I have been using Darktable for some now for digital pictures but I
recently borrowed a slide/negative scanner and used it to scan some old
negatives I had. While searching I also discovered a couple of rolls of
undeveloped film from years back which I got developed.
Looking at the colours from the negatives they seemed a little odd but when
I got the recently discovered films back from developing complete with CD
scans from the developer these have turned out to have very different
colours from those I get if I scan the same negatives myself.
Here an example:
http://s16.postimg.org/tchwpmfwh/img22highres.jpg
This is the scan that came back from the people developing the film.
http://s16.postimg.org/l5pwy1ptt/img22raw.jpg
This is what I get if I scan the film myself.
I also have an example where water is a very obvious purple. As I have
already scanned quite a few images and had to return the scanner, at least
for the time being, I am wondering if can work out a colour correction to
be applied to the images from the scanner I borrowed. Moving the sliders
in the white balance or colour balance modules to search for useful
settings does not seem to work very well. Taking samples from patches of
what should be the same colour on each of the scan from the developers and
the scan from the scanner I borrowed, working out the ratio for each colour
and then setting that as the gain in the colour balance module works but
only for that patch, i.e. that patch then becomes the same colour in the
corrected version from my borrowed scanner and the scan from the developers
but the rest of the picture retains noticeably incorrectly coloured.
So this suggests to me that the film or some aspect of the scanner does not
respond linearly and the extent of the non-linearity is different for each
colour. This is presumably what the other sliders in the colour balance
module are intended to correct but again moving them around whilst visually
checking is not very productive.
I know there is a step in adding a new camera to produce a colour matrix
which can be done given an IT8 target where there are various shades of
exactly known colours but I can't see a way to do that in this case. IT8
targets do not seem to be available on negative film and the films I have
been scanning are old so even buying an IT8 target, finding a film camera
and a similar roll of new film, photographing the target, developing the
film and then running the new camera colour matrix tool on the resulting
scan would not necessarily yield the right results as the film formulation
may have changed in the intervening years (20 or more, in some cases).
Could a similar process work for multiple patches where for each patch a
sample (r,g,b) is taken from the desired colour and another (r,g,b) for the
uncorrected colour picking, for example, both strong colours and neutral
areas, shadows and highlights? Does anyone know of a tool that would be
able to process these to work out the parameters for colour balance (or as
a colour matrix to add as if it was a new camera)?
If not what other approach would you use?
------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=157005751&iu=/4140/ostg.clktrk
_______________________________________________
Darktable-users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/darktable-users
