Collation system, client terminal, server apparatus, collation method, and program
A collation system 10 includes divide registered information to first information and second information, and to provide the second information to a server apparatus 200 , perform a first step for similarity calculation between the registered information and collation information inputted for collation with the registered information, based on the collation information and the first information, send a calculation result of the first step to the server apparatus 200 , perform a second step for the similarity calculation based on the calculation result of the first step received from a client terminal 100 , and the second information, send a calculation result of the second step to the client terminal 100 , and perform a third step for the similarity calculation based on the calculation result of the second step received from the server apparatus 200 , and the first information to calculate similarity between the registered information and the collation information.
1. A collation system comprising:
a client terminal including a memory storing instructions, and one or more processors; and
a server apparatus including a memory storing instructions, and one or more processors, wherein
the client terminal is configured to divide registered information to first information and second information, and to provide the second information to the server apparatus,
the client terminal is configured to perform a first step for similarity calculation between the registered information and collation information inputted for collation with the registered information, based on the collation information and the first information,
the client terminal is configured to send a calculation result of the first step to the server apparatus,
the server apparatus is configured to perform a second step for the similarity calculation based on the calculation result of the first step received from the client terminal, and the second information,
the server apparatus is configured to send a calculation result of the second step to the client terminal, and
the client terminal is configured to perform a third step for the similarity calculation based on the calculation result of the second step received from the server apparatus, and the first information to calculate similarity between the registered information and the collation information.
2. The collation system according to claim 1 , wherein
the registered information is x, the first information is x 1 , x 2 , and the second information is x 3 , the registered information is divided to satisfy x=x 1 +x 2 ·x 3 .
3. The collation system according to claim 2 , wherein
the similarity calculation is an inner product calculation of the registered information and the collation information, represented by x·y=(x 1 +x 2 ·x 3 )·y when the collation information is y,
in the first step, x 2 ·y is calculated,
in the second step, the calculation result of the first step is multiplied by x 3 , and
in the third step, x 1 ·y is added to the calculation results of the second step.
4. The collation system according to claim 3 , wherein
the collation information is divided to satisfy y=y 1 +y 2 , and
the first step, the second step and the third step are performed for each of y 1 and y 2 .
5. The collation system according to claim 1 , wherein
the second step and the third step are performed in secure computation.
6. The collation system according to claim 1 , wherein
the registered information and the collation information are common n-dimensional vectors.
7. The collation system according to claim 1 , wherein
the client terminal and the server apparatus are configured to perform the similarity calculation based on a challenge-response method.
8. The collation system according to claim 1 , wherein
the registered information and the collation information represent a feature of biometric information.
9. A client terminal comprising:
a memory storing instructions; and
one or more processors configured to execute the instructions to:
divide registered information to first information and second information, to store the first information, and to provide the second information to a server apparatus;
perform a first step for similarity calculation between the registered information and collation information inputted for collation with the registered information, based on the collation information and the first information;
send a calculation result of the first step to the server apparatus; and
perform a third step for the similarity calculation based on a calculation result of a second step for the similarity calculation based on a calculation result of the first step received from the server apparatus and the second information, and the first information to calculate similarity between the registered information and the collation information.
10. A server apparatus comprising:
a memory storing instructions; and
one or more processors configured to execute the instructions to:
receive, from a client terminal, a calculation result of a first step for similarity calculation between first information divided from registered information and collation information inputted for collation with the registered information, and to perform a second step for the similarity calculation based on the calculation result of the first step and second information divided from the registered information; and
send, to the client terminal, a calculation result of the second step for the client terminal to perform a third step for the similarity calculation based on the calculation result of the second step and the first information.
11. A collation method in a collation system including a client terminal and a server apparatus, the collation method comprising:
dividing, by the client terminal, registered information to first information and second information;
providing, by the client terminal, the second information to the server apparatus;
performing, by the client terminal, a first step for similarity calculation between the registered information and collation information inputted for collation with the registered information, based on the collation information and the first information;
sending, by the client terminal, a calculation result of the first step to the server apparatus;
performing, by the server apparatus, a second step for the similarity calculation based on the calculation result of the first step received from the client terminal, and the second information;
sending, by the server apparatus, a calculation result of the second step to the client terminal; and
performing, by the client terminal, a third step for the similarity calculation based on the calculation result of the second step received from the server apparatus, and the first information to calculate similarity between the registered information and the collation information.