Elephants Trunk Re-Visited

[Rodd]

How do I get 300 dpi--I read in a post that he processed in photoshop at 300 dpi.

[jtlandreneau]

your latest picture I think has had a pretty dramatic increase in improvement. I personally am not the biggest fan of the hubble palette (something about mapping H-a to green doesn't agree with me). The Heart nebula is awesome in the HOO palate if you want something to experiment with

I think my only comment would be tweaking the background back to "black" (don't clip it of course) and away from the purples you have now

[Rodd]

Anything I do to reduce the background color also reduces the quality of the nebulosity.  I have placed around with the CFHT and HSO pallets (2 other ones--I think they both have Ha as red).  Every time I shoot in narrowband I end up with 3 times as much work because I process the pic in all of them!  In the end, I like RGB (with Ha or L) the best--well, maybe not--I like them all.  I do like producing a "natural" colored pic though.  Something about trying to accurately reproduce the human color pallet.  Now, if I was a grasshopper......

[Stub Mandrel]

How do I get 300 dpi--I read in a post that he processed in photoshop at 300 dpi.

DPI has no influence on how you process an image, unless you change it without changing the image dimensions.

An image has three 'measures' of size:

Pixels is the only meaningful one for processing, if increased you blur details, if reduced you lose detail.

The 'physical size' is what size the image will come out if printed 1:1.

DPI is the number of dots (pixels) per inch if the image is printed 1:1 It is just a scaling factor, so if your image is 3000 pixels wide, at 300dpi it will print 10 inches wide, at 100 dpi it will print 30 inches wide.

Printed media can have DPIs of up to 1200 or more, but 300 DPI is usual as this is a good match to the resolution of the eye at a comfortable reading distance. Computer screens often have resolutions below 100 dpi, and are often viewed for further away. An image that looks good on screen won't look as good printed at the same size if you look at it closely.

If you change the DPI you have to be careful to check the box that prevents the pixel size of the image changing or you will degrade the image.

Changing the dpi of your image to 300 (without resampling the pixels) will just make it print much smaller, but it will look sharper on the paper.

In publishing, where images need to be scaled to fit a space, if this results in DPI below 300, we usually resample the image to increase the pixel count. This usually has to be followed by judicous tweaks, including sharpening as otherwise the image often appears soft. If this isn't done it will pixelate which looks worse.

[Rodd]

That's about what the printer said.  He also said my pic was set at 72 dpi, which seems quite low.  There is no question the resolution of my pics is very low.  I don't know how to fix it.  According to respected members of thius forum my scope (Televue np101is) is capable of taking very sharp, detailed images.   That's why I said its either my processing or the camera.  

[Stub Mandrel]

The image is 1600 x 1209 pixels, 5.8 megapixels, that's higher resolution than most dedicated astro cameras.

I think you are looking at people's huge mosaic photos and expecting to much from a single frame.

[Rodd]

Actually my camera is 8.3 megapixels--but no, I have seen awesome photos that are not mosaics that are so clear you think you are looking out a window in space.  there is no question that my image above is low resolution--it is not sharp and clear, but grainy.  The question is why.  If it is because I do not know how to process, then I can live with that as it will develop over time.  But I get the sinking feeling that there is an underlying problem that no one has identified yet--a setting, something.  I have tried all the tools available to me in processing and no amount of work improves the image.

[Pompey Monkey]

Actually my camera is 8.3 megapixels--but no, I have seen awesome photos that are not mosaics that are so clear you think you are looking out a window in space.  there is no question that my image above is low resolution--it is not sharp and clear, but grainy.  The question is why.  If it is because I do not know how to process, then I can live with that as it will develop over time.  But I get the sinking feeling that there is an underlying problem that no one has identified yet--a setting, something.  I have tried all the tools available to me in processing and no amount of work improves the image.

Two things:

Patience;

Experience.

Deep sky astrophotography cannot be rushed.

[ChrisLX200]

I can see an improvement in dynamic range - the highlights are no longer blown out - so that's a positive. However, you have changed so many things I've gone dizzy trying to keep track. If you have a problem like this you should change one thing and examine the result before moving on to the next thing to test. So the question of resolution (sharpness/detail/background noise) cannot be answered by us looking at your processed low-res jpg image. Only by looking at the raw stack (16-bit minimum, callibrated preferably but I'm not even sure what you're doing for your callibration method). If I had access to that then I could tell you if the data is missing something or if it is your processing.

ChrisH

[Davey-T]

Actually my camera is 8.3 megapixels--but no, I have seen awesome photos that are not mosaics that are so clear you think you are looking out a window in space. 

Bear in mind that many images in astro mags are taken up mountains in Arizona with many thousands of dollars of equipment so don't get hung up on trying to emulate them from your backyard.

Dave

[Rodd]

I can see an improvement in dynamic range - the highlights are no longer blown out - so that's a positive. However, you have changed so many things I've gone dizzy trying to keep track. If you have a problem like this you should change one thing and examine the result before moving on to the next thing to test. So the question of resolution (sharpness/detail/background noise) cannot be answered by us looking at your processed low-res jpg image. Only by looking at the raw stack (16-bit minimum, callibrated preferably but I'm not even sure what you're doing for your callibration method). If I had access to that then I could tell you if the data is missing something or if it is your processing.

ChrisH

Not sure how to get that to you Chris.  I realize you can't analyze my JPEG--that's why I work on it until I can't make it better and post it--if it is better, I am going in the right direction.  Other JPEGs on this forum look amazing, so it is possible to post one that is great.  I am trying everything at my disposal to improve the resolution (Clarity, detail, sharpness etc).  I have not come upon any one thing that improves the picture very much--hardly worth posting those minuscule increments of improvement--probably would get lost in the JPEG compression anyway.  I gave it my all (and then some) and the above pic is the result. So I posted it.  Short of that, there was no significant change--even this one is really the same with modified color and contrast and other, aesthetical things--though I did calibrate with darks, biases and flats for the first time--I've only ever used darks before this.  I do see a difference.    You asked about calibration--I made a master dark out of 21, 20 minute dark frames, subtracted a master bias from that (10 bias frames in the master).  Finally, I subtracted a master flat from each sub (10 Ha flats, 10 OIII flats and 10 SII flats).  I did not shoot my dark flats right so I could not use them.  Instead of just turning the illumination of the foil off, I actually closed the shutter and shot full darks--which I am assuming is wrong as it ruined the stacks. 

If you think its my processing, wonderful.  When I decide on a permanent processing platform I will improve. But I am not sure that is what I am hearing here.  I'm not convinced yet   Resolution should be an underlying attribute of the pic no? able to be tweeked with processing, but not wholly dependent on it.  How likely is it.  The resolution (my terminology for clarity, sharpness, etc) is pretty bad.  Just click on the image and compare it with any one of a number on other pages.   It gets pixelated with the slightest of zoom--like it was taken with a 1mp camera.

[Rodd]

Dave--my system cost over $20,000.    I have seen images from single shot color  DSLRs taken from the city limits that appear clear and sharp.  Taken by folks who do not use biases and flats, or even master darks,   On a weighted grade--taking into consideration the  quality of the equipment, the sky conditions in CT, and my experience (or lack thereof) I give it a D-.  My lack of experience is what saved me from an F.  I'm being objective--not taking about aesthetics at all-color, framing, etc.  I am talking about anobjective element--resolution (clraity, detail, sharpness)--pixellation at the slightest size increase.  Can that even be processed out?  One can't change a zebras stripes after all.

[ChrisLX200]

Not sure how to get that to you Chris.  I realize you can't analyze my JPEG--that's why I work on it until I can't make it better and post it--if it is better, I am going in the right direction.  Other JPEGs on this forum look amazing, so it is possible to post one that is great.  I am trying everything at my disposal to improve the resolution (Clarity, detail, sharpness etc).  I have not come upon any one thing that improves the picture very much--hardly worth posting those minuscule increments of improvement--probably would get lost in the JPEG compression anyway.  I gave it my all (and then some) and the above pic is the result. So I posted it.  Short of that, there was no significant change--even this one is really the same with modified color and contrast and other, aesthetical things--though I did calibrate with darks, biases and flats for the first time--I've only ever used darks before this.  I do see a difference.    You asked about calibration--I made a master dark out of 21, 20 minute dark frames, subtracted a master bias from that (10 bias frames in the master).  Finally, I subtracted a master flat from each sub (10 Ha flats, 10 OIII flats and 10 SII flats).  I did not shoot my dark flats right so I could not use them.  Instead of just turning the illumination of the foil off, I actually closed the shutter and shot full darks--which I am assuming is wrong as it ruined the stacks. 

If you think its my processing, wonderful.  When I decide on a permanent processing platform I will improve. But I am not sure that is what I am hearing here.  I'm not convinced yet   Resolution should be an underlying attribute of the pic no? able to be tweeked with processing, but not wholly dependent on it.  How likely is it.  The resolution (my terminology for clarity, sharpness, etc) is pretty bad.  Just click on the image and compare it with any one of a number on other pages.   It gets pixelated with the slightest of zoom--like it was taken with a 1mp camera.

Well it's not going to be the camera - either it works or it doesn't, it won't decide to produce low-res images on its own. Whether the resolution is within your data I couldn't say for certain (which is why I wanted to look at your stack to be sure) - but anyway that will be dependant upon exposure, focus and tracking. Looking at what you have as a jpg I can't believe your raw data is so poor that it is the root cause. So that leaves your processing - are you working with (processing) the full 16-bit or 32-bit (if stacked with DSS) data - either as TIF or FIT files, or are you actually processing saved JPGs? If the latter then you've no chance of producing all that which the data has to give. It is basic things like this that needs to be looked at - not all the fancy stuff with dark flats and so on. There is a fundemental error happing here but we cannot follow what you're doing.

ChrisH

[Rodd]

I process in FITs files.  Yes they do look almost black before I start stretching.  You all helped me with that mistake.  The image above represents the first image that I truly calibrated and exclusively used FITS.  Prior to this I processed in TIFF.  PixInsight saves as a floating 32 bit, and I think Nebulosity does too.  Its either 16 or 32.  Not 8

[Davey-T]

-pixellation at the slightest size increase.  Can that even be processed out?  One can't change a zebras stripes after all.

How are you increasing the size ? just zooming in will show pixels.

Your JPEG is about A4 size so would print without looking pixelated at 72 DPI

Dave

[Rodd]

Yes its 72 dpi according to the printer.  All I know is the "resolution" (clarity, detail, sharpness, depth) is poor compared to most pics on this forum.  I don't really zoom-but when you click on the pic it opens up bigger.  In Nebulosity or pixinsite you increase zoom by %.  How else to do it? Rezise and re-sample?

[Rodd]

The image is 1600 x 1209 pixels, 5.8 megapixels, that's higher resolution than most dedicated astro cameras.

I think you are looking at people's huge mosaic photos and expecting to much from a single frame.

For god sake take a look at "a couple of shots from s newbie" and tell me my resolution does not stink!!!

[ChrisLX200]

For god sake take a look at "a couple of shots from s newbie" and tell me my resolution does not stink!!!

But the highly compressed JPG image you posted here is what - 50% full size? So only 1/4 the actual size of the image. Sign up to a free hosting site (I use PhotoBucket & Astrobin, but Flikr is good too) then you can post the whole thing at full resolution.

ChrisH

[Rodd]

Trust me--it does not look any better as a FITs file on my computer--in fact, there is no discernible difference.  Why are his pics so much clearer--they are JPEGs too?

[Rodd]

But the highly compressed JPG image you posted here is what - 50% full size? So only 1/4 the actual size of the image. Sign up to a free hosting site (I use PhotoBucket & Astrobin, but Flikr is good too) then you can post the whole thing at full resolution.

ChrisH

Trust me--it does not look any better as a FITs file on my computer--in fact, there is no discernible difference.  Why are his pics so much clearer--they are JPEGs too?

[Stub Mandrel]

> "resolution" (clarity, detail, sharpness, depth)

I'm just a beginner, but I think that by focusing on 'resolution' you are at risk of chasing shadows. Here's a clip from your image, blown up without anti-aliasing so pixel detail is obvious:

Resolution is the smallest scale at which you can distinguish to objects as being separate. In your image the smallest stars are one or two bright pixels with slight overspil onto adjacent pixels, which is spot on in my book.

I suspect that what you really are trying to do is bring out more detail and contrast in the gas clouds?

Try sharpening (such as L-R deconvolution) on your image, plus a bit of noise reduction. I had a go but although the image can be made to look more 'depthy' the jpeg artefacts became really unpleasant. I think that subtle denoising and subtle sharpening of the background might do what you want, but these need to be done on the uncompressed image.

Another route is the s-shaped tone curve that increases contrast in the middle range of the image.

[Rodd]

> "resolution" (clarity, detail, sharpness, depth)

I'm just a beginner, but I think that by focusing on 'resolution' you are at risk of chasing shadows. Here's a clip from your image, blown up without anti-aliasing so pixel detail is obvious:

zoom.png

Resolution is the smallest scale at which you can distinguish to objects as being separate. In your image the smallest stars are one or two bright pixels with slight overspil onto adjacent pixels, which is spot on in my book.

I suspect that what you really are trying to do is bring out more detail and contrast in the gas clouds?

Try sharpening (such as L-R deconvolution) on your image, plus a bit of noise reduction. I had a go but although the image can be made to look more 'depthy' the jpeg artefacts became really unpleasant. I think that subtle denoising and subtle sharpening of the background might do what you want, but these need to be done on the uncompressed image.

Another route is the s-shaped tone curve that increases contrast in the middle range of the image.

That's pretty cool-not sure I completely understand your demonstration, but it appears empirical.  So--perhaps it is my lack of processing skills.  Thank you

[jtlandreneau]

For god sake take a look at "a couple of shots from s newbie" and tell me my resolution does not stink!!!

well, time to unfollow this thread. Rodd, you have a bunch of people trying to offer advice, and to be completely frank you are responding with a lot of hostility. I'll repeat Pompey's post:

Patience.

Experience.

Deep sky astrophotography cannot be rushed.

This hobby demands huge amounts of patience, both in acquisition and post processing. Your equipment is top notch, your data looks very good. Your post processing in this thread alone has shown very nice improvement. You are only going to be more frustrated if you push the people who are giving support away.

At this point, I can only suggest to throw your raw, stacked data onto a file sharing site like Dropbox and let the community have a go at it. Personally I've learned a lot doing so with my shots just by seeing what's possible with the dataset.

[Davey-T]

No idea about Nebulosity but in P'Shop you can resize images using the resize image tab.

If you don't want to be bothered registering a on hosting site you could upload an original HA image to One Drive or something and email me a link I can have a look.

Dave

[Rodd]

well, time to unfollow this thread. Rodd, you have a bunch of people trying to offer advice, and to be completely frank you are responding with a lot of hostility. I'll repeat Pompey's post:

Patience.

Experience.

Deep sky astrophotography cannot be rushed.

This hobby demands huge amounts of patience, both in acquisition and post processing. Your equipment is top notch, your data looks very good. Your post processing in this thread alone has shown very nice improvement. You are only going to be more frustrated if you push the people who are giving support away.

At this point, I can only suggest to throw your raw, stacked data onto a file sharing site like Dropbox and let the community have a go at it. Personally I've learned a lot doing so with my shots just by seeing what's possible with the dataset.

No hostility intended-frustration perhaps, but I don't see where you get hostility.  Hmmm.  Heres my last post--perhaps an improvement--perhaps not.  Cheers.