[fprint] New driver for Validity VFS101: tester wanted

[Sergio Cerlesi]

Hi Kunal,

i have done some test and reproduce the first issues (image height = 1).
Removing the break isn't correct because without it empty lines at end
of finger image are not removed. I rewrite a logic of first block of
img_screen and now is more robust.

I can't reproduce the second issue, on my system the driver work
correctly. However flushing the buffer isn't the solution because, when
send a command to the sensor, the buffer are already reset (see m_swap
function). If you send me the dumped logs i will try to investigate.

Attached version two of the patch that i hope fix your issue.

Bye
Sergio

Il giorno lun, 28/02/2011 alle 11.07 +0100, Sergio Cerlesi ha scritto:
> Hi Kunal,
> 
> thank for feedback and patch.
> 
> I will investigate on the two issues, could you send me the log ?
> 
> The issues you report are occasional or systematic ?
> 
> Bye
> Sergio
> 
> Il giorno dom, 27/02/2011 alle 16.20 +0530, Kunal Gangakhedkar ha
> scritto:
> > Hi Sergio,
> > 
> > Awesome work !!
> > Got the patch working on my HP dv6 notebook having the VFS101 sensor.
> > 
> > Two issues:
> > o. 	on my sensor, the img_screen() function was always returning
> > 	image height = 1. I added a fp_dbg() call to dump the values 
> > 	in the loop variables to figure out what was going on.
> > 	The 'break' statement in the 'find bottom of image' part
> >         is not correct since it aborts the screening prematurely.
> >         Due to this break, the screened image height always ended up = 1
> >         in my case.
> > 
> >        	This happens even during the enrollment phase.
> > 	I have the logs dumped - let me know if you want to take a look at them.
> > 
> > o.	during verification, if I end up just touching the sensor and lifting the 
> > 	finger, I get a message 'Unexpected finger find, remove finger from sensor'.
> > 	But, it goes into an infinite loop with the same msg since the device's 
> > 	bufffer still contains the old image data. The device buffer needs to be
> > 	flushed upon any such error conditions.
> > 
> > I'm attaching a patch on top of your code that tries to fix both the problems.
> > Please let me know if it looks OK.
> > 
> > Thanks,
> > Kunal
> > 
> > On Friday 25 Feb 2011 2:19:23 pm Sergio Cerlesi wrote:
> > > Hi all,
> > > 
> > > i finish to write the driver fo Validity VFS101. I use it for a couple
> > > of day with good result.
> > > 
> > > For verify the implementation i need tester. If i receive positive
> > > feedback i will work to merge it into library.
> > > 
> > > Attached the patch for libfprint 0.3.0. To compile it run:
> > > 
> > > autoreconf && ./configure && make
> > > 
> > > Important: the match score is much variable (on my experience from 10 to
> > > 100) so, to have a good chance to match the fingerprint, it's much
> > > important who scan is done. I use fprint_demo to find the best way to
> > > scan my finger.
> > > 
> > > Waiting feedback.
> > > 
> > > Bye
> > > Sergio
> > > 
> > 
> 
> 

-------------- next part --------------
A non-text attachment was scrubbed...
Name: libfprint-0.3.0-vfs101.v2.patch
Type: text/x-patch
Size: 35202 bytes
Desc: not available
Url : http://lists.reactivated.net/pipermail/fprint/attachments/20110301/d2faebf8/libfprint-0.3.0-vfs101.v2-0001.bin