Re: What is "detection / false positive rates" in fingerprint recognition?
From: Bernie (a.buisson_at_nextamp.com)
Date: 08/26/04
- Next message: Martin Brown: "Re: Using fftw to get a power spectrum and working with phase images"
- Previous message: GrahamH: "Re: VGA input frame grabber"
- In reply to: XQWU: "What is "detection / false positive rates" in fingerprint recognition?"
- Messages sorted by: [ date ] [ thread ]
Date: Thu, 26 Aug 2004 11:43:55 +0200
I think that some expert could respond to your question an a more specific
list
try to post-it on the watermarking list : http://www.watermarkingworld.org/
i work in watermarking but i have not enough theorical background to give a
math expression of false-positive.
good luck
"XQWU" <csxqwu@yahoo.com> a écrit dans le message de
news:250ad5da.0408231914.3336ade3@posting.google.com...
> Hi all,
> I have plotted the matching score distribution curves and ROC
> curves for fingerprint recognition. But a reviewer suggested that "use
> standard metrics to characterize classification performance; e.g.
> confusion matrix or detection / false positive rates in combination
> with statistical testing. At the moment one is reduced to eyeballing
> graphs."
>
> What's the "confusion matrix" and "detection / false positive rates"
> here? Are the FAR/FRR curves?
>
> Thanks a lot!
- Next message: Martin Brown: "Re: Using fftw to get a power spectrum and working with phase images"
- Previous message: GrahamH: "Re: VGA input frame grabber"
- In reply to: XQWU: "What is "detection / false positive rates" in fingerprint recognition?"
- Messages sorted by: [ date ] [ thread ]