Mike Chaney's Tech Corner
August 11, 2020, 01:17:17 AM *
Welcome, Guest. Please login or register.
Did you miss your activation email?

Login with username, password and session length
News: Feb 2013: Qimage Ultimate Challenges... have fun and explore features!
 
   Home   Help Search Login Register  

Professional Photo Printing for Windows
Print with
Qimage and see what you've been missing!
Pages: [1]
  Print  
Author Topic: Determining scanner "quality"  (Read 23145 times)
Gourdfather
Newbie
*
Posts: 15



View Profile WWW
« on: June 14, 2009, 01:57:19 AM »

Is there a target and/or procedure for quantitatively determining the quality/capabilities of a scanner??  I have a scanner that claims it will scan up to 4800 ppi (optical, not software).  I'd like to determine if 4800 is actually better than 1200 or 2400 ppi, maybe "better" isn't the correct term....I'd like to test 1200, 2400 and 4800 and see the actual results...is the 2400 or 4800 fuzzy or sharp??

Any ideas??  Thx

John  Phf
Logged

John  Phf
admin
Administrator
Forum Superhero
*****
Posts: 3111



View Profile Email
« Reply #1 on: June 14, 2009, 01:51:02 PM »

That's an interesting question.  The challenge might be finding something that actually has 4800 PPI of detail.  Test prints may not be the way to go: maybe you could scan some sort of woven fabric or something with lots of micro detail and then just scan at various resolutions to see how much each one picks up.

Mike
Logged
Gourdfather
Newbie
*
Posts: 15



View Profile WWW
« Reply #2 on: June 14, 2009, 05:31:18 PM »

Yes, that might give me a visual on how detailed the scanner would go, and God knows we have plenty of fabric!!  (Our business is  www.fabricdepotco.com  )  LOL

Seems that I have seen a printed "target" for a better word, where different resolutions were printed out for comparison (It sort of looked like a TV test pattern??).  That could be scanned and resolutions checked to see whih ones it starts to get fuzzy on.   I would imagine that it couldn't be a file that you print yourself because then you introduce the printer resolution capabilities.  I'll try the fabric and keep looking,  Thx

John  Phf
Logged

John  Phf
admin
Administrator
Forum Superhero
*****
Posts: 3111



View Profile Email
« Reply #3 on: June 14, 2009, 05:47:44 PM »

You might be able to print a really fine pattern on some inkjet shrink film: remember the old Shrinky Dinks?  Smiley  Print some patterns at maybe 300 PPI and then it shrinks to 20% it's original size and you get 1500 PPI.  I don't know how well that stuff holds together when it shrinks though.

Mike
Logged
Gourdfather
Newbie
*
Posts: 15



View Profile WWW
« Reply #4 on: June 15, 2009, 03:29:18 AM »

LOLOL, "Shrinky Dinks" were AFTER my time!!!  ;-)  BUT......

http://www.shrinkydinks.com/

Ya never know....

John  Phf
Logged

John  Phf
Gourdfather
Newbie
*
Posts: 15



View Profile WWW
« Reply #5 on: June 15, 2009, 11:18:26 PM »

This might work.....

http://dpnow.com/1263.html
Logged

John  Phf
Stephen Brennan
Newbie
*
Posts: 9



View Profile WWW
« Reply #6 on: July 07, 2009, 09:46:28 PM »

When I worked in IT we used Kodak targets for testing scanners, something like this one:

http://www.aig-imaging.com/mm5/merchant.mvc?Screen=PROD&Store_Code=AIIPI&Product_Code=QA-60&Category_Code=Video-Scanner-Resolution-Charts
Logged
Gourdfather
Newbie
*
Posts: 15



View Profile WWW
« Reply #7 on: July 07, 2009, 09:53:54 PM »

Is this target "printable" with a good printer??  Or must an original be ordered to get the resolution needed??

Thx,  John  PhF
Logged

John  Phf
Stephen Brennan
Newbie
*
Posts: 9



View Profile WWW
« Reply #8 on: July 08, 2009, 05:04:03 AM »

Unfortunately they had to be bought.
Logged
Ernst Dinkla
Sr. Member
****
Posts: 394


View Profile Email
« Reply #9 on: July 08, 2009, 09:37:39 AM »

Is there a target and/or procedure for quantitatively determining the quality/capabilities of a scanner??  I have a scanner that claims it will scan up to 4800 ppi (optical, not software).  I'd like to determine if 4800 is actually better than 1200 or 2400 ppi, maybe "better" isn't the correct term....I'd like to test 1200, 2400 and 4800 and see the actual results...is the 2400 or 4800 fuzzy or sharp??

Any ideas??  Thx

John  Phf


John,

Imatest Studio from the Norman Koren stable. It will do the same tests for your camera and lenses. That will make its price easier to digest. If you can do the job within 10 tests you could use the demo.

http://www.imatest.com

Depending on the manufacturer and the scanner model the "resolution" specs can be the upsampling possible with the driver or the physical sampling rate (what I like to call SPI). The first category are cheap consumer flatbeds, the second category better flatbeds and desktop filmscanners. None give a true optical resolution but Nikon usually is close with its numbers. Epson etc for the film scanner flatbeds gives a 200-300 % exaggeration of the optical resolution in their sampling rate numbers. So there are actually three categories of "resolution".

Remember that with some scanner drivers if you select 1200, 2400, the scanner actually will reduce the sampling rate which most of the time gives a lower true optical resolution but also a lower dynamic range. Resolution testing then isn't telling all of the image quality. Though the MTF chart like Imatest delivers is telling more. The highest sampling rate possible on the scanner (that is what the scanner manufacturer usually quotes as resolution but isn't resolution) will give the highest score too, and oversampling like more flatbeds do has some advantages. To avoid the reduced sampling rate of lower driver resolution settings you can select an odd resolution number just above the 1200, 2400. The driver will then use the scanner's highest sampling rate and downsample to the odd number in the driver. But that again relies on the quality of the downsampling routine. So I always scan at the highest sampling rate and use for example Qimage's downsampling to get a workable file size. And the Nikon 8000 doesn't reduce the sampling rate when lower resolution settings are selected, it will always downsample from the 4000 SPI. It gives a true 3800 PPI in practice. Most of the flatbeds do not get near the 50% of the highest sampling rate for their true optical resolution.

Edit: there is a cheaper alternative for Imatest: QuickMTF http://www.quickmtf.com/  70$ Shareware?
To make a slanted edge target for film scanner tests you can use a cut razor blade in a slide frame.
I made a slanted edge film target and had some processed by an image setter shop on lith film so the target could be used with the wet mounting I do. A razor blade wet mounted would show optical distortions in the fluid at the edge.


met vriendelijke groeten, Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/
« Last Edit: July 09, 2009, 08:47:34 AM by Ernst Dinkla » Logged
Seth
Sr. Member
****
Posts: 322



View Profile
« Reply #10 on: August 04, 2009, 03:05:38 AM »

If you are not using Vuescan, it could help you.  Sometimes with lower resolution scanners, simply doing two or three passes improves quality.  Vuescan also supports scanners that use infrared scans.
Logged

Seth
<CWO4 (FMF) USN, Ret.>
Pages: [1]
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.21 | SMF © 2015, Simple Machines Valid XHTML 1.0! Valid CSS!