Skip navigation
Currently Being Moderated

HotPixels show up after several LR processing steps

Aug 10, 2012 3:00 AM

Tags: #saturated #lr4 #hotpixels

Dear all,


I am having two problems with Nikon Raw Processing and LR4.1.


1st: I am producing and processing content for ESO (European South Observatory), especially TimeLapses. Those sequences consist of hundreds of Images in hundreds of folders. Source files are either from D3s, D4, D700, D7000 or the like.


Several development changes are applied to single key images in a sequence, and those later applied to all images in that sequence. Problem is that after a certain amount of development changes (that are written in to the XMP sidecar files with LR4), hot pixels start to show up in the images that were not visible before.


Seems like the oversaturated pixel information (Hot Pixel) that is always stored within the RAW images, is lost after a certain amount of development changes.


2nd: is there a known trick how to force LR4 into updating ALL thumbnails after development changes, including those, that are outside the currently visible grid view?


It is very time consuming to manually page up and down the grid to have LR4 updating the thumbnails on each page (of for example a folder of 1200 images) so that one gets a preview of the TL sequence with the current development changes in Library View.


Thanks for any help



Bildschirmfoto 2012-08-10 um 11.57.59.png

  • Currently Being Moderated
    Aug 14, 2012 3:52 PM   in reply to starrysky01

    what do you mean with HOTPIXELS?

    are you sure it´s not the highlight clipping warning?


    can you show a screenshot of s single image with hotpixel visible?



    ps: in the 90s i build some stuff for your observatory. 

    Mark as:
  • Currently Being Moderated
    Aug 15, 2012 1:14 PM   in reply to starrysky01

    yeah i know what hotpixels are.

    but i never noticed that they appear or disappear after editing in LR.


    but then i don´t have seen much of them.



    starrysky01 wrote:




    about the stuff: cool. did you visit them too when they incorporated them into the designs? if so? where have you been at? so far I've been to ALMA and Paranal but not La Silla.



    nope unfortunately not.

    i worked for the gauß observatory in göttingen (germany) and we build some meassuring equipment for la silla and the observatory del teide on teneriffa.

    Mark as:
  • Currently Being Moderated
    Aug 15, 2012 1:19 PM   in reply to -Agfaclack-

    i see you used the panogear panoramic head.. what do you think about it?

    Mark as:
  • Currently Being Moderated
    Aug 15, 2012 3:08 PM   in reply to starrysky01

    starrysky01 wrote:




    no screenshot needed... With "hot pixels" I meant noise on chip during long exposures. oversaturated pixels, either yellow, white, green, blue etc etc. usually information about hot pixels is stored within the RAW file (as they are easily recognizable), but it seems that LR4 seems to loose that information after some processing runs.

    It's my understanding that only "manually mapped" hot/dead pixel information is contained in the raw file and this is usually done when the camera is manufactured or at a factory depot.


    Did you try increasing LR's Color and/or Luminance noise reduction settings? Astrophotography is a "worst case" scenario for making hot pixels visible (i.e. long exposure + high ISO + black background). Every digital camera has them, it's just a question of the ISO and exposure time setting required to make them visible.


    If you are still seeing a few hot pixels that don't respond to LR's noise reduction controls, try removing them at pixel level  zoom (8:1) using the adjustment brush at +100 Noise, Feather = 0, Flow = 100, Density = 100, and then 'Sync' the Local Adjustments> Brush settings across like images (i.e. same shoot and camera).

    Mark as:
  • Currently Being Moderated
    Aug 15, 2012 4:28 PM   in reply to starrysky01

    starrysky01 wrote:


    Hi Trshaner,


    yes, to all your suggestions. Tried them alreadym and it's in our workflow.


    With Astro-TimeLapse and/or panning motions it is even more complex. You immediately see the static hot pixels during a timelapse motion on a moving landscape.


    First you cannot use the adjustment brush on stars - it looks odd when they move trough and brush areas fill and empty with other stars during the Timelapse Motion. Second once you pan or have a dolly movement some brush areas "move" trough trees or other parts of the landscape. it looks very ugly. what you can do is to stop brushing the hot pixels there for the the couple of images, or do all by hand means adjusting the healing brush location outside the tree or other part of the landscape moving by.

    OK, I see how masking out the hot pixels just moves the problem. You really need a way to detect them and remove them on each image by sampling nearest neighbors and replacing the data values. I am not sure why more hot pixels seem to crop up after making a few develop adjustments. If you can provide an image with XMP file (or DNG) for download I'll be happy to see if I can "reverse engineer" the cause.


    During capture production don't use 'live view' mode on the camera. It heats up the image sensor and will greatly increase the hot pixel levels. What is the typical per image exposure time and ISO setting?



    starrysky01 wrote:


    Hi Trshaner,


    any idea on how to force LR4 to update ALL previews also those outside the grid are still very appreciated.


    this here illustrates my problem:

    (watch it until the end how it should look like on preview)



    The only way I know how to rebuild previews in the Library module is to first select the images, and then go to toolbar Library> Previews> Render 1:1 Previews.

    Mark as:
  • Currently Being Moderated
    Aug 16, 2012 6:48 AM   in reply to starrysky01

    Some Adobe employees do monitor the posts here, tbut he best way to get Adobe's attention is to post the Library previews issue at the Photoshop Feature Request site:



    I'll send you a PM concerning the test files and see if I come come up with some other suggestions.

    Mark as:

More Like This

  • Retrieving data ...

Bookmarked By (0)

Answers + Points = Status

  • 10 points awarded for Correct Answers
  • 5 points awarded for Helpful Answers
  • 10,000+ points
  • 1,001-10,000 points
  • 501-1,000 points
  • 5-500 points