Hi - to anyone else who wants to use open_bayer() before anything later than 0.18.2 is stable-released, this is a link to the Github page at the earliest (least risky) version which supports open_bayer(). https://github.com/LibRaw/LibRaw/tree/d4c4f5c8e4a2c5db56287f805d44b590ac...
This works perfectly for 16bit (EDIT: and now 8bit) RAW images without any header or metadata.
(wish I could delete comments but I can only edit them so I must look really silly with two comments in a row)
I was serious that I do color, whitebalance, exposure, etc corrections to the raw data before putting it into LibRaw. I wrote the adjustments, and only use LibRaw to demosaic. No dcraw or LibRaw emulation of dcraw involved for color-correction / whitebalance / etc.
(Output with output_bps = 8, no_auto_bright = 1, no_auto_scale = 1, user_black = 0.) Every pixel is 0, and this was an overexposed 8bit raw. Do what you will with this information; I was just trying to repay a favor, thinking I might have found unintended behavior. I see now that LibRaw uses 16bit internally, even if the input buffer was only 8 BPP.
So, what seems to be happening...
1. 8 BPP data in the buffer has max value of 255
2. The buffer is read into LibRaw, which uses 16 BPP data internally
3. no_auto_bright = 1, so values are never scaled up to 65535
4. output_bps = 8, so values are all divided by 257 to scale down to 255 maxval... but at most, we had (255 / 257), which with integer rounding down equals 0. Anything less than 255 will of course also become 0.
5. Every pixel is 0 in the resulting image!
If that's correct (?), let this post just be a warning to any 8bit - LibRaw users who read this!
Hi Iliah,
I am using dcraw_emu (as I used dcraw 9.27 previously) as a simple tool to convert a raw file to a tiff. I'm trying to keep it very simple and use existing tools created by people such as yourself who are experts in many of these things. My program acts as a controller and uses these existing tools (such as dcraw_emu) to perform specific tasks. I'm not a C programmer and thus I don't have the ability to rebuild a custom version of dcraw_emu.
My need is simple: I just need dcraw_emu to create tiff files that are good quality (color-accurate) versions of the image that was captured in the raw file. The result should be quite similar to the original embedded JPEG and to the results produced by other raw converters. In almost all cases, dcraw_emu is doing that. The greens in my X-E2 sample pictures seem to be an exception.
As a further comparison, I converted my sample file with Lightroom and with Affinity Photo (I can provide samples if you wish). The result was:
* The original embedded JPEG and the tiff file produced by dcraw 9.27 were the brightest green (almost the same).
* The version produced by Lightroom was close to the embedded JPEG but the greens weren't quite as bright.
* The Affinity Photo version was quite a bit yellower than the previous versions. I don't think that it provided a very accurate version of the greens.
* The Dcraw_emu version came out similar to the Affinity Photo version (but lighter). Converting it with the brightness turned down further might produce a very similar version to Affinity Photo.
Going back to the dcraw 9.27 color profiles could be an unnecessary step backward. In my testing of the Sony ARW files, I found that the version produced by dcraw_emu has more accurate colors than the dcraw 9.27 version (which was much yellower). Thus, for my purposes, a small change to the color profile for the X-E2 (and the other X cameras if they are the same) should give me everything that I need from dcraw_emu.
To match JPEG colors you need not dcraw.c color profiles, but profiles used by camera vendors for in-camera processing (unavailable in most cases).
Also, the main problem is to repeat vendor's tone/contrast curve (which, in turn, changes when photographer change camera contrast/saturation/color mode) to match midtone brightness and highlights compression.
So, if your target is 'embedded JPEG match', than you (very likely) need own RAW postprocessing code instead of LibRaw::dcraw_process()
I'm sorry I do not understand why you can't use this solution? On a side note, why use colour processing from dcraw at all, why not to use quality colour transforms, since you develop anyway?
If that means that an upcoming version of dcraw_emu will use the dcraw color profile(s) that would be really good. Mainly, I just need the colors to match up with the colors in the embedded JPEGs. So far that seems to be the case, except for this one example.
dcraw.c has not changed for more than one year now (also, may-2016 change was minor, so real freeze is about 2 years).
Because of that, maintaining full dcraw.c compatibility is obsolete goal :)
Meanwhile, for cameras supported in (last) dcraw.c we (very likely) will change color profile to dcraw's even if we think our profile is better.
Since I'm using dcraw_emu, I can't use that solution. If the intent with dcraw_emu was to keep it consistent with dcraw, wouldn't it make sense to use that (revised) profile in dcraw_emu?
auto brightness calculates image histogram, than adjust (increase) values to fit full 16-bit range (with 1% of pixels at saturation with default auto_bright_thr). So, if you use linear exposure correction (w/o highlights compression), than your changes will be ruined (indeed), because brightness is adjusted to use full data range.
Do not know what 'color adjustment' you use,because no such mechanics exists in dcraw_process() options (excluding white balance).
Without auto-bright, 8-bit images use only 0..255 range (to be correct,it may be more if white balance is used). This is 8 stop below full range (0..65535), so resulting image will look 'near black' (not completely black).
Just rebuilt raw.dll (only ended up having to edit dcraw_common.cpp, thanks for the tips), and openbayer_sample() works with the patch! Thank you so much for your fast help. This is awesome!
I have found that to get what I want from LibRaw, I need to set no_auto_bright = 0 for 8bit images, and no_auto_bright = 1 for 16bit images. This is no problem to me, but the behavior is a little bizarre. The rest of this comment is to help you figure out if this is an unintentional error within LibRaw.
I do some precise exposure compensation and color correction before using LibRaw. After some testing, I have found a bizarre result when I set imgdata.params.no_auto_bright:
__________________|___8bit image____|__16bit image___ no_auto_bright = 0 |___100% correct___|__Earlier exposure / color adjustments are ruined no_auto_bright = 1 |_Completely black_|_100% correct___
(By the way, when I say "100% correct," I mean that my earlier exposure adjustments were conserved :) )
The results for 16bit images in the table above make perfect sense.
The results for 8bit images don't make any sense to me! I would expect results exactly like 16bit. Instead, in 8bit, no_auto_bright = 0 behaves like I would expect no_auto_bright = 1 to behave, and no_auto_bright = 1 makes the entire image all-black no matter what.
The difference is in camera color profiles used in LibRaw and dcraw.
To get the same color as in dcraw, please go to adobe_coeff() function in (LibRaw)/internal/dcraw_common.cpp
and replace our color profile (two lines:
{ "Fujifilm X-E2", 0, 0,
{ 12066,-5927,-367,-1969,9878,1503,-721,2034,5453 } },
to dcraw.c profile:
{ "Fujifilm X-E2", 0, 0,
{ 8458,-2451,-855,-4597,12447,2407,-1475,2482,6526 } },
The sample file in question was created with my X-E2 camera. I uploaded some screenshots and a copy of the file. The converted files are both a bit brighter than the embedded JPEG because I used a brightness of 1.
Hi Alex,
First of all, this specific problem was my own doing. I was using the -4 DCRaw parm that apparently does create some issues. Some of the files that I compared with had been converted without that parm. When I removed that, DCRaw_emu worked the way I had hoped and in a very similar way to DCRaw 9.27 (but faster).
There are still a couple of small differences from DCRaw 9.27:
* Fuji X files - these are coming out with the green a sort of yellow-green. This is different from the result with DCRAW 9.27 and different from the embedded JPEG.
* The resulting file name when I convert to a Tiff file is in the format filename.raf.tiff instead of filename.tiff
Otherwise, DCRaw_emu seems to do what I need it to do.
Chris
PS: I looked on the Data Structures page for -c so that's why I couldn't find it.
Hi Alex,
Thank you for the quick reply. I tried it with -c 0. I had the brightness set to 2.3 (leaving it at 1 made everything too dark). I did not see any difference using -c 0. Some images came out darker than the result from DCRaw 9.27, some came out lighter and one came out very similar. Obviously, changing the brightness level would change which was most similar. For most files, the version from DCRaw 9.27 (with brightness set to 1) was pretty consistent with the brightness level of the embedded JPEG from the RAW files (the version from DCRaw_emu was not).
Any other thoughts? I'd really like to use your version because of its other benefits but this complicates it.
NOTE: I did not see - c mentioned in your documentation. In the DCRAW 9.27 documentation, it says that -c is used to: "Write decoded images or thumbnails to standard output."
Yes, this is bug in open_bayer() code, the standard tone curve (color.curve[] array) was not initialized to linear curve, but it is used in 8-bit decoder.
Just add these lines:
for (int i = 0; i < 0x10000; i++)
curve[i] = i;
To void CLASS initdata() (anywhere in the function body). This call is in
* dcraw/dcraw.c if you're using snapshot (and re-generate script/makefile)
* internal/dcraw_common.cpp (if you're do not regenerate this file from LibRaw's dcraw.c).
(you may patch both source files if you're not sure, if re-generate script is used, than dcraw_common.cpp to be generated from dcraw.c if needed)
I found a workaround to make LibRaw accept the file (and other 8bit files from the same camera)... I store each 8bit pixel in memory as if it were a 16bit pixel, then have LibRaw output 8bits. I cannot keep using this because it is exactly twice the amount of IO that should be done!
Hope it's an easy fix
P.S. might be helpful -
This one does not have 255-level pixels at the beginning, they are quite dark in the upper left.
It is in the "RGGB" RGRGRG.../GBGBGB.../ Bayer pattern
LibRaw contains 'automatic adjust maximum' code to not get 'pink clouds' problem when maximum data level reported by camera is higher than real data maximum (this is common problem with canon files esp. if intermediate iso are used).
To disable this feature use -c 0 option of dcraw_emu (imgdata.params.adjust_maximum_thr = 0 if you use LibRaw directly from your code, not dcraw_emu.exe)
Hi - to anyone else who wants to use open_bayer() before anything later than 0.18.2 is stable-released, this is a link to the Github page at the earliest (least risky) version which supports open_bayer().
https://github.com/LibRaw/LibRaw/tree/d4c4f5c8e4a2c5db56287f805d44b590ac...
This works perfectly for 16bit (EDIT: and now 8bit) RAW images without any header or metadata.
(wish I could delete comments but I can only edit them so I must look really silly with two comments in a row)
Alex fixed the problem with open_bayer() 8-bit images. Fix is documented here https://www.libraw.org/comment/4678#comment-4678 :)
I was serious that I do color, whitebalance, exposure, etc corrections to the raw data before putting it into LibRaw. I wrote the adjustments, and only use LibRaw to demosaic. No dcraw or LibRaw emulation of dcraw involved for color-correction / whitebalance / etc.
I wish it were the case that the resulting 8bit images without auto_bright were 'near black,' but take a histogram of this result and see for yourself:
https://drive.google.com/file/d/0B41mEzFmIzubTkhfSk9zMmtVVnc/view?usp=sh...
(Output with output_bps = 8, no_auto_bright = 1, no_auto_scale = 1, user_black = 0.) Every pixel is 0, and this was an overexposed 8bit raw. Do what you will with this information; I was just trying to repay a favor, thinking I might have found unintended behavior. I see now that LibRaw uses 16bit internally, even if the input buffer was only 8 BPP.
So, what seems to be happening...
1. 8 BPP data in the buffer has max value of 255
2. The buffer is read into LibRaw, which uses 16 BPP data internally
3. no_auto_bright = 1, so values are never scaled up to 65535
4. output_bps = 8, so values are all divided by 257 to scale down to 255 maxval... but at most, we had (255 / 257), which with integer rounding down equals 0. Anything less than 255 will of course also become 0.
5. Every pixel is 0 in the resulting image!
If that's correct (?), let this post just be a warning to any 8bit - LibRaw users who read this!
Thank you for the explanation, and we hear you. It is just that the word "accurate" is not the one I would use.
Hi Iliah,
I am using dcraw_emu (as I used dcraw 9.27 previously) as a simple tool to convert a raw file to a tiff. I'm trying to keep it very simple and use existing tools created by people such as yourself who are experts in many of these things. My program acts as a controller and uses these existing tools (such as dcraw_emu) to perform specific tasks. I'm not a C programmer and thus I don't have the ability to rebuild a custom version of dcraw_emu.
My need is simple: I just need dcraw_emu to create tiff files that are good quality (color-accurate) versions of the image that was captured in the raw file. The result should be quite similar to the original embedded JPEG and to the results produced by other raw converters. In almost all cases, dcraw_emu is doing that. The greens in my X-E2 sample pictures seem to be an exception.
As a further comparison, I converted my sample file with Lightroom and with Affinity Photo (I can provide samples if you wish). The result was:
* The original embedded JPEG and the tiff file produced by dcraw 9.27 were the brightest green (almost the same).
* The version produced by Lightroom was close to the embedded JPEG but the greens weren't quite as bright.
* The Affinity Photo version was quite a bit yellower than the previous versions. I don't think that it provided a very accurate version of the greens.
* The Dcraw_emu version came out similar to the Affinity Photo version (but lighter). Converting it with the brightness turned down further might produce a very similar version to Affinity Photo.
Going back to the dcraw 9.27 color profiles could be an unnecessary step backward. In my testing of the Sony ARW files, I found that the version produced by dcraw_emu has more accurate colors than the dcraw 9.27 version (which was much yellower). Thus, for my purposes, a small change to the color profile for the X-E2 (and the other X cameras if they are the same) should give me everything that I need from dcraw_emu.
To match JPEG colors you need not dcraw.c color profiles, but profiles used by camera vendors for in-camera processing (unavailable in most cases).
Also, the main problem is to repeat vendor's tone/contrast curve (which, in turn, changes when photographer change camera contrast/saturation/color mode) to match midtone brightness and highlights compression.
So, if your target is 'embedded JPEG match', than you (very likely) need own RAW postprocessing code instead of LibRaw::dcraw_process()
> I just need the colors to match up with the colors in the embedded JPEGs. So far that seems to be the case
Ummm.... "Match" means, in a very relaxed environment, "within maximum of 4..6 dE".
I'm sorry I do not understand why you can't use this solution? On a side note, why use colour processing from dcraw at all, why not to use quality colour transforms, since you develop anyway?
If that means that an upcoming version of dcraw_emu will use the dcraw color profile(s) that would be really good. Mainly, I just need the colors to match up with the colors in the embedded JPEGs. So far that seems to be the case, except for this one example.
dcraw.c has not changed for more than one year now (also, may-2016 change was minor, so real freeze is about 2 years).
Because of that, maintaining full dcraw.c compatibility is obsolete goal :)
Meanwhile, for cameras supported in (last) dcraw.c we (very likely) will change color profile to dcraw's even if we think our profile is better.
Since I'm using dcraw_emu, I can't use that solution. If the intent with dcraw_emu was to keep it consistent with dcraw, wouldn't it make sense to use that (revised) profile in dcraw_emu?
These results are expected (more or less).
auto brightness calculates image histogram, than adjust (increase) values to fit full 16-bit range (with 1% of pixels at saturation with default auto_bright_thr). So, if you use linear exposure correction (w/o highlights compression), than your changes will be ruined (indeed), because brightness is adjusted to use full data range.
Do not know what 'color adjustment' you use,because no such mechanics exists in dcraw_process() options (excluding white balance).
Without auto-bright, 8-bit images use only 0..255 range (to be correct,it may be more if white balance is used). This is 8 stop below full range (0..65535), so resulting image will look 'near black' (not completely black).
Just rebuilt raw.dll (only ended up having to edit dcraw_common.cpp, thanks for the tips), and openbayer_sample() works with the patch! Thank you so much for your fast help. This is awesome!
I have found that to get what I want from LibRaw, I need to set no_auto_bright = 0 for 8bit images, and no_auto_bright = 1 for 16bit images. This is no problem to me, but the behavior is a little bizarre. The rest of this comment is to help you figure out if this is an unintentional error within LibRaw.
.............................................................................................................................
I do some precise exposure compensation and color correction before using LibRaw. After some testing, I have found a bizarre result when I set imgdata.params.no_auto_bright:
__________________|___8bit image____|__16bit image___
no_auto_bright = 0 |___100% correct___|__Earlier exposure / color adjustments are ruined
no_auto_bright = 1 |_Completely black_|_100% correct___
(By the way, when I say "100% correct," I mean that my earlier exposure adjustments were conserved :) )
The results for 16bit images in the table above make perfect sense.
The results for 8bit images don't make any sense to me! I would expect results exactly like 16bit. Instead, in 8bit, no_auto_bright = 0 behaves like I would expect no_auto_bright = 1 to behave, and no_auto_bright = 1 makes the entire image all-black no matter what.
.............................................................................................................................
Hope that was helpful for moving on with Release 0.19!
The difference is in camera color profiles used in LibRaw and dcraw.
To get the same color as in dcraw, please go to adobe_coeff() function in (LibRaw)/internal/dcraw_common.cpp
and replace our color profile (two lines:
{ "Fujifilm X-E2", 0, 0,
{ 12066,-5927,-367,-1969,9878,1503,-721,2034,5453 } },
to dcraw.c profile:
{ "Fujifilm X-E2", 0, 0,
{ 8458,-2451,-855,-4597,12447,2407,-1475,2482,6526 } },
Than re-compile/rebuild libraw/dcraw_emu
Thanks.
dcraw processed file (with -w to get camera balance) is definitely more green.
We'll investigate the difference.
Here is another link directly to the file:
https://1drv.ms/u/s!AhvHWo83K_oNjxRpT9I4Xzo00Z1X
Sorry, your link allows only download of screeenshot_embedded_JPEG.JPG, no RAW to play with
The sample file in question was created with my X-E2 camera. I uploaded some screenshots and a copy of the file. The converted files are both a bit brighter than the embedded JPEG because I used a brightness of 1.
Link
dcraw's -4 option results in
- linear gamma (-g 1 1)
- no automatic brigtening (-W)
- 16 bit output (-6)
There is no -4 option for dcraw_emu, but you may use -g 1 1 -W -6 instead.
Fuji X: what exact camera and what Libraw version?
.rawext.tiff - yes, this is intended option to allow convert filename.cr2 and filename.dng in the same folder.
Hi Alex,
First of all, this specific problem was my own doing. I was using the -4 DCRaw parm that apparently does create some issues. Some of the files that I compared with had been converted without that parm. When I removed that, DCRaw_emu worked the way I had hoped and in a very similar way to DCRaw 9.27 (but faster).
There are still a couple of small differences from DCRaw 9.27:
* Fuji X files - these are coming out with the green a sort of yellow-green. This is different from the result with DCRAW 9.27 and different from the embedded JPEG.
* The resulting file name when I convert to a Tiff file is in the format filename.raf.tiff instead of filename.tiff
Otherwise, DCRaw_emu seems to do what I need it to do.
Chris
PS: I looked on the Data Structures page for -c so that's why I couldn't find it.
-с is, indeed, documented: https://www.libraw.org/docs/Samples-LibRaw.html
-c float-value
This key sets params.adjust_maximum_thr parameter.
Use -c 0 to completely disable automatic maximum calculation.
Default value: 0.75
Could you please provide some samples to test against dcraw/dcraw_emu. With -c 0 it should be the same, at least with default settings.
Hi Alex,
Thank you for the quick reply. I tried it with -c 0. I had the brightness set to 2.3 (leaving it at 1 made everything too dark). I did not see any difference using -c 0. Some images came out darker than the result from DCRaw 9.27, some came out lighter and one came out very similar. Obviously, changing the brightness level would change which was most similar. For most files, the version from DCRaw 9.27 (with brightness set to 1) was pretty consistent with the brightness level of the embedded JPEG from the RAW files (the version from DCRaw_emu was not).
Any other thoughts? I'd really like to use your version because of its other benefits but this complicates it.
NOTE: I did not see - c mentioned in your documentation. In the DCRAW 9.27 documentation, it says that -c is used to: "Write decoded images or thumbnails to standard output."
Thanks for the sample.
Yes, this is bug in open_bayer() code, the standard tone curve (color.curve[] array) was not initialized to linear curve, but it is used in 8-bit decoder.
Just add these lines:
for (int i = 0; i < 0x10000; i++) curve[i] = i;To
void CLASS initdata()(anywhere in the function body). This call is in* dcraw/dcraw.c if you're using snapshot (and re-generate script/makefile)
* internal/dcraw_common.cpp (if you're do not regenerate this file from LibRaw's dcraw.c).
(you may patch both source files if you're not sure, if re-generate script is used, than dcraw_common.cpp to be generated from dcraw.c if needed)
Link to full patch: https://www.dropbox.com/s/xqi4ox7ui4r5cqg/8bit-openbayer.patch?dl=0
(sorry, our forum does not allow attachments to topic comment, it is easier to put it on dropbox than re-tune our forum).
This full patch also patches openbayer_sample.cpp to get correct parameters to this call.
Hi Alex, no worries! Hope it was good vacation time.
Here is my 8bit raw file on Google Drive:
https://drive.google.com/file/d/0B41mEzFmIzubcmxKMnVEcE14MEE/view?usp=sh...
Here it is demosaiced to a .tif using other software (with worse algorithms!):
https://drive.google.com/file/d/0B41mEzFmIzubZW0wMXE3YWo3eFk/view?usp=sh...
I found a workaround to make LibRaw accept the file (and other 8bit files from the same camera)... I store each 8bit pixel in memory as if it were a 16bit pixel, then have LibRaw output 8bits. I cannot keep using this because it is exactly twice the amount of IO that should be done!
Hope it's an easy fix
P.S. might be helpful -
This one does not have 255-level pixels at the beginning, they are quite dark in the upper left.
It is in the "RGGB" RGRGRG.../GBGBGB.../ Bayer pattern
LibRaw contains 'automatic adjust maximum' code to not get 'pink clouds' problem when maximum data level reported by camera is higher than real data maximum (this is common problem with canon files esp. if intermediate iso are used).
To disable this feature use -c 0 option of dcraw_emu (imgdata.params.adjust_maximum_thr = 0 if you use LibRaw directly from your code, not dcraw_emu.exe)
Pages