[RCC] Heart Nebula: Narrowband normalization

3.01 akshay87kumar ... · view views · Share link
Dear Forum Members and Experts I am struggling to edit the Heart Nebula in HOO palette. My calibrated frame is made of: Lights: 58 frames x 300 seconds at 120 gain (~4.5+ hours of image) Darks: 33 frames x 300 seconds Bias: 60 frames x 0.001 seconds Flats:40 frames x 0.14 seconds (auto generated by ASIAIR) Equipment: ZWO ASI294MC Pro Color camera, Optolong L Extreme filter Other context: Bortle 8-9 sky, Taken at elevations 30 to 60 degree over 1 night So far, my workflow has been: Dynamic Background Extraction > Spectrophotometric Color Calibration > StarNet2 Star Removal > Noise Exterminator > GHS > Masked contrasts and curves. Any attempt to bring out more details ruins the images because the data quality is not very great. Once I have done this, I tried to run narrowband normalization (HOO, SHO), but the images I am getting are very weird. I have posted two images - the first one is without any normalization, the second one is after HOO normalization. Is this normal to expect in narrowband normalization? If yes, would like to see your recommendations on post-processing. While I definitely understand that higher integraton length and better seeing will improve the quality, I am curious to see what is possible to generate out of this dataset! I am attaching the link to calibrated frame generated using the following data set, as well as the outcome of processing. Posting only starless image as I did not have RGB lower exposure of stars yet. I would concur that the seeing was not very good (there were high clouds), so likely the image quality is not very good compared to the integration time. However, I would still want to understand how to make best use of OSC using dual narrowband filters. Link to calibrated frame (Contains .fit and .xisf format in zip folder): https://drive.google.com/drive/folders/1XLXK4TMTQYj7i1_OLs1FH_MKpwaGIq1t?usp=sharing Like

3.01

...

Dear Forum Members and Experts

I am struggling to edit the Heart Nebula in HOO palette. My calibrated frame is made of:
Lights: 58 frames x 300 seconds at 120 gain (~4.5+ hours of image)
Darks: 33 frames x 300 seconds
Bias: 60 frames x 0.001 seconds
Flats:40 frames x 0.14 seconds (auto generated by ASIAIR)

Equipment: ZWO ASI294MC Pro Color camera, Optolong L Extreme filter
Other context: Bortle 8-9 sky, Taken at elevations 30 to 60 degree over 1 night

So far, my workflow has been: Dynamic Background Extraction > Spectrophotometric Color Calibration > StarNet2 Star Removal > Noise Exterminator > GHS > Masked contrasts and curves. Any attempt to bring out more details ruins the images because the data quality is not very great.

Once I have done this, I tried to run narrowband normalization (HOO, SHO), but the images I am getting are very weird. I have posted two images - the first one is without any normalization, the second one is after HOO normalization. Is this normal to expect in narrowband normalization? If yes, would like to see your recommendations on post-processing.

While I definitely understand that higher integraton length and better seeing will improve the quality, I am curious to see what is possible to generate out of this dataset!

I am attaching the link to calibrated frame generated using the following data set, as well as the outcome of processing. Posting only starless image as I did not have RGB lower exposure of stars yet. I would concur that the seeing was not very good (there were high clouds), so likely the image quality is not very good compared to the integration time. However, I would still want to understand how to make best use of OSC using dual narrowband filters.

Link to calibrated frame (Contains .fit and .xisf format in zip folder): https://drive.google.com/drive/folders/1XLXK4TMTQYj7i1_OLs1FH_MKpwaGIq1t?usp=sharing

7.29 D. Jung ... · view views · 2 likes · Share link
You have some nasty gradients in your image. Once you get those removed, you can get a passable edit out, considering the low exposure time. In case you have a widefield view of the heart, you can use this technique to better remove the gradients. https://pixinsight.com/tutorials/multiscale-gradient-correction/ Though accourding to PixInsight announcement, the new MARS data base will be released with PI1.9 any day now. https://pixinsight.com/mars/ Edited ... Like

7.29

D. Jung

...

· 2 likes

· Share link

You have some nasty gradients in your image. Once you get those removed, you can get a passable edit out, considering the low exposure time.

In case you have a widefield view of the heart, you can use this technique to better remove the gradients.
https://pixinsight.com/tutorials/multiscale-gradient-correction/
Though accourding to PixInsight announcement, the new MARS data base will be released with PI1.9 any day now.
https://pixinsight.com/mars/

Edited ...

3.01 akshay87kumar Topic starter ... · view views · Share link
D. Jung: You have some nasty gradients in your image. I had been using DBE, with subtraction. Anything major and different that you tried out? I really like the way your image has come out. Like

7.29 D. Jung ... · view views · 1 like · Share link
D. Jung: You have some nasty gradients in your image. I had been using DBE, with subtraction. Anything major and different that you tried out? I really like the way your image has come out. For this image I used seti's AutoDBE (masking the heart) and PIs DBE and gradient removal process. However, if it was my own image, I would have used my wide field shots. Usually gives superior results when the gradients are that messy. You can see that in my edit here, the contrast in the central region lacks. Probably due to false background removal. Edited ... Like

7.29

D. Jung

...

· 1 like

· Share link

D. Jung:
You have some nasty gradients in your image.

I had been using DBE, with subtraction. Anything major and different that you tried out? I really like the way your image has come out.

For this image I used seti's AutoDBE (masking the heart) and PIs DBE and gradient removal process.
However, if it was my own image, I would have used my wide field shots. Usually gives superior results when the gradients are that messy.
You can see that in my edit here, the contrast in the central region lacks. Probably due to false background removal.

Edited ...

8.11 Tony Gondola ... · view views · Share link
Have you done any data culling? Some of your imaging time was at 30 degrees which is very low. This will for sure hurt your image quality. I think PI has some great tools for looking at the quality of your frames so I'd start there before you worry too much about colors. I know it's hard to throw away data but it's worth doing in the end. You're also not helping yourself by taking such long subs. 120 sec or even 60 sec will be sharper. The read noise is very low at a gain just above 120 so that's where you should be if your going to do this. If you do cull your data you'll loose less time overall with this approach. Like

8.11

Tony Gondola

...

· Share link

Have you done any data culling? Some of your imaging time was at 30 degrees which is very low. This will for sure hurt your image quality. I think PI has some great tools for looking at the quality of your frames so I'd start there before you worry too much about colors. I know it's hard to throw away data but it's worth doing in the end. You're also not helping yourself by taking such long subs. 120 sec or even 60 sec will be sharper. The read noise is very low at a gain just above 120 so that's where you should be if your going to do this. If you do cull your data you'll loose less time overall with this approach.

3.01 akshay87kumar Topic starter ... · view views · Share link
D. Jung: D. Jung: You have some nasty gradients in your image. I had been using DBE, with subtraction. Anything major and different that you tried out? I really like the way your image has come out. For this image I used seti's AutoDBE (masking the heart) and PIs DBE and gradient removal process. However, if it was my own image, I would have used my wide field shots. Usually gives superior results when the gradients are that messy. You can see that in my edit here, the contrast in the central region lacks. Probably due to false background removal. Thanks D. Jung for the detailed insights. I have tried AutoDBE from SetiAstro, but didnt explore masked AutoDBE (i realize how versatile masks can be!). One more question please - what is your recommendation on the exposure length? I have been using 300s, havent really explored playing around much with 60s, 120s, 180s vs 300s subs as I am relatively new. I believe guiding will be of key importance here to answer the question - I have been able to typically get guiding in the range of 2" accuracy so far. I have read that for my camera (ASI294MC Pro), Telescope (WO 71GT APO) and iOptron CEM40 mount, I should target getting 1" accuracy, but I have miserably failed achieving anything lower than 2" in stable manner. I keep dithering on at 3px after every frame, stability 2", and settle time 1 seconds. Appreciate all your insights and help, Thanks! Edited ... Like

3.01

akshay87kumar

Topic starter

...

· Share link

D. Jung:
D. Jung:
You have some nasty gradients in your image.

I had been using DBE, with subtraction. Anything major and different that you tried out? I really like the way your image has come out.

For this image I used seti's AutoDBE (masking the heart) and PIs DBE and gradient removal process.
However, if it was my own image, I would have used my wide field shots. Usually gives superior results when the gradients are that messy.
You can see that in my edit here, the contrast in the central region lacks. Probably due to false background removal.

Thanks D. Jung for the detailed insights. I have tried AutoDBE from SetiAstro, but didnt explore masked AutoDBE (i realize how versatile masks can be!). One more question please - what is your recommendation on the exposure length? I have been using 300s, havent really explored playing around much with 60s, 120s, 180s vs 300s subs as I am relatively new.

I believe guiding will be of key importance here to answer the question - I have been able to typically get guiding in the range of 2" accuracy so far. I have read that for my camera (ASI294MC Pro), Telescope (WO 71GT APO) and iOptron CEM40 mount, I should target getting 1" accuracy, but I have miserably failed achieving anything lower than 2" in stable manner. I keep dithering on at 3px after every frame, stability 2", and settle time 1 seconds.

Appreciate all your insights and help, Thanks!

Edited ...

7.29 D. Jung ... · view views · 1 like · Share link
D. Jung: D. Jung: You have some nasty gradients in your image. I had been using DBE, with subtraction. Anything major and different that you tried out? I really like the way your image has come out. For this image I used seti's AutoDBE (masking the heart) and PIs DBE and gradient removal process. However, if it was my own image, I would have used my wide field shots. Usually gives superior results when the gradients are that messy. You can see that in my edit here, the contrast in the central region lacks. Probably due to false background removal. Thanks D. Jung for the detailed insights. I have tried AutoDBE from SetiAstro, but didnt explore masked AutoDBE (i realize how versatile masks can be!). One more question please - what is your recommendation on the exposure length? I have been using 300s, havent really explored playing around much with 60s, 120s, 180s vs 300s subs as I am relatively new. I believe guiding will be of key importance here to answer the question - I have been able to typically get guiding in the range of 2" accuracy so far. I have read that for my camera (ASI294MC Pro), Telescope (WO 71GT APO) and iOptron CEM40 mount, I should target getting 1" accuracy, but I have miserably failed achieving anything lower than 2" in stable manner. I keep dithering on at 3px after every frame, stability 2", and settle time 5 seconds. Appreciate all your insights and help, Thanks! That's a tricky question and usually triggers a lengthy discussion. Mathematically speaking, it doesnt matter if you do 100x1min or 10x10min, you get exactly the same result; but in reality you should take guiding, seeing, choice of filter, camera specs and so on into consideration. The main drawback with shorter exposures is the amount of data you collect and the increased demand in storage, memory and pre-processing time. PI now offers fast batch processing to help with that. Longer exposure times suffer from loss of sharpness due to guiding and seeing and saturation issues. In your case I'd, go with shorter exposure times, somewhere between 2-3 minutes with a narrowband filter. In my opinion, there is no right or wrong answer, just balancing differnt pros and cons. Like

7.29

D. Jung

...

· 1 like

· Share link

D. Jung:
D. Jung:
You have some nasty gradients in your image.

I had been using DBE, with subtraction. Anything major and different that you tried out? I really like the way your image has come out.

For this image I used seti's AutoDBE (masking the heart) and PIs DBE and gradient removal process.
However, if it was my own image, I would have used my wide field shots. Usually gives superior results when the gradients are that messy.
You can see that in my edit here, the contrast in the central region lacks. Probably due to false background removal.

Thanks D. Jung for the detailed insights. I have tried AutoDBE from SetiAstro, but didnt explore masked AutoDBE (i realize how versatile masks can be!). One more question please - what is your recommendation on the exposure length? I have been using 300s, havent really explored playing around much with 60s, 120s, 180s vs 300s subs as I am relatively new.

I believe guiding will be of key importance here to answer the question - I have been able to typically get guiding in the range of 2" accuracy so far. I have read that for my camera (ASI294MC Pro), Telescope (WO 71GT APO) and iOptron CEM40 mount, I should target getting 1" accuracy, but I have miserably failed achieving anything lower than 2" in stable manner. I keep dithering on at 3px after every frame, stability 2", and settle time 5 seconds.

Appreciate all your insights and help, Thanks!

That's a tricky question and usually triggers a lengthy discussion.
Mathematically speaking, it doesnt matter if you do 100x1min or 10x10min, you get exactly the same result; but in reality you should take guiding, seeing, choice of filter, camera specs and so on into consideration.
The main drawback with shorter exposures is the amount of data you collect and the increased demand in storage, memory and pre-processing time. PI now offers fast batch processing to help with that.
Longer exposure times suffer from loss of sharpness due to guiding and seeing and saturation issues.
In your case I'd, go with shorter exposure times, somewhere between 2-3 minutes with a narrowband filter.
In my opinion, there is no right or wrong answer, just balancing differnt pros and cons.

8.11 Tony Gondola ... · view views · Share link
I don't know what mount you have but 2 arc/sec RMS is pretty poor. I don't know what mount you are using but 1" RMS should be reachable with standard settings with 0.6" being very possible. That is certainly something to look at. Like

3.01 akshay87kumar Topic starter ... · view views · Share link
Took a second pass at trying to edit the image in standard RGB palette, and an HOO palette. It is my first attempt to post process in HOO Palette, so request your feedback and improvement areas! Like

7.29 D. Jung ... · view views · 1 like · Share link
Some things you could consider: - A little bit of noise helps retaining sharpness. You could try noisexterminator at .5 and use it multiple times until you reach the desired nose level. And use a mask to protect the nebula - maybe apply a little less blurrxterminator. Over sharpening stars tends to move towards saturation and lots of color. A little blurred states often look more natural - you could play a bit with the colors. Use an inverted mask on the nebula and change RGB until you reach a good level of red/blue. That helps keeping the background black while changing the color of the nebulosity Don't change anything if you like your image. It's for you not for me Like

7.29

D. Jung

...

· 1 like

· Share link

Some things you could consider:
- A little bit of noise helps retaining sharpness. You could try noisexterminator at .5 and use it multiple times until you reach the desired nose level. And use a mask to protect the nebula
- maybe apply a little less blurrxterminator. Over sharpening stars tends to move towards saturation and lots of color. A little blurred states often look more natural
- you could play a bit with the colors. Use an inverted mask on the nebula and change RGB until you reach a good level of red/blue. That helps keeping the background black while changing the color of the nebulosity

Don't change anything if you like your image. It's for you not for me smile

3.01 akshay87kumar Topic starter ... · view views · Share link
D. Jung: Some things you could consider: - A little bit of noise helps retaining sharpness. You could try noisexterminator at .5 and use it multiple times until you reach the desired nose level. And use a mask to protect the nebula - maybe apply a little less blurrxterminator. Over sharpening stars tends to move towards saturation and lots of color. A little blurred states often look more natural - you could play a bit with the colors. Use an inverted mask on the nebula and change RGB until you reach a good level of red/blue. That helps keeping the background black while changing the color of the nebulosity Don't change anything if you like your image. It's for you not for me Thanks for the feedback - you inspire to continue improving I agree, i pushed noisexterminator to 0.9 in first go. I havent purchased blurxterminator, so I used PixInsights built in Deconvolution. But I agree it has made it a bit too sharp. I did not have RGB colours, and my narrowband stars were indeed very saturated due to long exposure. I just used Narrowband to RGB stars script for now, but definitely will revisit once I have more data for narrowband and separate shorter data for RGB stars. Like

3.01

akshay87kumar

Topic starter

...

· Share link

D. Jung:
Some things you could consider:
- A little bit of noise helps retaining sharpness. You could try noisexterminator at .5 and use it multiple times until you reach the desired nose level. And use a mask to protect the nebula
- maybe apply a little less blurrxterminator. Over sharpening stars tends to move towards saturation and lots of color. A little blurred states often look more natural
- you could play a bit with the colors. Use an inverted mask on the nebula and change RGB until you reach a good level of red/blue. That helps keeping the background black while changing the color of the nebulosity

Don't change anything if you like your image. It's for you not for me

Thanks for the feedback - you inspire to continue improving

I agree, i pushed noisexterminator to 0.9 in first go. I havent purchased blurxterminator, so I used PixInsights built in Deconvolution. But I agree it has made it a bit too sharp.
I did not have RGB colours, and my narrowband stars were indeed very saturated due to long exposure. I just used Narrowband to RGB stars script for now, but definitely will revisit once I have more data for narrowband and separate shorter data for RGB stars.

8.21 Miguel T. ... · view views · Share link
D. Jung: D. Jung: D. Jung: You have some nasty gradients in your image. I had been using DBE, with subtraction. Anything major and different that you tried out? I really like the way your image has come out. For this image I used seti's AutoDBE (masking the heart) and PIs DBE and gradient removal process. However, if it was my own image, I would have used my wide field shots. Usually gives superior results when the gradients are that messy. You can see that in my edit here, the contrast in the central region lacks. Probably due to false background removal. Thanks D. Jung for the detailed insights. I have tried AutoDBE from SetiAstro, but didnt explore masked AutoDBE (i realize how versatile masks can be!). One more question please - what is your recommendation on the exposure length? I have been using 300s, havent really explored playing around much with 60s, 120s, 180s vs 300s subs as I am relatively new. I believe guiding will be of key importance here to answer the question - I have been able to typically get guiding in the range of 2" accuracy so far. I have read that for my camera (ASI294MC Pro), Telescope (WO 71GT APO) and iOptron CEM40 mount, I should target getting 1" accuracy, but I have miserably failed achieving anything lower than 2" in stable manner. I keep dithering on at 3px after every frame, stability 2", and settle time 5 seconds. Appreciate all your insights and help, Thanks! That's a tricky question and usually triggers a lengthy discussion. Mathematically speaking, it doesnt matter if you do 100x1min or 10x10min, you get exactly the same result; but in reality you should take guiding, seeing, choice of filter, camera specs and so on into consideration. The main drawback with shorter exposures is the amount of data you collect and the increased demand in storage, memory and pre-processing time. PI now offers fast batch processing to help with that. Longer exposure times suffer from loss of sharpness due to guiding and seeing and saturation issues. In your case I'd, go with shorter exposure times, somewhere between 2-3 minutes with a narrowband filter. In my opinion, there is no right or wrong answer, just balancing differnt pros and cons. Digitally speaking it does matter. You'll eventually reach a point, whether it's the exposure time being ridiculously too low or the object being too faint, where even a 16 bits number doesn't have enough precision to accommodate the difference between two shade of luminance given a short exposure time. 16 bits being the storage capacity of the file format, there's some camera model that only have 10, 12 or 14 bits ADC. And there is much more to that. I wouldn't recommend looking only at disk space. Like

8.21

Miguel T.

...

· Share link

D. Jung:
D. Jung:
D. Jung:
You have some nasty gradients in your image.

I had been using DBE, with subtraction. Anything major and different that you tried out? I really like the way your image has come out.

For this image I used seti's AutoDBE (masking the heart) and PIs DBE and gradient removal process.
However, if it was my own image, I would have used my wide field shots. Usually gives superior results when the gradients are that messy.
You can see that in my edit here, the contrast in the central region lacks. Probably due to false background removal.

Thanks D. Jung for the detailed insights. I have tried AutoDBE from SetiAstro, but didnt explore masked AutoDBE (i realize how versatile masks can be!). One more question please - what is your recommendation on the exposure length? I have been using 300s, havent really explored playing around much with 60s, 120s, 180s vs 300s subs as I am relatively new.

I believe guiding will be of key importance here to answer the question - I have been able to typically get guiding in the range of 2" accuracy so far. I have read that for my camera (ASI294MC Pro), Telescope (WO 71GT APO) and iOptron CEM40 mount, I should target getting 1" accuracy, but I have miserably failed achieving anything lower than 2" in stable manner. I keep dithering on at 3px after every frame, stability 2", and settle time 5 seconds.

Appreciate all your insights and help, Thanks!

That's a tricky question and usually triggers a lengthy discussion.
Mathematically speaking, it doesnt matter if you do 100x1min or 10x10min, you get exactly the same result; but in reality you should take guiding, seeing, choice of filter, camera specs and so on into consideration.
The main drawback with shorter exposures is the amount of data you collect and the increased demand in storage, memory and pre-processing time. PI now offers fast batch processing to help with that.
Longer exposure times suffer from loss of sharpness due to guiding and seeing and saturation issues.
In your case I'd, go with shorter exposure times, somewhere between 2-3 minutes with a narrowband filter.
In my opinion, there is no right or wrong answer, just balancing differnt pros and cons.

Digitally speaking it does matter. You'll eventually reach a point, whether it's the exposure time being ridiculously too low or the object being too faint, where even a 16 bits number doesn't have enough precision to accommodate the difference between two shade of luminance given a short exposure time. 16 bits being the storage capacity of the file format, there's some camera model that only have 10, 12 or 14 bits ADC. And there is much more to that. I wouldn't recommend looking only at disk space.

8.11 Tony Gondola ... · view views · Share link
Are you talking about a single frame or after stacking? Like

	Distinct awarded users	Total awarded images
Image of the day
Top picks
Top pick nominations

	Image of the day	Top pick	Top pick nominations	Total submitted
Deep sky
Solar system
Extremely wide field
Star trails
Northern lights
Noctilucent clouds
Landscape
Artificial satellite

	Image of the day	Top pick	Top pick nominations	Total submitted
Backyard
Traveller
Own remote observatory
Amateur hosting facility
Public amaeteur data
Professional, scientific grade data
Mix of multiple sources
Other
Unknown

AstroBin

AstroBin

Image Index:

Contribution Index (beta):

[RCC] Heart Nebula: Narrowband normalization Requests for constructive critique · akshay87kumar · ... · 12 · 481 · 4

AstroBin

AstroBin

Image Index:

Contribution Index (beta):

[RCC] Heart Nebula: Narrowband normalization Requests for constructive critique · akshay87kumar · ... · 12 · 481 · 4

Sorry

Sorry

Sorry :-(

Sorry

Share this forum post

Post editing not available

Sorry

Sorry

Sorry :-(

Sorry

Share this forum post

Post editing not available

Sorry

Sorry

Sorry :-(

Sorry

Share this forum post

Post editing not available

Sorry

Sorry

Sorry :-(

Sorry

Share this forum post

Post editing not available

Sorry

Sorry

Sorry :-(

Sorry

Share this forum post

Post editing not available

Sorry

Sorry

Sorry :-(

Sorry

Share this forum post

Post editing not available

Sorry

Sorry

Sorry :-(

Sorry

Share this forum post

Post editing not available

Sorry

Sorry

Sorry :-(

Sorry

Share this forum post

Post editing not available

Sorry

Sorry

Sorry :-(

Sorry

Share this forum post

Post editing not available

Sorry

Sorry

Sorry :-(

Sorry

Share this forum post

Post editing not available

Sorry

Sorry

Sorry :-(

Sorry

Share this forum post

Post editing not available

Sorry

Sorry

Sorry :-(

Sorry

Share this forum post

Post editing not available

Sorry

Sorry

Sorry :-(