New project: camera color profile calculation - page 17 - RAW Shooting and Processing - CHDK Forum
« previous next »
Pages: 1 ... 12 13 14 15 16 [17] 18 19 20 21 22 ... 49   Go Down

New project: camera color profile calculation

  • 480 Replies
  • 319214 Views
*

DawMatt

Re: New project: camera color profile calculation
« Reply #160 on: 25 / May / 2008, 05:24:58 »
Advertisements
Hi vit40,

Quote from: vit40 on 27 / January / 2008, 09:44:47
As I explained in DngSetmatrix readme, brightness of image made from dng depends on BaselineExposure tag value. So here is "DngSetExposure.dpr" (correct line marked "exposure value here" up or down if needed; value in header files is +50/100)
I've only ever been able to find the exe for DngSetmatrix (someone hosted it on zshare), not the readme.  Can you please post a link to the official hosting site for this program, including the readme file?

Update:
I managed to locate a full copy of the program after all.  The DPReview thread vit40 mentioned previously is still online, and links to the full DngSetMatrix package.

Thanks,
Matt
« Last Edit: 26 / May / 2008, 21:30:35 by DawMatt »
Logged
*

Offline fe50

  • ******
  • 3147
  • IXUS50 & 860, SX10 Star WARs-Star RAWs
    • fe50
Re: New project: camera color profile calculation
« Reply #161 on: 25 / May / 2008, 07:21:46 »
Quote from: DawMatt on 24 / May / 2008, 10:35:51
...
I've just finished adding support for the IXUS 860 IS to my copy and am testing it now, so its definitely doable.

Hi Matt, can you share this with us ?  :P
Logged
*

Offline brainwash

  • **
  • 95
  • Nikon D40x & A460
Re: New project: camera color profile calculation
« Reply #162 on: 26 / May / 2008, 07:42:00 »
I've tested DNG4PS2 with my S2IS and it works perfectly! Last time I used DNG4PS and the white balance/hue was all off and almost impossible to correct in Camera Raw (Photoshop). Now, with the custom profile loaded for S2 IS the RAW in its raw state is almost as good as JPG. Throw in some noise reduction technology and you've got DSLR-like quality ;)
Logged
*

DawMatt

Re: New project: camera color profile calculation
« Reply #163 on: 26 / May / 2008, 21:10:38 »
Quote from: fe50 on 25 / May / 2008, 07:21:46
Hi Matt, can you share this with us ?  :P
Have already done so, just hadn't advertised it yet!

I've added IXUS 860IS support to DNG4PS2, and have also made the calibration function more accessible.  Denis has released these changes in beta 25 of the product (available here).  So if you have downloaded a DNG4PS2 beta since Sunday then you already have this functionality (though it hasn't been mentioned in the changes log yet).

When I get a moment I'm planning to write up the calibration process in the wiki.  Hopefully we can streamline adding support for new cameras by allowing us camera owners to do more of the prep work for the developers.

Thanks,
Matt
Logged
*

DawMatt

Re: New project: camera color profile calculation
« Reply #164 on: 26 / May / 2008, 21:44:08 »
FYI: The matrix used for the IXUS 860 IS in beta 25:
Code: [Select]
IXUS 860 IS
0.747939, -0.268233, -0.129266,
-0.050633, 0.515687, 0.023509,
-0.021341, 0.076495, 0.186385
Logged
*

DawMatt

Re: New project: camera color profile calculation
« Reply #165 on: 26 / May / 2008, 22:51:55 »
Hi Cor,

Quote from: Cor on 20 / February / 2008, 05:37:45
I am new to RAW files, but the beauty in RAW files for me is that no in-camera manipulations has been done. The in-camera de-noising and color manipulations are the main reasons for me to shoot RAW.

If you compare jpg with RAW and make an "colour matrix", an important point of shooting in RAW is lost.
I understand the question, but Denis has taken that into consideration.

De-noising shouldn't be an issue because we are calibrating using slabs of monotone colour.  Thus the de-noising algorithm will (hopefully) use the monotone colour when removing noise, and we can average out the result to minimise the impact of noise.  The program also monitors the error level when calculating the matrix, so we can confirm the quality of the matrix before using it in the program.

In the first post in the thread you will notice a number of camera settings to be used before taking a calibration photo.  These are designed to reduce the amount of colour manipulation (especially the underexposing setting) the camera performs when creating the JPEG.  It also helps blur the colour swatches, which should help reduce noise.

So while Denis has clearly stated that these are approximate colour matricies, steps have been taken to minimise the issues you are concerned about.  And in most cases they will produce an adequate result.  More accurate colour matricies could be produced, but most people don't have the equipment/software/knowledge/patience to do so (I don't. Yet?).

Thanks,
Matt
Logged
*

Offline lks

  • *
  • 16
  • [A570IS, S5IS]
Re: New project: camera color profile calculation
« Reply #166 on: 27 / May / 2008, 00:42:28 »
Quote from: DawMatt on 26 / May / 2008, 21:10:38
When I get a moment I'm planning to write up the calibration process in the wiki.  Hopefully we can streamline adding support for new cameras by allowing us camera owners to do more of the prep work for the developers.

Sounds great -- I would like to try it on the S5IS.  If you do write up the procedure, please let us know.
Logged
*

Offline fe50

  • ******
  • 3147
  • IXUS50 & 860, SX10 Star WARs-Star RAWs
    • fe50
Re: New project: camera color profile calculation
« Reply #167 on: 27 / May / 2008, 03:14:45 »
Thx Matt for your work, i'll test it the next days...

Quote from: DawMatt on 26 / May / 2008, 21:10:38
...When I get a moment I'm planning to write up the calibration process in the wiki.  Hopefully we can streamline adding support for new cameras by allowing us camera owners to do more of the prep work for the developers.
This would be great, perhaps i can work at a SD400 profile then...
Logged
*

Offline hiker_jon

  • ****
  • 284
Re: New project: camera color profile calculation
« Reply #168 on: 28 / May / 2008, 10:50:12 »
I have an A720.  DNG4PS works great for me, but I would like to also use DCraw directly on my camera raw files.  DCraw can read the data, but its colors are way wrong.  Does anyone know how to modify the source for DCraw to get the colors right?  I find the code in that program particularly hard to understand.
Logged
*

Offline brainwash

  • **
  • 95
  • Nikon D40x & A460
Re: New project: camera color profile calculation
« Reply #169 on: 28 / May / 2008, 13:05:40 »
Are you using DCRaw as-is without any frontend? You should try UFRaw as frontend. RAW files usually have no color balance set, this is usually and advantage since you could set it manually without inducing banding in the histogram. These front-ends have options for setting color balance (cyan-magenta is the most used in-camera) and also an auto-balance setting.
Logged
Pages: 1 ... 12 13 14 15 16 [17] 18 19 20 21 22 ... 49   Go Up
« previous next »
 

Related Topics