Authentication system and registration system related to facial feature information

An electronic registration system for registering information for face authentication, comprising:

• two cameras configured to take two face images of a first person at the same time;
• a controller, said controller comprising:
• a generator configured to generate face information including three-dimensional shape information in the face of the first person to be authenticated on the basis of measurement information of the first person, said measurement information being obtained by using the principal of triangulation from corresponding feature points between two face images of the first person;
• a first calculator configured to calculate a correlation value based on the inverse of an average differential absolute value of luminance values between at least two of the corresponding feature points in the two face images;
• a model modifier configured to modify a standard model of a human face by using the face information, thereby generating an individual model of the face of the first person;
• a second calculator configured to calculate a first model perfection level as a perfection level of the individual model on the basis of reliability of the face information, the perfection level being calculated using the correlation value;
• an extractor configured to extract feature information of the first person from the individual model;
• a registry configured to register the model perfection level and the feature information as the information for face authentication;
• and
• a camera controller configured to take the two face images by the two cameras repeatedly until the correlation value necessary for a required authentication accuracy is obtained.

A computer software program stored on a non-transitory computer readable medium for causing a computer to execute a method comprising:

• a generating step of generating face information including three-dimensional shape information in the face of the first person to be authenticated on the basis of measurement information of the first person, said measurement information being obtained by using the principal of triangulation from corresponding feature points between two face images of the first person;
• a first calculating step for calculating a correlation value based on the inverse of an average differential absolute value of luminance values between at least two of the corresponding feature points in the two face images;
• a model modifying step of modifying a standard model of a human face by using the face information, thereby generating an individual model of the face of the first person;
• a second calculating step of calculating a first model perfection level as a perfection level of the individual model on the basis of reliability of the face information, the perfection level being calculated using the correlation value;
• an extracting step of extracting first feature information as feature information of the first person from the individual model;
• an obtaining step of obtaining second feature information stored in said computer as preregistered
• feature information of a second person to be compared with the first feature information;
• an authenticating step of performing an authenticating operation on the first person by using the first model perfection level in addition to similarity between the first feature information and the second feature information;
• and
• a controlling step of causing the two face images to be repeatedly acquired until the correlation value necessary for a required authentication accuracy is obtained.

A controller for authenticating facial feature information, said controller comprising:

• a memory;
• and
• a computer processor programmed to perform the steps of:
• generating face information including three-dimensional shape information in the face of a first person to be authenticated on the basis of measurement information of the first person, said measurement information being triangulation data from corresponding feature points between two face images of the first person taken simultaneously by two cameras;
• calculating a correlation value based on the inverse of an average differential absolute value of luminance values between at least two of the corresponding feature points in the two face images;
• modifying a standard model of a human face by using the face information, thereby generating an individual model of the face of the first person;
• calculating a first model perfection level as a perfection level of the individual model on the basis of reliability of the face information, the perfection level being obtained from the correlation value;
• extracting first feature information as feature information of the first person from the individual model;
• obtaining second feature information stored in said controller as pre-registered feature information of a second person to be compared with the first feature information;
• performing an authenticating operation on the first person by using the first model perfection level in addition to similarity between the first feature information and the second feature information;
• and
• causing the two face images to be repeatedly acquired until the correlation value necessary for a required authentication accuracy is obtained.

A controller for registering facial feature information, comprising:

• a memory;
• and
• a computer processor programmed to perform the steps of:
• generating face information including three-dimensional shape information in the face of a first person to be authenticated on the basis of measurement information of the first person, said measurement information being triangulation data from corresponding feature points between two face images of the first person taken simultaneously by two cameras;
• calculating a correlation value based on the inverse of an average differential absolute value of luminance values between at least two of the corresponding feature points in the two face images;
• modifying a standard model of a human face by using the face information,
• thereby generating an individual model of the face of the first person;
• calculating a model perfection level as a perfection level of the individual model on the basis of reliability of the face information, the perfection level being obtained from the correlation value;
• extracting feature information of the first person from the individual model;
• causing the two face images to be repeatedly acquired until the correlation value necessary for a required authentication accuracy is obtained;
• and
• registering the model perfection level and the feature information as the information for face authentication.

An electronic authentication system comprising:

• two cameras configured to take two face images of a first person each other at the same time;
• and
• a controller, said controller including:
• a generator configured to generate face information including three-dimensional shape information in the face of the first person to be authenticated on the basis of measurement information of the first person, said measurement information being obtained by using the principal of triangulation from corresponding feature points between the two face images of the first person;
• a first calculator configured to calculate a correlation value based on the inverse of an average differential absolute value of luminance values between at least two of the corresponding feature points in the two face images;
• a model modifier configured to modify a standard model of a human face by using the face information, thereby generating an individual model of the face of the first person;
• an extractor configured to extract first feature information as feature information of the first person from the individual model;
• an obtainer configured to obtain second feature information stored in said controller as pre-registered feature information of a second person to be compared with the first feature information;
• an authenticator configured to perform an authenticating operation on the first person by using the first model perfection level in addition to similarity between the first feature information and the second feature information;
• and
• a camera controller configured to take the two face images by the two cameras repeatedly until the correlation value necessary for a required authentication accuracy is obtained.