Jump to content

Banner.jpg.b89429c566825f6ab32bcafbada449c9.jpg

Mars 3rd March 14"Dob

Clayton

By Clayton
in Imaging - Planetary

Recommended Posts

cgarry

cgarry

Posted
Posted

I usually avoid playing with other people's data (sorry Rob), especially in JPEG form, but I have had a play with the curves on Rob's image that was processed with the equalised histograms and come up with this:

post-21918-13387774062_thumb.jpg

Okay, I have brightened the highlights too much, but I think it shows that equalising the histograms has produced a less 'speckled' image than cleaning up the grid noise with Image analyzer after stacking.

Cheers,

Chris

Link to comment
Share on other sites
  • Replies 52
  • Created
  • Last Reply
cgarry

cgarry

Posted
Posted
Me too, ive had two cameras, one was better than the other for this weird as it sounds

I did wonder if it was due to the bayer filter not being located in exactly the correct position on the sensor. That is just wild speculation but could be tested by shooting a light source through R, G and B filter and seeing how much light make it to the wrong pixels.

Cheers,

Chris

Link to comment
Share on other sites
Clayton

Clayton

Posted
  • Author
Posted
I only get it at full gain and then not always, low histos and full gain is the real danger for them ive found. I never noticed it here, but then im getting to old for fine inspectionicon9.gif looking harder i see it now in the middle just going to check what analyzer does with it Rob

Image analyzer detected it as 3 blobs of noise im not sure which one it was it might be all three i would have to experiment, but its now gone, i didnt blur this away rob i subtracted the grid noise, it should ( hopefully ) be as is but with out the grid, were you at full gain Rob, its getting difficult to tell if its all gone

Chris is it all gone now compared to the png link Blowing it up and turning analyzer off and on its pretty much all gone you would need to see me switching off and on at high blow up to see, im sure its done the job. wonder what caused that. What sharpening do you use Rob ?

Hi Neil,

From memory the exposure was 1/500th, gain about 800+ a bit, Gamma was default (100 I think??) Histo was at about70%

Sharpening was the top 3 wavelets @ 100 with a touch or two of denoise on the top slider

>CS3 had a touch of unsharp mask, some smart sharpen and a despeckle:icon_scratch:

I think that's about it:rolleyes:

Link to comment
Share on other sites
Clayton

Clayton

Posted
  • Author
Posted
That process has completely cured the fine grid, though I preferred the final red colouring of your initial processing, well maybe a little less red! I guess you can get that back in CS3. I just tried a quick RGB balance of your image in R6 and that gave it nice natural colours to my eye.

Cheers,

Chris

Thanks again for the feedback on this Chris:icon_salut:

I'm not too worried about colour ATM, but I'm flying blind re the grid, it's just not visible on either of my 2 quite different monitors:(

Link to comment
Share on other sites
Clayton

Clayton

Posted
  • Author
Posted
I was just wondering if it was possible to determine which light areas are clouds and which just light parts of the surface of mars. Could a filter do this ? Fabulous photo BTW.

Thanks Perrin6

Yes it is possible, a blue filter will highlight the clouds and a red filter will highlight the surface features, also a map will help you to learn and recognize permanent features and distinguish them from cloud.

Google has some nice map images, but I have about 3 of them to work from as they all differ:)

Link to comment
Share on other sites
Clayton

Clayton

Posted
  • Author
Posted
That big 'ole Dob is bearing fruit at last! Lovely image Rob! Chris's repro gets my vote! Does drizzling make any difference to the bayer pattern?

Thanks Stuart ;)

Slowly getting there:rolleyes:

I doubt if drizzling will help, but I will give anything a go:D

Link to comment
Share on other sites
Clayton

Clayton

Posted
  • Author
Posted
Thanks Rob. Might try a piece of blue plastic from a pair of 3d glasses as I don't have any colour filters and don't really want to buy any.

Should work fine ;)

And the Red as well:icon_salut:

Link to comment
Share on other sites
Kokatha man

Kokatha man

Posted
Posted
Thanks again for the feedback on this Chris:icon_salut:

I'm not too worried about colour ATM, but I'm flying blind re the grid, it's just not visible on either of my 2 quite different monitors:(

Looks like you go in for big, blurry images Stuart :icon_scratch: - Rob, that pattern is only visible when the image is over-blown to Hades and appears mainly in the centre and some peripheral areas.....totally inconsequential in the scheme of planetary imaging imho - but really I just want to reiterate that I think this has been a major progression in your imaging and more particularly that your getting that auto-dob beastie functionaing correctly, well done again..!:):icon_salut::)

Link to comment
Share on other sites
Clayton

Clayton

Posted
  • Author
Posted
Looks like you go in for big, blurry images Stuart ;) - Rob, that pattern is only visible when the image is over-blown to Hades and appears mainly in the centre and some peripheral areas.....totally inconsequential in the scheme of planetary imaging imho - but really I just want to reiterate that I think this has been a major progression in your imaging and more particularly that your getting that auto-dob beastie functionaing correctly, well done again..!:):icon_salut::icon_scratch:

Thanks Darryl :) I finally found a program that displays the offending pattern and tend to agree. I usually overdo the resampling at 200% anyhow:rolleyes: and if you bump that up by 400% :eek:the detail is going to be degraded to the point where I wouldn't expect anyone to be happy looking at it. Naturally I would prefer that it wasn't present at all, but I think I can manage it by moderating or altering my sharpening techniques:)

Link to comment
Share on other sites
Kokatha man

Kokatha man

Posted
Posted
You think Chris's version is big and blurry? :)

Well Stuart, I'm not here to bag anyone's images even if they're reworks of someone elses.....and in Chris's defence it is no bigger (48mm diameter on my smaller DVI screen) than Rob's PLUS he "shows it as it is" and you can't click on the image and bring up the real "blowsy" job like my old mate Rob's do....!

One of the reasons I reduced the size somewhat in my rework (which Chris essentially emulated in his own rework!) is to keep the scale under control vis a vis detail resolution, but nah, I'm not canning anyone per se except making the point that size at the expense of clarity seems to impress a lot of folks.....and you did sound impressed in your response - but it's all just harmless fun mate....!:);)

and ps: Sorry Rob - that'll be payback for turning one of my images into a psychedelic rainbow way back whenever!.....but regardless of your "big & blowsy" I really do think this is a great image - and have you taken a gander at the forecasts for later this week...and not just SkippSky's....?:):icon_scratch:

Link to comment
Share on other sites
cgarry

cgarry

Posted
Posted
Well Stuart, I'm not here to bag anyone's images even if they're reworks of someone elses.....and in Chris's defence it is no bigger (48mm diameter on my smaller DVI screen) than Rob's PLUS he "shows it as it is" and you can't click on the image and bring up the real "blowsy" job like my old mate Rob's do....!

One of the reasons I reduced the size somewhat in my rework (which Chris essentially emulated in his own rework!) is to keep the scale under control vis a vis detail resolution, but nah, I'm not canning anyone per se except making the point that size at the expense of clarity seems to impress a lot of folks.....and you did sound impressed in your response - but it's all just harmless fun mate....!;):icon_scratch:

and ps: Sorry Rob - that'll be payback for turning one of my images into a psychedelic rainbow way back whenever!.....but regardless of your "big & blowsy" I really do think this is a great image - and have you taken a gander at the forecasts for later this week...and not just SkippSky's....?:):)

I was hesitate calling the image I did 'a rework', much less emulating anybody else's rework. All I wanted to do was use the 'RGB Balance' in Reg6 so that the colour was a little more natural so that the 'equalising the RGB histograms before stacking' method of processing could be fairly evaluated. This method of equalising the histograms and using Reg6's 'RGB Balance' does produce consistent colours in my experience and I am actually pleased that they match the natural looking colours in your version of the image as this method does not require any tweaking of the colours by eye and can be repeated consistently session after session.

I assume Rob generated this image larger than the previous ones to show up the grid artefact we were looking to fix. So it is really a bit unfair to compare it to the smaller versions.

- Rob, that pattern is only visible when the image is over-blown to Hades and appears mainly in the centre and some peripheral areas.....totally inconsequential in the scheme of planetary imaging imho

I completely disagree with this, I think the grid pattern is a big problem with some 618 OSC cameras that a lot of people have been complaining about. I also think it can obscure fine detail in the images, which is a real shame when you have collected data as good as Rob's.

What is curious is that for some reason not everybody is as sensitive to seeing and it does not appear in everybody's images. Personally, I can easily see it in images at their natural size yet Rob does not see it at all. The point I am trying to make is that I think it is better if this grid pattern can be prevented from appearing in the first place by a simple operation of equalising the histograms before stacking, rather than using clever filters after stacking. As good as these filters are it is hard to tell if they have removed all of the problem or even worse if they has removed some useful detail.

It is still not known yet if equalising the colour histograms always fixes this grid problem, but Rob's test was another example of it appearing to work. If this technique does constantly fix the grid pattern then maybe the correct solution is to be sure the colour histograms are all equal at capture time rather than just trying to get a natural looking image on laptop screen.

Anyway, one thing we do agree on is this discussion should not detract from recognising the jump in quality that Rob's images have just taken. Not that they were exactly shabby before!

Cheers,

Chris

Link to comment
Share on other sites
Kokatha man

Kokatha man

Posted
Posted

I'm not one to try and limit people's quest to extend their knowledge nor challenge any existing tenets of processing etc Chris, but I don't think there's enough evidence in your histo equalisation application to justify your inference re the grid.....it would seem that merely changing file-types removes that "artefact" (if indeed it is a bayer matrix effect...)

It appeared as if it was removed from my tweaking merely by the act of taking a bmp screenshot of Rob's image here and dropping it into CS4...

Also, and without denigrating Rob's image one iota (him and I are good friends!) it is nowhere near high enough in resolution to even reveal said "effect" even when his largish image is further magnified - it takes a more than doubling in scale to do so.....and certainly not in any way capable of "degrading" any fine detail as I'm reading you state in your post!:)

Btw - I simply used "auto colour" in CS4 which is often a bit better than the R6 controls.....but I don't necessarily leave it at that (although I did in that "tweaking.")

And I'd have to say that histogram equalisation isn't a mantra I adopt either before or after capture tbh.....it is very easy to blow out all details in the NPC with green & blue histo levels set similarly to red.....

None of this is meant to be critical of your comments per se nor your right to make them - I'd rather have folks debating or at least stating their opinions rather than everyone nodding their heads or all agreeing (I'm a contrary b*$$#@! in fact, and enjoy stirring the pot and folk's set ideas) - so even if I'm not agreeing with your appraisal please don't take offence......just that we diverge in opinions on this matter somewhat....!:)

As to the 618 chip in a bayer setup you may well be correct re it being an issue, although I've seen plenty of good images of pretty high resolution that don't seem to manifest this "problem.":icon_scratch::icon_scratch:

Link to comment
Share on other sites
Space Cowboy

Space Cowboy

Posted
Posted
Well Stuart, I'm not here to bag anyone's images even if they're reworks of someone elses.....and in Chris's defence it is no bigger (48mm diameter on my smaller DVI screen) than Rob's PLUS he "shows it as it is" and you can't click on the image and bring up the real "blowsy" job like my old mate Rob's do....!

One of the reasons I reduced the size somewhat in my rework (which Chris essentially emulated in his own rework!) is to keep the scale under control vis a vis detail resolution, but nah, I'm not canning anyone per se except making the point that size at the expense of clarity seems to impress a lot of folks.....and you did sound impressed in your response - but it's all just harmless fun mate....!;):icon_scratch:

and ps: Sorry Rob - that'll be payback for turning one of my images into a psychedelic rainbow way back whenever!.....but regardless of your "big & blowsy" I really do think this is a great image - and have you taken a gander at the forecasts for later this week...and not just SkippSky's....?:):)

I agree that images can be over blown but to me Chris's softer version looks more natural and when comparing the images at same scale has superior cloud detail across the disk. Whether that is from the histo equalisation I don't know. A small amount of colour noise reduction cures the bayer pattern for me but I've rarely had any problem though AS!2 seems to show it up more (tried drizzle and its much worse).

If you want to see big and blurry Darryl wait till you see my f50 shots (if conditions allow) :). To be honest I think Rob is just showing off ;)

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

Go to topic listing

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue. By using this site, you agree to our Terms of Use.