Home · Register · Join Upload & Sell

Moderated by: Fred Miranda
Username  

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Post-processing & Printing | Join Upload & Sell

1
       2       end
  

Archive 2015 · // Significant color noise during long exposures //

  
 
Sigur Ros
Offline

Upload & Sell: Off
p.1 #1 · p.1 #1 · // Significant color noise during long exposures //


Hello guys ,

I have a issue here , it driving me crazy .. )

During long exposure shoots i have a significant colour noise on my pictures,i just bought 16-35 f/2.8 II L - lens and i didn't have this problem before.

I read online that i can try enabling the in-camera option to automatically implement noise reduction during long exposures but i don't now if that is a good compromise as it is an auto option done by camera?

Noise dots are all over my picture somewhere i have bigger colour artefacts to call them like that ) , please any help or advice would be appreciated.

I attached samples for review, thanks.




Jun 23, 2015 at 06:02 AM
dhphoto
Offline
• • • • • •
Upload & Sell: Off
p.1 #2 · p.1 #2 · // Significant color noise during long exposures //


That doesn't look like noise, they look like hot pixels which inevitably show up on long exposures. The older your sensor gets the more dots you will probably see

If you can use a RAW converter like Lightroom they will magically disappear, alternatively you can use the Canon pixel mapping software (that I have never used)

You can try this non-destructive technique which some swear by - take off the lens and put on a body cap, put the camera into manual sensor cleaning mode and leave it for a minute or two, apparently this can trigger the internal pixel mapping software, but I have never tried (because I use Lightroom)

Edited on Jun 23, 2015 at 06:24 AM · View previous versions



Jun 23, 2015 at 06:17 AM
Sigur Ros
Offline

Upload & Sell: Off
p.1 #3 · p.1 #3 · // Significant color noise during long exposures //


dhphoto wrote:
That doesn't look like noise, it looks like hot pixels which inevitably show up on long exposures. The older your sensor gets the more dots you will probably see

If you can use a RAW converter like Lightroom and they will magically disappear, alternatively you can use the Canon pixel mapping software (that I have never used)



Thanks dhphoto,

I actually opened them in Lightroom rather than in my default DxO and i can't see the the hot pixels anymore,care to explain why i can't see them in Lighroom ? , is the LR program automatically mapping them out ?

Also this lens is less then 1 year old as i bought it second hand but i have guarantee list , so how its possible it has so many hot pixels already as the sensor is pretty new ?

Thanks



Jun 23, 2015 at 06:23 AM
dhphoto
Offline
• • • • • •
Upload & Sell: Off
p.1 #4 · p.1 #4 · // Significant color noise during long exposures //


It's not the lens that causes the pixels, it's the sensor, it's just one of those things, a bit like flecks of dust on a negative.

See my edit above, you can try that to see if you can eliminate the dead pixels.

Yes LR just edits them out, some RAW converters do, some don't (DPP doesn't)



Jun 23, 2015 at 06:26 AM
Sigur Ros
Offline

Upload & Sell: Off
p.1 #5 · p.1 #5 · // Significant color noise during long exposures //


Thanks great explanation


Jun 23, 2015 at 06:33 AM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.1 #6 · p.1 #6 · // Significant color noise during long exposures //


The red and green dots will be reduced considerably by using the Long Exposure Noise Reduction feature. As yo may know, this makes the camera take a second so-called dark frame exposure following your initial exposure. It uses the same exposure time and ISO but records only the non-scene image data — e.g. noise and hot pixels. These data are then subtracted from your "real" photograph of the scene to remove (or at least greatly reduce) the hot pixels and similar artifacts that can otherwise become more obvious during very long exposures on digital cameras.

This is a different sort of noise than what we usually are concerned with. It is not the random noise spread across the entire image that we might see at high ISOs. Instead, this is the result of some photo sites being "hot" and recording a strong signal in the red, green, or blue channel — hence the little points of red, green, or blue light.

Some post processing software (e.g. - Lightroom, ACR) will automatically map out the hot pixels and you may find that at least some of them simply disappear when you open the files there.

Dan

Sigur Ros wrote:
Hello guys ,

I have a issue here , it driving me crazy .. )

During long exposure shoots i have a significant colour noise on my pictures,i just bought 16-35 f/2.8 II L - lens and i didn't have this problem before.

I read online that i can try enabling the in-camera option to automatically implement noise reduction during long exposures but i don't now if that is a good compromise as it is an auto option done by camera?

Noise dots are all over my picture somewhere i have bigger colour artefacts to call them like that ) ,
...Show more

Edited on Jun 25, 2015 at 10:03 AM · View previous versions



Jun 23, 2015 at 10:53 AM
redcrown
Offline
• • •
Upload & Sell: Off
p.1 #7 · p.1 #7 · // Significant color noise during long exposures //


Long ago I read a technical paper that explained (and demonstrated) that both noise and "hot" pixels are increased by the heat of the sensor. Higher ISO and longer exposures both increase the heat in the sensor. Heat also builds up with multiple exposures, as little time is allowed for heat to dissapate.

In the demo, the author fired off multiple frames at constant long exposures. I don't remember the number of frames, ISO, or shutter speed, but when the first frame was compared to the last frame the difference was clear. More noise and more hot pixels.

The main lesson was to let camera rest between shots when doing long exposure.

Of course, that was long ago when sensor technology was less advanced. But I'd guess that newer sensors with much higher pixel density are actually more prone to the problem of heat build up.



Jun 23, 2015 at 01:11 PM
Aaron D
Offline
• • •
Upload & Sell: Off
p.1 #8 · p.1 #8 · // Significant color noise during long exposures //


What you explained in the demo you saw can be true. My old XS/1000D would act like that. These are referred to as either "hot" or "stuck" pixels. However, I once read that if its red, green, or blue...it's a "stuck" pixel. If its white (r,g,b all firing)...it's a "hot" pixel. If its black...it's a "dead" pixel.

LR does a good job making them disappear. There's also a few other programs available specifically for this. I've used a couple, but forget the names. One only worked on jpg's. The other had to be given referrence images, as in lens cap on, long exposure, so that the resulting referrence pic was black and showed the troublesome pixels only. These programs were usually more trouble than they were worth.

I found LR dealt with the problem best. In some cases LR might miss one, in which case I'd just clone it out.



Jun 23, 2015 at 04:22 PM
RustyBug
Offline
• • • • • • •
Upload & Sell: On
p.1 #9 · p.1 #9 · // Significant color noise during long exposures //


Aaron D wrote:
I once read that if its red, green, or blue...it's a "stuck" pixel. If its white (r,g,b all firing)...it's a "hot" pixel. If its black...it's a "dead" pixel.


Interesting info ... I thought that folks were just tossing the terms around loosely (likely some were) to all mean the same thing.



Jun 23, 2015 at 04:32 PM
Sigur Ros
Offline

Upload & Sell: Off
p.1 #10 · p.1 #10 · // Significant color noise during long exposures //


Great info Thanks


Jun 24, 2015 at 08:48 AM
Greg Campbell
Offline
• • • •
Upload & Sell: Off
p.1 #11 · p.1 #11 · // Significant color noise during long exposures //


Noise and Hot Pixels are much the same thing, just taken different degrees.

There are several freeware astronomy related programs that can subtract a dark frame from your images. (You can do a quick job and dirty with a few layers.) The idea is that you have a library of dark frames representing various shutter speed and ambient temperatures. find the one that best matches and load it into the software. The program will subtract the noise and attempt to fill the resultant holes with data sampled from nearby pixels.



Jun 24, 2015 at 04:36 PM
Greg Campbell
Offline
• • • •
Upload & Sell: Off
p.1 #12 · p.1 #12 · // Significant color noise during long exposures //


The in-camera reduction works much the same way, but requires the camera to take an equally long dark exposure immediately after every 'real' exposure. This delay can be annoying.


Jun 24, 2015 at 04:38 PM
dhphoto
Offline
• • • • • •
Upload & Sell: Off
p.1 #13 · p.1 #13 · // Significant color noise during long exposures //


Greg Campbell wrote:
Noise and Hot Pixels are much the same thing, just taken different degrees.



?

Noise can be controlled or eliminated by careful exposure, using lower ISO's and noise reduction software

Hot/stuck pixels are entirely different and are essentially 'flaws' in the sensor, no amount of noise removal will touch them



Jun 25, 2015 at 01:16 AM
atwl77
Offline
• •
Upload & Sell: Off
p.1 #14 · p.1 #14 · // Significant color noise during long exposures //


dhphoto wrote:
no amount of noise removal will touch them


While true in the strictest/technical sense, I have to mention that some noise reduction software/algorithms include hot pixel removal. In Capture One Pro, for example, this is under the Single Pixel setting of their noise reduction tab.



Jun 25, 2015 at 05:29 AM
dhphoto
Offline
• • • • • •
Upload & Sell: Off
p.1 #15 · p.1 #15 · // Significant color noise during long exposures //


atwl77 wrote:
While true in the strictest/technical sense, I have to mention that some noise reduction software/algorithms include hot pixel removal. In Capture One Pro, for example, this is under the Single Pixel setting of their noise reduction tab.


Very possibly, but still hot pixels and noise are certainly not the same thing at all



Jun 25, 2015 at 07:15 AM
gdanmitchell
Offline
• • • • • • •
Upload & Sell: Off
p.1 #16 · p.1 #16 · // Significant color noise during long exposures //


Greg Campbell wrote:
The in-camera reduction works much the same way, but requires the camera to take an equally long dark exposure immediately after every 'real' exposure. This delay can be annoying.


A few points.

You are absolutely right that the dark frame exposure method (Long Exposure Noise Reduction, or LENR) increases the time spent exposing each photograph. If you are doing a 10 minute exposure for your photograph, the ensuing dark frame exposure takes another 10 minutes — a total of 20.

Is it worth it? The answer is a bit subjective and each photographer will have to decide.

Since the in-camera dark frame exposure always has exactly the same settings as the original photograph it is likely to be more accurate than the method described above in a previous post, which uses benchmark photographs made at a different time and perhaps under different conditions (e.g. temperature) as the starting point for determining the data to be subtracted from the frame.

Looked at one way, the doubled exposure time (regular exposure + dark frame exposure) is an annoyance or a problem, especially since you are most likely to need it with the longest exposures. When I mentioned a 10 minute exposure taking 20 minutes I selected those times because they are in the range where you might want to employ LENR. Make a 30 minute exposure and you are talking about spending one hour.

That may be bad, but because digital cameras do not suffer from reciprocity failure in the way that film cameras do, the exposure times are still shorter than they were back in the good old days of film. In addition, if you are getting the R, G, and B hot pixels in your long exposure images, and you are trying to produce high quality work, the time is worth it. And for really long exposures there are alternatives, including blending multiple shorter exposures.

One time some years ago I made a few long exposures using LENR while on a backpacking trip into the eastern Sierra Nevada. It was a very dark night, lit only by less-than-full moon light, and I think that exposures were in the 14-15 minute range. Each photograph took over a half hour by the time I set things up, composed it (in the dark), selected camera settings, completed the main exposure, and then waited for the LENR dark-frame exposure to complete. I was initially frustrated by the delay. But then I found a comfortable place to sit along the edge of the lake, sat back, and enjoyed the deep quiet and stillness of the night. (It did not turn out to be one of my best night photographs, but that's the way it goes.) :-)

Dan

http://gallery.gdanmitchell.com/gallery/var/albums/NaturalWorld/TheLandscape/California/SierraNevada/OtherSierraNevada/BlackAndWhite/ThousandIslandMoonCove20070727BW.jpg

Edited on Jun 25, 2015 at 04:34 PM · View previous versions



Jun 25, 2015 at 08:09 AM
butchM
Offline
• • • • •
Upload & Sell: Off
p.1 #17 · p.1 #17 · // Significant color noise during long exposures //


gdanmitchell wrote:
if you are getting the R, G, and B hot pixels in your long exposure images, and you are trying to produce high quality work, the time is worth it.

Dan


I agree. I quite often use LENR when shooting night metro traffic scenes or fireworks and much prefer to handle the noise and/or hot pixel issues in-camera. It can be annoying waiting for the process to complete ... but quite often, if I do my calculations correctly before tripping the shutter release, there is little to do once I download the images. Which is a great bonus when working on tight deadlines.



Jun 25, 2015 at 08:18 AM
redcrown
Offline
• • •
Upload & Sell: Off
p.1 #18 · p.1 #18 · // Significant color noise during long exposures //


I'm confused. Why would you spend an extra 10, 20, or 30 minutes letting the camera do "LENR" when many post processing options can do the same thing in seconds? The disadvantage is obvious, what's the advantage?


Jun 25, 2015 at 11:52 AM
AmbientMike
Offline
• • • • • •
Upload & Sell: Off
p.1 #19 · p.1 #19 · // Significant color noise during long exposures //


I haven't read all the way down , but as far as the op's question, I get a lot of noise at a minute or two on aps and m4/3. FF might give a longer ss, since ff is generally a lot better on noise.

Odd colored pixels seem to disappear in my older version of DPP. Just haven't had a problem with them.



Jun 25, 2015 at 12:02 PM
Peter Figen
Offline
• • • • •
Upload & Sell: On
p.1 #20 · p.1 #20 · // Significant color noise during long exposures //


"I'm confused. Why would you spend an extra 10, 20, or 30 minutes letting the camera do "LENR" when many post processing options can do the same thing in seconds? The disadvantage is obvious, what's the advantage?"

Because it's not the same as any other type of noise reduction. It uses something called dark field subtraction and that requires, in order to be most effective, a second exposure exactly the same length as the initial, but without the shutter being open, so it's just dark, which is then effectively layered with the original but using the equivalent of the Subtract blending mode in Ps. That second, dark field exposure, needs to be the same time as the original in order to accurately reproduce the inherent noise in the sensor at that exposure time, which is then subtracted from the image file.

Now, you can shoot your dark field exposures at a later time with a lens cap on and do the dark field subtraction in Ps as part of your post production, but most people find it more convenient to have the camera do it automatically.



Jun 25, 2015 at 12:26 PM
1
       2       end




FM Forums | Post-processing & Printing | Join Upload & Sell

1
       2       end
    
 

You are not logged in. Login or Register

Username       Or Reset password



This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.