iTC Meeting Minutes 2023 10 17 - biometricITC/cPP-biometrics GitHub Wiki
Call started at 10:02am EDT
-
Brian will create a TD for the proposed change to the fingerprint PAD test
-
Brian will create a private repo for discussion of the ISO documents (for commenting)
The call started with a quick review of the task list. There were no updates at this time.
The group then had a quick discussion about the ISO drafts. Brian added Clare and Greg O to the folder where they are stored in OnlyOffice. Naruki will send Greg F a link to the ISO documents to see if he can directly access them from there, and if not he will create an account for OnlyOffice to access them there. Naruki is going to try and set up a call with that group next week (not sure when the call will be) to discuss collaboration. The comments are due back at the end of November, so the next call (after the ICCC) will be focused on this.
The call then moved to the PAD testing question.
Brian summarized the current discussion that the mold tests are listed as non-cooperative, and hence within the proper AP, but the test steps are actually cooperative, which would put them outside the AP. The original reason was for consistency and to provide a worse-case scenario, which ended up being too difficult. Stephanie from FIDO also responded via email that cooperative tests are specifically out of scope for FIDO.
Greg O asked why this was a problem when the other two tests used cooperative tests to capture a flat image via the sensor. Brian stated that it seems that the problem is in large part about the depth of the ridges, which isn’t captured in the flat image but is in the molds. Greg O also pointed out that given that it seems those tests do seem to pass, it appears they are good ones, even using cooperative collection, and so is probably good.
Brian then said the TD should probably change only the two attacks and the fingerprint overview. The attack docs should have the cooperative listing which would then change the AP calculation, and the overview should note that the 3D molds are now considered out of scope, but not to remove any of these for the time being as they are likely useful for the PAD levels project. Everyone agreed with this proposal, so Brian said he would put it together before the ICCC meetings so it could be sent to NIAP (and he could try to talk to people face-to-face about it while there).
Brian also stated that we should probably look at the Attack Potential tables we are using as they don’t match the normal CC one. Naruki thought this may have been taken from ISO work and would check. Brian said we should review this as we work on the new PAD since it would have an impact on how we rate everything, and so a good time to look at it.
The call ended at 10:56am EDT.