Ultra Quadband, do I select narrowband filters mode? I don't get good graphs when I do. thx..
My SPCC has Wavelength & Bandwidth boxes for the RGB filters, not a way to select filters. And looking forward to seeing updates on the Narrowband filters mode. Thanks and a wonderful video.
Really great stuff as usual Adam, thank you. I'm wondering... how would this be applied to a narrowband workflow, SHO for instance? It's unclear what the Narrowband check box does..?
I am going to get some clarification... but SPCC calibrates super accurately the flux ratios using GAIA as a reference and your own equipment (filters). So what you get back from SPCC are the *true* strengths of each NB channel. This isn't going to make a pretty picture though...since the standard procedure is to boost the weaker channels. But... your initial result is astrophysicaly faithful. Does this answer your question>?
I was wondering the same thing... I have OSC (ZWO ASI294MC Pro) and use the Optolong L-eXtreme filter).
Hopefully you can make a part 3 and discuss the use of SPCC when using narrowband filters?
Perhaps... likely Part 3 will literally be me fixing some errors that the developers will likely "inform" me of. :)
Gotta love the application of advanced calculus to astrophotography!
Really awesome video. I’m learning to use PI and, as a scientist myself, I really appreciate the in depth discussions about how and why things work. My only minor frustration with SPCC is that right now I am using a Canon DSLR (as I develop my skills) and it is not clear what the best settings are for sensor and filters. But the capabilities of the process and database are astounding!
Bravo Adam! This is content on an academic level and gave me a different perspective of the data I'm capturing. Thanks!
Thanks for watching!
Adam,
These 2 videos are awesome! Great work!
I've learned so many things from you (here or on AdamBlock Studio), thank you! I really appreciate how you go deeply into processes, it helps a lot to understand "what am I doing".
You've defenitely contributed to improve my skill on PI (I still have tons of things to learn though)
Just a small question: why "pro" camera has so big pixel size (I guess instrument on JWST have 17µ pixel size; that's huge)? I would bet that the smaller they are, the better it is but that's not case! Maybe a matter of noise/chipset size constraint? (small pixels => more pixel => more circuit => more (read)noise)?
Thanks again for all your work.
Regards;
In general "pros" want to collect as much light as possible to improve the S/N. Resolution isn't necessary the priority and nor is worrying about saturation of non-related things (like stars). JWST is a different consideration. JSWT observes at the Infrared wavelengths of light. 17micro pixels ARE SMALL for this wavelength!
@@AdamBlock Ooooh, ok I see!
Thanks for answering :)
Thanks Adam for all your content.
In another video, the user suggested changing the white balance reference when using filters such as Optolong L-Extreme. Instead they used the white reference as 'photon flux' and not the 'average spiral galaxy'
Can you elaborate on why this would be done, when using eg: ASI294MC Pro with an Optolong L Extreme filter ?
That is a longer explanation (that I have on my site). The issue is that these isn't a white balance for narrowband. In fact, all SPCC gives you is flux calibrated answers- but this isn't color determinate.
Excellent!! Questions: I’ve been tunning DBE AFTER doing SPCC, does it matter? It does seem like when I run DBE now I’m getting better results. But I’ll change it to before SPCC if that is optimal.
Yes, DBE after SPCC can technically change the color balance (depending on how you do it). So the optimal order is DBE first followed by SPCC. At the end of the day... it should make a big difference. Just remember to use the background neutralization of SPCC correctly and you should be OK.
Thank you for an excellent and educational presentation. Appreciate the background information which really helps our understanding of SPCC and the advancement of both precision and accuracy. Having a background in Metrology, those are important, distinct and well chosen words.
Well done Adam and Thank You!
thank you!
Excellently represented. Pace, technical details and the theory behind explained incredibly well. Thank you very much Adam. 🎉
Thank you Emre! I appreciate you saying that.
So much information!🤯
You answered all my questions before I even knew I had them. Thank you for explaining all of this in such great detail.
Thanks Bill.
Hi Adam,
great video, very informative and going a bit deeper into the physics and mathematics behind it than usual “click here and then there” tutorials!
I have one question which arose at 29:30 in the video. When I plug the R/G x value into eq. 1, I get the y value shown there (0.815758). But if I plug the B/G x value into eq. 2, I get y = 0.7873 and not the shown y = 0.68755. But 0.7873 is the G white balance factor… I don’t get this sorted. Am I thinking wrong?
Regards, Bernd
Edit: Found the problem. The R/G Whitepoint x is transformed by the R/G linear fit, which is correct. But the B/G Whitepoint is transformed by the R/G linear fit as well instead of the B/G linear fit! That leads to 0.68755, which is indeed wrong. Correct is 0.7873, which you get when using the correct B/G linear fit with the B/G Whitepoint. But the white balance factors are correct again. These are normalized by the maximum of the inverses of the R/G, 1, and B/G image factors, and there we find the correct value 0.7873 for G (which is 1/(1/B/G) = B/G. So there seems to be a bug in the log output, but not in the final result.
Correct. You will note that I discovered this bug on my own and made Edoardo aware of it last week. I did my due diligence Bernd. However... I did not want to highlight the bug since it will be fixed.
@@AdamBlock Adam, absolutely no offense! I tried to reproduce the calculations to see if I understood it or not. This is a great video (as usual), so thanks for sharing!
Adam, thank you for this video. It is the first video I've watched that makes SPCC understandable rather than just a fill in the blanks and run the programme situation. I do have a question with regard to filters. I use an Atik Horizon II OSC with Hyperstar on my C11. I also use an IDAS LPS-D2 or an Optolong L-Extreme in front of my camera. As these introduce another filter in addition to the GRBG Horizon II chip filters I'm wondering what impact this would have, if any, on SPCC results and, if so, how would this be accommodated?
Kind regards,
Gord
For the light pollution filter...I do not think there will be a big effect (but, I am not an expert on this). For the NarrowBand- there is some thought as to what your goal is. For NB there isn't a color blaance- SPCC will just give you the signal strengths for each emission line for your system. But usually people stretch NB data and do not maintain this relationship.
@@AdamBlock Thanks Adam,
That was my thought as well. I don’t know for sure but I expect that given the LP filter transmission lines there would only be a small, if any, effect.
Cheers,
Gord
Adam, I really appreciate your videos. I have one question I hope you will be able to comment on. Within the first minute of the official PixInsight video, "SPCC - Working with Color Sensors" they say that when working with color sensors, we should always use the Ideal QE curve setting. In this video of yours, between 25 and 26 minutes in, you select the particular Sony sensor from the OSC camera that was used in the example image that you are working with. It seems to me that if my sensor is listed in SPCC I should select it (as you did in your example) but if that sensor is in a OSC camera, do you think I should be selecting the Ideal QE curve instead?
Yeah, one of several errors. You have to remember I made this video very shortly after the process became available... no oter content creator nor even the developers attempted to go into this detail at the time. There is risk involved in creating this stuff... e.g getting something wrong... I stepped up! So the answer is, for many OSC sensors the filter specifications are not known. Usually you specify the filter transmission ANd the sensor sensitivity at given wavelengths. (These values are just mulitplied/convolved together). But in the case of OSC, the filter values are not know. So you should specify "1" for the Ideal QE and then just specify the sensor you are using. This folds the total transmission and sensitivity together into one parameter since you do not have more
numbers to work with.
@@AdamBlock Thanks so much for the quick reply. Please allow me to ask a follow on question. If I’m using a duo narrowband filter with my OSC camera, should I select the Ideal QE curve for the sensor and the specific filter used in each of the R, G, and B filter fields?
@@kennethklock7419 That is more complicated. You are effectively using two filters. In addition, you would be using SPCC in the "Narrowband mode" to give you calibrated fluxes..but not calibrated colors. For NB imagery... calibrated fluxes may or may not be critical. Many people adjust the signal strengths in such a way they are far from what physics determines... so I think you should look things up on the official forum. This is not an area of strength for me (I actually find it perplexing as to why it is done.)
Awesome video.
Do you know if SPCC background neutralization (and/or balance) is local or global?
I.e. does it fit whole image to curve, or for each pixel of image it looks at stars (weighted by distance, I guess) and applies that?
What I'm trying to figure out is whether you need ABE/DBE at all after SPCC (if it's local, I think there should be no need?).
Your video made me think. A process like SPCC makes it possible for amateur images to be relevant in a scientific way, maybe. I'm just wondering how the PI workflow would look like to make amateur images useful.
I understood and learned a lot from Part 1, but Part 2 was way over my head. I think I’ll have to watch it again.
This was a great explanation Adam..thank you!!!! After working with Pixinsight for the better part of a year now and then coming back to watch your videos, I'm having a lot of "AH-HA!!!" moments of revelation, which I suppose is rather like putting the cart before the horse, but the hands on experience makes the theory a little easier to understand. That in turn helps immensely when I start work on a new project. More specifically concerning SPCC, as someone else posted here, I think the "change" took everyone by surprise, but your explanation makes the entire process far less intimidating knowing what its potential is. Keep up the great work!!
Thank you for watching the video! Good Lord... please pass on the word. lol
I love your enthusiasm as well as the well-done explanation for the science behind SPCC! wow how incredible. Very inspired and thanks for these 43 minutes of my day :)
Thank you very much for watching. This video has not been vetted by PixInsight developers... I am prepared to make an errata video...but I haven't heard any complaints since publishing it...so I suspect I am close with my explanation.
Excellent video and tutorial. Content is very well explained and visualized.. Appreciate your effort and will certainly check out your web side.
Thanks Olaf!
Thank you so much for this information! This level of precision make me humble that this is now available to me along with the technology and teachers like you! I can't wait to process and maybe reprocess some of my data just to see the results! Would you suggest doing a DBE before or after SPCC? I am a student of your course but I could never figure that out.
Thank you very much. Yes,you should use DBE before SPCC... to take caer of the gradients.
I’m a little lost when it comes to calculating the Catalog ratios for every star. It seems like there is only one ratio for R/G and B/G that we can get from the catalog (which are the ideal white balance ratios). How do you get the individual catalog ratio for every star so it’s not just a vertical line on the plot?
I think you might be mixing things. When you compare your stars to the catalog you are measuring a ratio of instrument response (your sensor versus GAIA data). The ratio for each star (of whatever color) will be about the same... but there will be scatter...so you do a linear fit and get a line. The slope of the line is the ratio. With this in hand (a correction)... you can then apply a white balance to your image. I haven't been thinking about this for a while... but this is my quick answer.
@@AdamBlock Yeah, I was definitely mixing things up. I was confusing the ideal white balance flux ratio with the individual star flux ratios. Must've been a long day or something...
My only question now is that once I normalize the "y" values shown in the video, I get white balance factors of .8428/.6876/1 instead of .9651/.7873/1. Is there something else happening in the background or am I failing at algebra?
@@SeanDWalker4 Hmm...When it was originally released I did discover an error in that the output was correct...but the equation values were not. I thought everything was fixed as far as I know. I would need to see your particular example to know and check it. Run the values through the equations and make certain they come up with the output ratio- then you can normalize however you would like. I not certain how SPCC chooses a particular channel to be "1" ... but it doesn't matter.
@@AdamBlock I was just looking at the values on this video since I was curious. Once I get to processing more of my own images I'll have to see if the numbers makes sense. Thanks!
Oh my...indeed! I think you found the same error! Yes, it should be correct now. Good job. I thought I was special by identifying the issue when it was released. I couldn't wait for a updated version before making this video..so the error is recorded here...though I doubted anyone would ever notice. I guess it took a year... :)
Great explanation of this great development which brings science closer to the amateur photographer! It’s astonishing to see the possibilities of thus development. And it brings the discussion of what is the real color of the color you see to a next level
Thanks for watching!
Thank you Adam Block!!!
Since we know the true spectrum of most stars now, can't we just take very short exposures of the stars in the target, and rely on SPCC to get the colors right. Then we can take longer exposure of nebulae, galaxies, etc. without going back for several hours of integration of RGB stars?
Thanks, John
Sorry.. how does taking longer exposures of nebula, galaxies..etc not permit getting the color right? (what is the issue?)
@@AdamBlock If we are now following the workflow of removing stars, processing the starless target, and later adding back minimalized or morphed down RGB stars, then time required to get a good spectral representation of the stars used should only be a short time, right? It is no longer necessary to get an hour each through RGB filters for good stars to PixelMath back in. Is SPCC now filling in the full spectrum that might be missed with longer integrations? Is my thinking correct here?
@@johnadastra1754 I think there may be a misunderstanding at what SPCC is doing. Stars are removed from the same image from which the nebula (or whatever) exists. It sounds like you are asking if you can create a starless image from a long exposure (which HAS LONG DURATION EXPOSED STARS) and also take a short exposure RGB and add that into the starless version. #1 SPCC will give you the same color calibration/white balance for the long exposure star image as the short exposure one. #2 You never did have to spend an hour through each RGB filter.... other than getting as much information as necessary. SPCC doesn't affect this - there will be the same answer. We are likely not understanding each other.
@@AdamBlock Thanks Adam. Still working on exploring the full potential of this new tool. I was talking about plugging RGB stars back into a NB image. I guess I was trying to see if SPCC would lessen the amount of integration time on the stars I would use for this. Will keep trying to see how these processes are benefiting my workflow. Seems like helpful addition to PI.
@@johnadastra1754 I don't think SPCC directly affects lessening the exposure time for RGB stars that you might insert elsewhere. It will just make certain your stars..whatever exposure time, are colored accurately.
This is amazing, thank you so much !
Thank you