[sane-devel] Please help for my Xsane settings
Hi All, I've been playing with the settings for quite a while and the scanning results are still not good enough. I was using Xsane 0.996 under Ubuntu 9.10 to scane blank writing pad paper with rule-lines. Note the paper is white and has no ink at all, except those rule-lines. When I scanned this kind of paper, the result image is always too bright to see the rule-lines (which are my concerns). Case1: true-gray model, 600dpi, gamma = 0.3 (minumun), brightness = -100(minumun), contrast = 100(maximum), I get a result with partial visible rule-lines, but the quality in general is not good. I tried to modify the threshold parameters in the Preference->setup->enhancement, but no luck -- maybe I can't fully understand those threshold minimum(0), maximum(20), multiplier(12.5). Anybody can help? Case2: Black&White, 600dpi, It didn't matter what gamma, birghtness, etc were, the results were almost always a white image -- no rule lines are visible at all. Note that if I scanned it in a color model, I can see the rule-lines easily. However, I don't want to do the follow-up converting work; I want everything that comes out of the scanner to be original and raw. If possible, I'd like to scanned in both gray model and BW model respectively, and at the same time make the rule-lines as clear as possible. If anybody has done this before or have any ideas how to play with those parameters, I'd be more than happy to hear. I appreciate your help! Sincerely, Jin Chen -- next part -- An HTML attachment was scrubbed... URL: <http://lists.alioth.debian.org/pipermail/sane-devel/attachments/20100125/950c58a3/attachment.htm>
[sane-devel] Has anyone managed to get Snapscan backend working with FreeBSD 8.0 (which uses libusb)?
Hi, On Sun, Jan 24, 2010 at 9:51 PM, Oliver Schwartz wrote: > > what does you snapscan.conf look like? If libusb is used, you mustn't use a > device entry such as "/dev/..." or "libusb:/dev/...". All that is needed is > the entry containing the vendor and product ID of your scanner, i.e > "usb 0x06bd 0x2061" and the firmware entry. > > My snapscan.conf is correct, it only contains the firmware line and vendor and product ID, sorry for not mentioning this in my first posting. In the meantime, another FreeBSD user found a way to fix it, it is just a one-line patch: http://lists.freebsd.org/pipermail/freebsd-usb/2010-January/008172.html With Andrew's patch, my scanner now works: root at kg-v2# sane-find-scanner -q found USB scanner (vendor=0x03f0 [HP], product=0xc202 [Photosmart 8200 series]) at libusb:/dev/usb:/dev/ugen1.2 found USB scanner (vendor=0x06bd [AGFA], product=0x2061 [ Snapscan1212u_2]) at libusb:/dev/usb:/dev/ugen2.2 root at kg-v2# scanimage -L device `snapscan:libusb:/dev/usb:/dev/ugen2.2' is a AGFA SNAPSCAN 1212U_2 flatbed scanner and also as a normal user: tingo at kg-v2$ scanimage -L device `snapscan:libusb:/dev/usb:/dev/ugen2.2' is a AGFA SNAPSCAN 1212U_2 flatbed scanner -- Regards, Torfinn Ingolfsen -- next part -- An HTML attachment was scrubbed... URL: <http://lists.alioth.debian.org/pipermail/sane-devel/attachments/20100125/d50caed9/attachment.htm>
[sane-devel] Anyone interested in a Medion MD6190 ?
I have a Medion MD6190 collecting dust for a while now. This USB1.1 scanner is only supported under Windows XP. Already did some USB tracing, but never got to the point of grasping the protocol and implementing anything. For example I know there is some firmware sent, and the device then reboots under a new USB-id. Was planning to do some automated tracing using Python-TWAIN, but well, never got that far. If there's anybody interested and knowledgeable, just drop me mail, and I'll see if I can ship it to you without me breaking the bank. Henk Poley
[sane-devel] Mustek ScanExpress A3 USB 1200 PRO
On Sun, Jan 24, 2010 at 11:34 PM, Scot wrote: >> by m. allan noah-3 Jan 18, 2010; 08:52pm >> >> On Sat, Jan 16, 2010 at 2:57 PM, scar wrote: >>> >>> since i require an inexpensive A3 scanner, i plan on purchasing one of >>> these soon, i have not found any other options. ?Mustek appears to have >>> a linux driver on their Taiwanese website, although i read reports here >>> of it not working. ?if all else fails i will attempt to install Windows >>> XP in a virtual machine and use the scanner through there, which was >>> also a possible solution i read on this list. >>> >>> naturally i would want to help develop a working sane driver. ?i do not >>> program but i have a degree in computer engineering so my background >>> should be adequate enough to provide some help, given there is someone >>> to provide me with suitable advice on how precisely to help. >> >> Sure- If you get one of these, we would need you to get a trace of the >> windows driver making a small black and white scan, using this tool: >> >> http://www.pcausa.com/Utilities/UsbSnoop/ >> >> That should give us enough info to at least tell if the protocol is >> similar to other scanners. >> >> allan > > The requested snoop output is bundled together with the corresponding > small black and white image and posted at: > > ?http://www.drivehq.com/file/df.aspx/publish/mustekscanner/ > ?PublicFolder/mustek-a3-1200-pro.zip > > Please let me know what else I can do to help! Well, it does alot of small control transfers, probably setting up registers. I'm not well-versed in such machines. Perhaps someone else will recognize it, or have time to compare to the existing mustek backends. allan -- "The truth is an offense, but not a sin"
[sane-devel] Xerox Travel Scanner 100 could not be detected
> Stef, > On my Visioneer Strobe XP 100, revision 3, I can calibrate and clear > calibration from within XSane without a problem. > > But when I try: > $scanimage -d genesys --clear > or > $scanimage -d genesys --calibrate > I get the following: > scanimage: sane_start: Error during device I/O > > (With the command line calibrate, it does feed the calibration sheet into > the scanner, just like it does with xsane). > > So, maybe Andrey could try to calibrate using xsane, instead of the command > line. In xsane, click on the "window" tab and check "show advanced > options" and you should get a calibrate button. > > Jack > There's a problem when you calibrate with command line. Once calibration is done, a scan is started. You have to put in a document so the scan can go on so there is no error. To avoid this you have to add the -n or --dont-scan option on the scanimage command line used: scanimage -d genesys --calibrate --dont-scan When default scanner settings aren't goo enough, it may not be able to recognize the calibration target which must consist in a pure white sheet with a full width horizontal black strip. Regards, Stef
