Equipment & Techniques > Beginner's Questions

Noise reduction in camera or in post

(1/2) > >>

sanfairyanne:
I heard some people saying they fire off a dark exposure prior to shooting night photography (perhaps a shot with the lens cap on) then they use that dark exposure whilst processing. This method apparently negates the need to do in camera NR which of course doubles the length of time to take a photograph.

I wonder if anyone can tell me what is the processing technique whereby this dark shot (i.e the shot with say the lens cap on) is used in processing.

Also is there one method which is considered best. i.e. In camera NR, NR in a raw convertor or NR through a plug in?

Many thanks in advance.

razrblck:
If you shoot RAW files, all processing settings (as well as NR) do not apply to them but only to their previews. The RAW data is preserved and if you load the files in Lightroom you start basically from scratch. Any additional NR data saved in the proprietary files can only be read and used by propietary software.

Basically any additional in camera processing only makes you waste time if all you shoot is RAW files.

fdisilvestro:
There are two different types of noise reduction in most cameras:

- High Iso or standard Noise Reduction: Applied after the image is taken, similar to what you can do in post processing and it does not affect the RAW file.

- Long Exposure Noise Reduction (LENR): The method you referred in the OP where a dark exposure is made and then substracted to the image. This does affect the RAW file, since the camera performs the substraction of the image and dark frame before recording the raw file.

The inconvenience of this method is that for every image you take, the camera records a dark frame, doubling the time it takes to have the camera ready for the next image.

One option to overcome this is to capture a dark frame separately and then do the substraction in post. The dark frame shoud be taken with the same ISO and exposure time as your image.

The issue here is that popular raw converters such as LR or C1, do not have the option to handle the dark frames, and you are left with less user-friendly applications such as RawTherapee, DCRaw or astrophotography programs. Some people perform the substraction in photoshop but it is not technically correct, unless you develop you raws using a linear profile.

sanfairyanne:
Thanks for clarifying this Francisco. perhaps Long Exposure Noise Reduction should be renamed Hot Pixel removal or Dark Frame Subtraction.

fdisilvestro:

--- Quote from: sanfairyanne on January 13, 2016, 08:35:38 am ---perhaps Long Exposure Noise Reduction should be renamed Hot Pixel removal or Dark Frame Subtraction.

--- End quote ---

Hot Pixel removal is different from Dark Frame subtraction. Even if you don't have hot pixels, the Dark Frame subtraction will be effective for long exposures.

I think the name Long Exposure Noise Reduction is easier to understand to most people and you don't have control of the process anyway (The in-camera)

Hot Pixel removal or suppression is another type of noise reduction, which I did not include in my previous post. For instance most Nikon DSLRs apply Hot Pixel Suppression at shutter speeds longer than 1/4 sec and you cannot disable or bypass it unless you hack the firmware (not recommended).

Main difference:

Hot Pixel Suppression: Analysis of neighbour pixels looking for outliers (hot pixels)
Dark Frame Subtraction: Subtract the pixel value when there is no image (noise) from the pixel value with image (noise + signal)

Navigation

[0] Message Index

[#] Next page

Go to full version