I tried Anthony's aes1610 driver code which he attached in an email
and also with libfprint/fprint_demo v0.0.4 but I get the attached
images with img_capture. I doubt the minutae count is anywhere near
what's needed for a successful enroll :-)

I was getting "no image height assigned" and the -5 error code so I
changed the -1 to 8 in aes1610.c's .img_height field which suppresses
this error, but:

Whether I place my finger or thumb on the reader slowly and swipe it
slow and steady, or if I flick it across quickly I get similar finger
images (in either swipe direction) which look nothing like a
fingerprint scan... I do get the "reversed scan direction" log on one
direction. I have the reader on an HP Pavilion tx1302au tablet PC.

Anything I should be doing differently, or do you need more information?

Actually, I've attached finger scan images which look a bit better
after some trial and error, but not good enough...

Duh, while writing this email and experimenting with fprint_demo, I
got a "successful" scan with 11 minutae for my right index finger.
Trying my left thumb gives a segfault after logging "reverse
direction"... but most of the time it just says bad scan...

I'll try some more swipes to get more successful scans...

-- 
"Thou shalt not follow the null pointer for at it's end madness and chaos lie."

<<attachment: finger.pgm>>

<<attachment: finger_standardized.pgm>>

Attachment: unpatched-fast.out
Description: Binary data

Attachment: patched.out
Description: Binary data

_______________________________________________
fprint mailing list
[email protected]
http://lists.reactivated.net/mailman/listinfo/fprint

Reply via email to