Methods, apparatus and systems for authentication

A method of enrolling a user in a biometric authentication system, the method comprising:

• obtaining ear biometric data for a user to be enrolled in the biometric authentication system, the ear biometric data comprising one or more features characteristic of the user's ear canal and an associated fit metric, the fit metric being indicative of a positioning of a personal audio device relative to the user's ear canal, the personal audio device comprising a transducer for application of acoustic stimulus to the user's ear to obtain the ear biometric data;
• and
• determining a plurality of target user model sets based on the ear biometric data, each target user model set being associated with a particular fit metric of a set of fit metrics and each target user model set comprising one or more target user models based on respective one or more features for the particular fit metric, wherein at least one of the target user model sets is associated with the determined associated fit metric and the one or more target user models of the at least one of the target user model set are based on the respective determined one or more features;
• wherein a user is identified as a particular authorised user based on user biometric data comprising one or more features and an associated fit metric determined from the user, and the target user model set associated with the associated fit metric determined from the user, wherein identifying the user comprises:
• determining one or more target user similarity scores indicative of the similarity of the one or more features to a corresponding target user model of the determined target user model set;
• determining a non-target user model set from a plurality of non-target user model sets based on the fit metric, each non-target user model set being associated with a particular fit metric of a set of fit metrics and comprising a plurality of non-target user models;
• determining a plurality of non-target user similarity scores, each indicative of the similarity of the one or more features to a respective non-target user model of the plurality of the non-target user model of the determined target user model set;
• and
• adjusting the one or more target user similarity scores based on the plurality of non-target user similarity scores.

A biometric authentication apparatus, comprising:

• a biometric module for performing a biometric process;
• a stimulus module for generation of an acoustic stimulus;
• a transducer for application of the acoustic stimulus to a user's ear, a microphone for detecting a response signal of the user's ear to the acoustic stimulus;
• a feature extraction module for extracting from the response signal one or more features characteristic of the user's ear canal for use in a biometric process;
• and
• a fit measure module for determining a fit metric associated with the one or more features;
• wherein the biometric module is configured to identify the user as a particular authorised user using both the determined one or more features and the determined associated fit metric by:
• determining a target user model set from a plurality of target user model sets based on the fit metric, each target user model set being associated with a particular fit metric of a set of fit metrics and comprising at least one target user model;
• determining one or more target user similarity scores indicative of the similarity of the one or more features to a corresponding target user model of the determined target user model set;
• determining a non-target user model set from a plurality of non-target user model sets based on the fit metric, each non-target user model set being associated with a particular fit metric of a set of fit metrics and comprising a plurality of non-target user models;
• determining a plurality of non-target user similarity scores, each indicative of the similarity of the one or more features to a respective non-target user model of the plurality of the non-target user model of the determined target user model set;
• and
• adjusting the one or more target user similarity scores based on the plurality of non-target user similarity scores.