Face recognition system and method capable of updating registered face templates

A face recognition system comprising:

• memory configured to store a registered template database including a registered face template and configured to store an authenticated template database including a plurality of authenticated face templates, wherein the registered face template includes a face template generated from a face image of a registered user and used as a reference for authenticating the face template of a user attempting a face authentication, and wherein each of the plurality of authenticated face templates includes a face template generated from a face image of the user attempting the face authentication, the user attempting the face authentication is determined as the registered user in the face authentication, the face template of the user attempting the face authentication used during the face authentication is stored in the authenticated template database as one of the authenticated face templates;
• a camera configured to capture an image including a face of the user attempting the face authentication;
• a processor comprising circuitry, the processor configured to:
• detect a face image from the image captured by the camera;
• extract feature point information from the detected face image;
• generate an aligned face template by normalizing the face image using the extracted feature point information;
• calculate a similarity degree between the aligned face template inputted and the registered face template stored in the registered template database;
• determine the user attempting the face authentication as the registered user when the similarity degree is higher than a reference value;
• obtain similarity degrees respectively for every combination between each of a plurality of the authenticated face templates by using the plurality of the authenticated face templates stored in the authenticated template database after elapse of a predetermined period during which the face authentication has been performed multiple times;
• obtain the value of an average similarity degree for each of the plurality of authenticated face templates by averaging the values of the similarity degrees associated with itself;
• select as a new registered face template one of the authenticated face templates having the value of the largest average similarity degree among the values of the average similarity degree;
• and
• update the registered face template by storing the new registered face template in the registered template database.

A face recognition method for updating a registered face template performed by a face recognition system including memory configured to store a registered template database including a registered face template and configured to store an authenticated template database including a plurality of authenticated face templates, wherein the registered face template includes a face template generated from a face image of a registered user and used as a reference for authenticating the face template of a user attempting a face authentication, and wherein each of the plurality of authenticated face templates includes a face template generated from a face image of the user attempting the face authentication, the user attempting the face authentication is determined as the registered user in the face authentication, the face template of the user attempting the face authentication used during the face authentication is stored in the authenticated template database as one of the authenticated face templates, and a camera configured to capture an image including a face of the user attempting the face authentication, the method comprising:

• detecting a face image from the image captured by the camera;
• extracting feature point information from the detected face image;
• generating an aligned face template by normalizing the face image using the extracted feature point information;
• calculating a similarity degree between the aligned face template inputted and the registered face template stored in the registered template database;
• determining the user attempting the face authentication as the registered user when the similarity degree is higher than a reference value;
• obtaining similarity degrees respectively for every combination between each of a plurality of the authenticated face templates by using the plurality of the authenticated face templates stored in the authenticated template database after elapse of a predetermined period during which the face authentication has been performed multiple times;
• obtaining the value of an average similarity degree for each of the plurality of authenticated face templates by averaging the values of the similarity degrees associated with itself;
• selecting as a new registered face template one of the authenticated face templates having the value of the largest average similarity degree among the values of the average similarity degree;
• and
• updating the registered face template by storing the new registered face template in the registered template database.