You are here

Converting RAW/NEF images to JPEG

As I already said 100 days ago, m68k hardware is still rocksolid:

spice:/home/ij# uname -a
Linux spice 2.6.23-m68k #1 Sun Nov 25 22:39:15 CET 2007 m68k GNU/Linux
spice:/home/ij# uptime
20:14:18 up 500 days, 6:42, 6 users, load average: 1.50, 1.37, 1.40
spice:/home/ij# uprecords
# Uptime | System Boot up
----------------------------+---------------------------------------------------
-> 1 500 days, 06:42:17 | Linux 2.6.23-m68k Wed Nov 28 11:56:41 2007
2 349 days, 23:37:09 | Linux 2.6.14-m68k-amiga Tue Dec 5 19:06:28 2006
3 66 days, 07:29:56 | Linux 2.6.14-m68k-amiga Fri Feb 3 01:14:11 2006

:-)

Kategorie: 
 

Comments

> for i in *.tiff ; do j=`echo $i | sed -e 's/tiff/jpg/g'` ; convert $i $j ; done

for i in *.tiff ; do convert "${i}" "${i%%.tiff}.jpg"; done

I don't use digikam, so I cannot comment on it's conversion quality. But one thing I'd like to suggest in my experience is that to get the best results, the conversion of a RAW image is best achieved by hand tuning each image. It takes time, but the result is often far superior. You can run a batch once you've tuned each file's conversion settings.

Well, currently I've >5000 files from last weekend. Some are duplicates (RAW+JPEG), some are just JPEGs, some are just RAWs. I think hand tuning >1000 is way too much work for a webarchive.
Of course that's something I'll do for selected images that will go to print... ;)

It sounds like mkdir, mv and rm will be one of the important tools in your photographic processing then ;)

I use ufraw for my raw images from a D50 Nikon camera. I remember I got better results than using dcraw. Also, if you have taken several pictures at the same light conditions, you can adjust some parameters and get a nice result after adjusting only the first one. Here is one example:

http://www.flickr.com/photos/tiagovaz/sets/72157609164944361/

you could try UFRAW

I already did try ufraw, but didn't get any better results, although it states better support for Nikon images (tone curve support or such) than dcraw. Maybe I missed some options that would do some magic... ;-)

I'd have to agree that hand-tuning is the best way to achieve good results, and that ufraw gives me better results than dcraw, but barring that, with NEFs (Nikon RAW) the closest to best is usually to extract the full-size jpeg from the NEF:

$ exiftool -b -JpgFromRaw somefile.nef > somefile.jpg

This requires off course that the full-size jpeg is in fact stored in the RAW file, but TTBOMK this is always the case with Nikon RAW photographs.

Thanks! This really is what I was looking for! :-)

I have tried digikam with D90 and it works fine if you use the RAW import tool to handle the image (select it form settings / image editor). There are various possibilities to choose in Raw decoding. When editing select in workspace the color profile you use (sRBG or adobe) and embedded camera profile and camera white balance.

Well, when I edited/batch converted the NEFs, I actually used the appropriate color profile embedded camera profile and camera white balance. But still the results were disappointing. Sadly. :(

Unless you manually set the white balance in the camera to match the conditions in which the images were exposed, the camera simply guessed what the white balance should be. Sometimes it does a good job, sometimes not, especially on a bright, sunny day (when the colour might be too cool). In post processing, one of the first things to do is set the correct white balance.

I shoot RAW-only (so no JPEG), but to have the F-Spot archive of a sane size I keep all the NEFs on an external drive and feed F-Spot the JPEGs resuliting from

dcraw -e *.nef

which provides the camera-generated ‘thumbnails’ – in reality, full-size JPEGs.

hi,
I shoot in RAW format and I need to convert and process my photographs in batch. I have been using contenta converter for almost a month now and I am very satisfied with the result.
Regards,
Gulati

But there's no Linux version of contenta converter. Bad, bad, bad...

Pages

Add new comment

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer