Kofax Blank detection and Fujitsu Fi-7260

Kofax Blank detection and Fujitsu Fi-7260

Postby archiegaw » Mon Jun 13, 2016 6:25 pm


Need help with Blank detection of Kofax Capture 10, with VRS 5.1.
Kofax doesn't seem to detect the blank pages, even when we used blank detection of KC, then KC and VRS, then VRS only.
the white pages keep coming out. anybody experienced this when installing Kofax?
the scanner is a new one and we find it troublesome to setup.

Fujitsu fi-7260 Component Installer from teh Kofax website has also been installed on the machine running windows 7.

Would appreciate tips/advise as we're not sure how to proceed :(

Posts: 3
Joined: Mon Mar 14, 2016 1:44 am

Re: Kofax Blank detection and Fujitsu Fi-7260

Postby bclaman » Wed Jun 15, 2016 6:55 am

have you tried just using the back delete threshold setting in Scan module - Scanner properties - Options - Threshold bytes?

Also are you using the ISIS or TWAIN driver? They seem to behave slightly different.
Posts: 258
Joined: Thu Sep 24, 2009 10:00 am
Location: Des Moines, IA

Re: Kofax Blank detection and Fujitsu Fi-7260

Postby russell@centuryc.com » Thu Jun 16, 2016 11:03 am

I can't tell you how the VRS blank page detection works, but I can tell you how the Capture one works.

As the scanner scans, the individual pages are saved as single-page TIFF files. If the resulting file is smaller than the threshold, then it will be discarded. Typical values I see are 1500-2000 as a starting point for letter sized pages.

The reason this works is the Group 4 TIFF compression. Instead of recording each pixel, Group 4 only records each transition between black and white. A completely white page will be very small, since there are no transitions. A normal page of text will have a very high number of transitions. Perhaps 50 or more per character. So it will be a significantly larger file. However, the system can't distinguish the reason for the transition. It could be text, it could be a black border, punched holes, dog ear, ragged edge, speckles, etc. ANYTHING that prevents the resulting page from being a solid white rectangle will affect the filesize and the resulting blank page detection.

And if you're doing grayscale or color, it really gets interesting. A grayscale/color will be far larger than a Group 4 black and white. And while I expect more complex pages to be larger than blank pages, I'm less confident on finding a reliable threshold.

And while I've never tested it, I think the whole thing goes out the window if the Batch Class properties are set to not compress the individual files. In that situation, filesize is strictly a function of the size of the image in pixels. In other words, a legal page will be larger than a letter. But all letter sized pages will be the same regardless of what is on them.
Posts: 3374
Joined: Wed May 17, 2006 12:53 pm
Location: USA

Return to Kofax Capture General Discussion

Who is online

Users browsing this forum: Google [Bot] and 5 guests