There was a discussion recently on Twitter over whether the ScanSnap was supported in Linux.
Out of the box, none of the models come with Linux drivers – the S510 and S300 come with Windows software, and the S510 and S300M come with Mac OSX software.
However, there is a project called SANE – Scanner Access Now Easy that’s goal is to enable as many scanners as possible in Linux.
This post by Mark J Cox outlines how he got the S510 to work in RedHat Enterprise Linux.
# echo “usb 0x04c5 0×1155″ >> /etc/sane.d/fujitsu.conf
Then you can use any scanning front-end, or from the command line say you wanted to scan at 150dpi colour, double-sided, then use “scanimage -L” to figure out where your scanner is, and replace the 005:004 below with the location:
# scanadf –device fujitsu:libusb:005:004 –source “ADF Duplex” –mode Color -v –resolution 150 –y-resolution 150
Mark’s post is from September 2007 and from looking at the SANE documentation, it looks like both the S510 and S300 are on the list of supported devices, so it may actually be that Mark’s hack is no longer required as long as you have the latest sane-backend (at least 1.0.19).
Do you have the ScanSnap working in Linux or any other *nix flavor? Let us know in the comments how it’s working for you and how easy it is to do.
Related posts:
Just plugged mine into my Ubuntu box. Started xsane. Scanned. To do nice document scanning, gscan2pdf seems to be a nice solution. (Just install it from repository on Ubuntu)
Cool, thanks for your feedback cycloon!
SANE works just fine with those scanners now – they're now under their supported scanner list.
For the S300/S300M scanners, you need to use the PC-based firmware file (300_0C00.nal) from the Windows device drivers (c:\windows\ssdriver\s300 directory). Once you get that file, you need to edit the epjitsu.conf configuration file to where you placed it or you can just put it in the default location specified in the config file.
I tried the the S300M firmware file from the MAC driver, but I couldn't get it working so I recommend just using the S300 firmware file even if you have an S300M (it seems to be compatible.)
You might also have to put a PCI DevID entry in for the S300M (as I seem to remember that's not in their by default) if you're using that scanner – which is 0x117F (Fujitsu VenID 0x04c5.)
The S300 doesn't work with the fujitsu driver – apparently it's different than the other fujitsu scanners. For the S300 you need to use the epjitsu backend.
Thanks a lot for letting us know about the S300 Richard. I appreciate it.
sane-backends 1.0.20 was recently released. It features significantly updated drivers for both fujitsu and epjitsu. Its a highly recommended upgrade.
I'm actually re-writing a good portion of the epjitsu driver now. I previously re-wrote the calibration routines to get better color balance and even lighting, but these new routines only work with the S300. Now I'm refactoring the whole driver so that the FI-60F can also be supported. Hopefully this will be included in the next sane release.
Has anyone looked at the S1500/M under Linux yet?
I've had no reports of it.
I got it running this evening after installing 1.20 under Jaunty. I had to use an "untrusted" repository, but it works great!! Thanks!!!
I just tried it under Ubuntu (sane-fujitsu 1.0.21) and it wasn't recognized. I'll be digging into it (hints welcome
)
OK, the installed version is 1.0.19 (the man page is 1.0.21) in Jaunty. The updated backend doesn't seem to be available in the trusted software sources.
To duplex scan with this printer, go to the Edit–>Preferences setting in gscan2pdf and choose the "scanadf" frontend instead of "scanimage". Then use "Extended Page Numbering" with an increment of "1" to prevent numbering by 2.
gscan2pdf is intelligent enough to advance the starting number of the next batch to the next number in the scans.
I am really digging my scanner!!
Awesome, thanks so much for the info!
Above post regarding the ScanSnap S510
how do i get the scansnap s1300 working?
The S1300 uses the sane-epjitsu backend, while the other ScanSnaps use sane-fujitsu. Could that be why you are having trouble? Also looks like if your main o/s is Linux, the S1500 is more thoroughly supported than the S1300, so it might be worth spending the extra $$$ if you have room for the bigger one.
I want to buy a Fujitsu Scansnap 1500 Deluxe to work under Ubuntu. My question is: Does the Fujitsu OCR work?
Well, the OCR software that ships with the ScanSnap is a Windows EXE so it wouldnt work under Linux without emulation of some sort.I know that some people do thru scanning and OCR in Windows and then do all the rest in Linux but that may or may not be religiously acceptable.
As for Linux OCR options, I'll have to dig into that.
Thanks a lot for answer! I know that these days OCR is really weak in Linux as an OSS application. So, we are waiting for OCRopus from Google (http://code.google.com/p/ocropus/) right?
All the best and
warm regards from Germany,
HK
Have you tried tesseract?
Any news about the ScanSnap s1100? Any help appreciated. Cheers.
i’m curious about the S1100 as well. Thinking of buying but want to make sure I can use it in linux first.
It seems from this as if the epjitsu backend is in testing for the s1100, so you might want to check in on bios you can help with the testing? No personal experience, I'm afraid. http://www.sane-project.org/cgi-bin/driver.pl?man…