Pixel Mapping or Noise Reduction?

Kerbouchard

Member
Messages
38
Reaction score
0
Location
Crete, GR
Hello, I'm a new E-500 owner (had it for a whole week now, absolutely loving it!!!) :-)

I do have a question for some of the more experienced E500 users out there, about pixel mapping and noise reduction. I have taken a long exposure nightshot and used the in-camera NR, no other PP, but there are still some brightish red pixels throughout the pic, apparently noise, visible in the original 5MB file. I'm wondering if these are a result of the Oly NR processing, or possibly hot pixels which can be marked and ignored via pixel mapping.

Do the hot pixels which can be removed by pixmap , not the ones in a long exposure, show up differently in a picture, i.e brighter, larger, various colors, one color,etc?

And will pixmap, and/or a brand new camera, generally only show say less than a dozen hot pixels, or up to a hundred or so, like the red pixels in my picture?

I have done a lot of night work and astrophotograpy with my previous Oly C2100UZ, so I'm familiar with DFS, Blackframe NR, Registax for stacking multiple exposures, etc. and removing electronic noise. I must say, I am extremely impressed with the E500 NR feature! This picture is an 8-min ISO 200 exposure, with in-camera NR, and no other PP! WAY better than my 2100UZ could ever have done, with a max. 16-sec exposure...

Link to a smaller version of the pic, big one is also available at my site:

http://kerbouchard.smugmug.com/gallery/1151328/1/53704573

Thanks for any reply!
Cheers
 
Hello, I'm a new E-500 owner (had it for a whole week now,
absolutely loving it!!!) :-)
Welcome aboard Matt!!
I do have a question for some of the more experienced E500 users
out there, about pixel mapping and noise reduction. I have taken a
long exposure nightshot and used the in-camera NR, no other PP, but
there are still some brightish red pixels throughout the pic,
apparently noise, visible in the original 5MB file. I'm wondering
if these are a result of the Oly NR processing, or possibly hot
pixels which can be marked and ignored via pixel mapping.
Do the hot pixels which can be removed by pixmap , not the ones in
a long exposure, show up differently in a picture, i.e brighter,
larger, various colors, one color,etc?
And will pixmap, and/or a brand new camera, generally only show say
less than a dozen hot pixels, or up to a hundred or so, like the
red pixels in my picture?
I downloaded your full size image and looked it over carefully. IMO the numerous hot pixels in your photo represent ones that heated during your original 8 minute exposure but did not heat equally in the dark frame exposure. I've never taken an 8 minute star trail shot but I did accidently take a 1 minute ISO 100 trail shot when my telescope drive clutch was slipping :-). This shot (embedded below) shows a dozen or so hot pixels that weren't "trapped" by the dark frame routine. I would assume that as the exposure time (and ISO) increases the number of rogue hot pixels increases exponentially... certainly just a guess on my part. I would find it totally unreasonable that you have a hundred or so true hot pixels that need to be mapped out (no reason not to try your in camera pixel mapping though as it will do no harm). To the best of my knowledge, TRUE hot pixels on the CCD look just like those you get from long exposures; variable in color and generally just one pixel, but I've never noticed a hot pixel that needed to be mapped out on my E-1, E-300 or E-500 (either lucky or blind!?).
I have done a lot of night work and astrophotograpy with my
previous Oly C2100UZ, so I'm familiar with DFS, Blackframe NR,
Registax for stacking multiple exposures, etc. and removing
electronic noise. I must say, I am extremely impressed with the
E500 NR feature! This picture is an 8-min ISO 200 exposure, with
in-camera NR, and no other PP! WAY better than my 2100UZ could ever
have done, with a max. 16-sec exposure...
It's a nice pic! I really like it! Can't help but wonder though why on an 8 minute exposure you wouldn't have used ISO 100 to help control noise and get a cleaner shot?
Link to a smaller version of the pic, big one is also available at
my site:

http://kerbouchard.smugmug.com/gallery/1151328/1/53704573

Thanks for any reply!
Cheers
The following shots aren't all that good, but are merely illustrative. They all are full frame pics reduced in size for web viewing.

Accidental star trail shot of Orion (telescope drive motor clutch was slipping). E-500, 40-150mm kit lens, fl 150mm, f/4.5, ISO 100, 60 seconds.
Note the hot pixels that escaped the dark frame subtraction routine.



Same shot, same lens and parameters after I tightened my clutch. Undoubtedly some hot pixels escaped in this shot too, but who would ever know? :-)



One final shot to show the difference in quality of the ZD 50-200mm lens vs. the kit lens.
E-500, ZD 50-200mm, fl 200mm, f/3.5, ISO 100, 60 seconds.



Hope this is of some help to you,

Fred
 
Kerbouchard,

After reading your comments in Theresa's "How many lurkers are out there?" thread I took the liberty to repost your questions here. There are actually quite a few posters on this forum who are knowledgeable about astrophotography/night-shots and hopefully you will get some responses. Many of us forum irregulars have the same feelings you do about being a serial thread killer :-), but with all the activity threads do move on quite quickly and far too often without any response.

While I am neither a serious nor knowledgeable astrophotographer, I have dabbled a bit with it and will post my 2 cents worth in a followup as time allows.

I hope you don't mind that I've reposted your questions here!

Fred
Hi everyone. I'm still a newbie to this forum, and I'll do my best
not to be offensive! :-)

My question I have about the ME1 has to deal with the primary--or
only--use that I bought it for, which is using it to get a better
manual focus on stars at night for either star trails or mounted
either on or through my telescope. Has anyone tried it for focusing
on stars at night, and if so did it make a difference? The manual
focus with the focus indicator assistance, where the focus light
flashes then goes solid when it thinks you have the right focus,
doesn't work very well so far, for me anyway, when focusing on a
black sky with a few stars in it.

What I've had to do so far is focus on a house or streetlight in
the distance, then switch to manual and make sure I don't touch the
focus ring while I adjust the tripod & camera. This can be a pain,
since I either have to pick up the camera and tripod, or focus then
attach camera to tripod. But, I don't know how well this will work
when I piggyback the camera onto my telescope, which automatically
tracks objects across the sky, if I can't get the focus precise.

Does anyone know how far away a ground-based object has to be in
order to achieve focus on infinity? I'm using the ZD lenses that
came with the 2-lens kit, nothing different or more complicated.

Here's an example of my best pic, but from my inexperience with my
new E-500 I can't tell if the stars are fuzzy from the focus being
slightly off, or just because of the light gathered from the 8
minute exposure. Any thoughts, opinions? (the original 5MB pic is
also available on my smugmug site, kerbouchard.smugmug.com)

I am very happy with the way the camera caught the purple streak of
the Orion Nebula in the middle of the pic, and really eager to slap
this baby on the back of my telescope for a long, tracked exposure
of Orion!



Thanks for any input!

Also, I'm living on the Greek island of Crete, where we'd better
not see snow--even though it's cold enough right now to do it!! :-)
 
Hi everyone. I'm still a newbie to this forum, and I'll do my best
not to be offensive! :-)

My question I have about the ME1 has to deal with the primary--or
only--use that I bought it for, which is using it to get a better
manual focus on stars at night for either star trails or mounted
either on or through my telescope. Has anyone tried it for focusing
on stars at night, and if so did it make a difference? The manual
focus with the focus indicator assistance, where the focus light
flashes then goes solid when it thinks you have the right focus,
doesn't work very well so far, for me anyway, when focusing on a
black sky with a few stars in it.
I would think the focus indicator light would only work if there was something with contrast (a star) right under the AF sensor. Since you have the ME-1 can you tell me what your impression is of it's usefullness? I have considered getting one for the very same purpose as you, but don't know if it would be worth it.
What I've had to do so far is focus on a house or streetlight in
the distance, then switch to manual and make sure I don't touch the
focus ring while I adjust the tripod & camera. This can be a pain,
since I either have to pick up the camera and tripod, or focus then
attach camera to tripod. But, I don't know how well this will work
when I piggyback the camera onto my telescope, which automatically
tracks objects across the sky, if I can't get the focus precise.
I currently use one of two methods for attaining star focus:

Using the center autofocus point only, I point the camera to a bright star or planet and use autofocus (works well more times than not) and then switch to manual focus for recomposition and shooting. Be sure to turn off the sleep mode on the camera so it doesn't reset your lens while you're mucking around. It may take 3 or 4 half presses to get focus right, but you usually can tell if it "took" by the pinpoint star image in the viewfinder. This is where the ME-1 would seem to me to be very useful.

Alternatively I've used Oly Studio to shoot quick test jpeg images in tethered mode. This is the poor man's not-quite-live preview but you can get focus dead-on in this manner.
Does anyone know how far away a ground-based object has to be in
order to achieve focus on infinity? I'm using the ZD lenses that
came with the 2-lens kit, nothing different or more complicated.
I have a freebie utility called E System DOF written by Kevin St. John that works fine for giving hyperfocal distances. Unfortunately, I don't know where in the heck I got it or where to direct you. For the shot you took (a nice one by the way!) at 14mm and f/5.6 the hyperfocal distance calculates to be 8 ft. (focusing on anything at or past this distance will result in all further objects being in "reasonably acceptable" focus). At 45mm and f/4.5 the hyperfocal distance is 101 ft. With the 40-150 kit lens at 150mm and f/4.5 it would be 1,116 ft to get into the hyperfocal range. With the ZD 50-200 at 200mm and f/3.5 this distance is 2,551 ft. These figures are all well and fine, but I certainly prefer to focus on a celestial object directly!
Here's an example of my best pic, but from my inexperience with my
new E-500 I can't tell if the stars are fuzzy from the focus being
slightly off, or just because of the light gathered from the 8
minute exposure. Any thoughts, opinions? (the original 5MB pic is
also available on my smugmug site, kerbouchard.smugmug.com)
IMO you are in good focus with this shot. The varying size of individual star trails has more to do with sensor blooming of bright stars vs. dim stars. As an aside, I would highly recommend using ISO 100 for this type of shot to not only reduce noise but to avoid "overcooking" the brighter stars. What little I've done at ISO 200 (guided) shows this blooming and noise gain to be a problem.
I am very happy with the way the camera caught the purple streak of
the Orion Nebula in the middle of the pic, and really eager to slap
this baby on the back of my telescope for a long, tracked exposure
of Orion!
You will love it in piggyback mode! I live within the Oklahoma City metroplex and find that anything over about 1 minute in exposure time from my back yard is pretty much limited by light pollution. From a dark sky location it should be quite nice though. I look forward to seeing your further attemps!

Best of luck!

Fred


Thanks for any input!

Also, I'm living on the Greek island of Crete, where we'd better
not see snow--even though it's cold enough right now to do it!! :-)
 
If he had used ISO 100, it would have taken 16 minutes to get the same results...and I think the E500 is capped at an 8 minute exposure. I am not saying that he would have been unable to capture an image, but it would have definately been different...right?
It's a nice pic! I really like it! Can't help but wonder though
why on an 8 minute exposure you wouldn't have used ISO 100 to help
control noise and get a cleaner shot?
--

Life is like a roll of TOILET PAPER; the closer you get to the end, the faster it goes.
 
You are absolutely correct CA but the difference would only be one stop. I personally think the exposure is slightly overcooked on the sky and would benefit some from levels adjustment to darken the long exposure sky fog. Using ISO 100 would help accomplish this and at the same time keep the star trails the same length with less image blooming around the brighter stars. But in the end, it's Matt's exposure and I think he did a VERY nice job with it as is! I was just trying to plant the seed that often with long exposures a lower ISO gives more pleasing results (just my observation).

Fred
It's a nice pic! I really like it! Can't help but wonder though
why on an 8 minute exposure you wouldn't have used ISO 100 to help
control noise and get a cleaner shot?
--
Life is like a roll of TOILET PAPER; the closer you get to the end,
the faster it goes.
 
Wow! I just checked this thread again for the first time in a while, and I'm overwhelmed by the responses!! Thanks Mucho for the feedback!

Freddyg: Thanks for reposting my questions from the other thread, for the input on the hyperfocal ranges, for suggesting to see if I can get it focus on a star or planet, and especially your examples. The pic where your clutch drive slipped really stood out and slapped me, because of course hotpixels would be dots instead of streaks since they're ON the ccd, and not being picked up from externally like the stars which are moving across the ccd.

Also, if you got those pics of Orion with that much detail from only a 60-second exposure, I can't wait to see what we can get with 4-8 mins! Hoping for Barnard's loop even... Have you tried any longer pics?

On my pic, I increased the ISO (to 200, if I remember correctly, maybe 160) rather than 100 because a previous pic at 100 was darker, and I wanted to capture the fainter stars as well, and to get an idea of how much diference the ISO made. I also wanted to see how it worked on the church and the olive trees in the foreground, and I like the result. Mght have to tweak the contrast & gamma, but it looks good to me.

The skies here can get pretty dark, especially on the south side of the island where some places the nearest lights you see could be at Gavdhos island, about 40 miles to the south and with a population of about 12 people. From one of the small villages down there, I've seen a bright globular star cluster with my Canon IS 10X30 binoculars! The pic that I posted, though, was taken from my backyard which is about 5 KM from the airport here, and they leave the lights on all night....

I just got my ME-1 eyecup, and haven't been out to use it yet. I put it on and had to severely adjust the diopter to correct, so I think I'll keep it on all the time. Fortunately, my vision is still 20/20, no glasses or other problems, so I can see the exposure info inside the viewfinder clearly and without crushing my nose into the screen. However, some people have reported that they can't see this info with the ME-1 on, so if anyone is considering one, read through other posts about it.

Kjebre: Sorry, but the church is pretty firmly planted in my backyard, and the landlord might not like me sending it up to Norway... but it certainly makes a nice foreground, especially for sunrises!

Thanks again for the info and feedback! I know it'll help out!!

Cheers, Yamas!

Matt
 

Keyboard shortcuts

Back
Top