I mostly concur with Volker on all of that. I'd also point out that there aree a few applications that are primarily designed to look after digital photos, so they do some of what you want.
My favourite is kimdaba, and digikam is also very good. Both are available for SuSE. kimdaba has a great indexing system, making it easy to ascribe keywords to images, for example a picture taken at my son's birthday party may be given a keyword for Miles (his name), party, the names of the other kids there, home (if thats where the party is), birthday - whatever else you like. You can then pull up all images by combinations of the keywords. The keywords are all in categories too, like people, location. Its really really easy to do the indexing and the searching. It uses certain preprogrammed directories to look for pics, so if you set up your scanner and fax programs to save to the right place, the images will be added to the database (but of course you have to index them). For photos all exif data is recognised, so the photos are already tagged with such esoteric stuff as the f-stop and focal length - if you ever need to know! The date and time is included in the exif data and that is very handy. There are links to the common editing programs, including gimp of course, as well kolourpaint. For scanning, xsane is the old standard, and in fact will view, save, fax, email or copy (to the printer) from the gui. sane is your basic scanning program mechanism, and in my experience has often got better drivers than the windows ones produced by the scanner manufacturers. In typical linux fashion it is divided into backends (which communicate with the scanner) and frontends (which communicate with you). The frontends go from command line to gui, and there is even a competent windows network client, so your stick in the mud win users can scan from the other end of the house. kde has a scanning frontend called kooka, which seems to do much the same stuff as xsane. OCR I have found to be a waste of time on any OS. Mind you I haven't tried it on windows for a long time. The good thing bout linux is tha ability script a lot of the stuff you need to do :-) On Sat, 2005-03-26 at 19:26 +1200, Volker Kuhlmann wrote: > With the multitude of your requirements, I'd say that all of it can > certainly be done in Linux, but there isn't much that does all of it in > a user-friendly way. The top two programs for image processing are gimp > (interactive use) and ImageMagick (non-interactive use). They won't > arrange your images for you though. I use my own method, involving cp, > mkdir, mv, rm, and a script which I've been using for a decade now and > which uses ImageMagick for the donkey work. All commandline though, as I > personally don't have a use for point'n'click in this situation. > > > documents together and also has a built in (or add on) OCR program which is > > quite effective, > > There is no usable OCR for Linux. The two open source solutions are for > specialised applications (scanning books) or practically useless. There > is a commercial one available for Linux which is almost certainly 100% > effective, but it is for seriously professional/commercial applications > and costs 4 digits. > > OCR is one of those problems for which there is no straightforward > algorithmic solution, and making an application therefore requires more > than just a bit of weekend programming. This takes it out of reach of > typical open source projects. Just like speech recognition, which is > even more extreme. > > Volker > -- Nick Rout <[EMAIL PROTECTED]>
