Electronic device for encrypting biometric information and method for operating the same
An electronic device encrypting biometric information includes memory for storing instructions and a processor. The instructions may be configured to, when executed by the processor, cause the electronic device to obtain a transformation matrix that maps a first feature value related to first biometric information to a first codeword, generate a third feature value by applying the transformation matrix to a second feature value related to second biometric information, compare a cosine similarity between a second codeword most similar to the third feature value and the third feature value with a threshold, map the second feature value to the second codeword based on the cosine similarity being greater than the threshold, and generate a hashed second codeword based on the second codeword, and transmit or store the hashed second codeword.
1. An electronic device encrypting biometric information, the electronic device comprising:
memory for storing instructions; and
a processor operatively connected with the memory,
wherein the instructions are configured to, when executed by the processor, cause the electronic device to:
obtain a transformation matrix that maps a first feature value related to first biometric information to a first codeword,
generate a third feature value by applying the transformation matrix to a second feature value related to second biometric information,
compare a cosine similarity between a second codeword most similar to the third feature value among designated codewords and the third feature value with a threshold,
map the second feature value to the second codeword based on the cosine similarity being greater than the threshold,
generate a hashed second codeword based on the second codeword, and
transmitting or storing the hashed second codeword.
2. The electronic device of claim 1 , wherein the instructions are configured to, when executed by the processor, cause the electronic device to:
obtain a reference cosine similarity related to biometrics recognition of the second feature value, and
determine the threshold based on the reference cosine similarity.
3. The electronic device of claim 1 , wherein the instructions are configured to, when executed by the processor, cause the electronic device to:
based on the cosine similarity being not greater than the threshold, determine encryption of the second biometric information as failed.
4. The electronic device of claim 1 , wherein at least one of the first codeword or the second codeword is selected from an error correction code (ECC) including designated real-valued codewords.
5. The electronic device of claim 4 , wherein the instructions are configured to, when executed by the processor, cause the electronic device to:
select the second codeword most similar in angular distance to the third feature value from among the codewords in the ECC.
6. The electronic device of claim 1 , wherein at least one of the first biometric information or the second biometric information includes a facial image.
7. The electronic device of claim 1 , wherein at least one of the first feature value or the second feature value is generated based on a facial feature extraction algorithm.
8. The electronic device of claim 1 , wherein the instructions are configured to, when executed by the processor, cause the electronic device to:
extract the first feature value from the first biometric information;
generate the transformation matrix that maps the first feature value to the first codeword; and
store the transformation matrix in the memory.
9. The electronic device of claim 8 , wherein the instructions are configured to, when executed by the processor, cause the electronic device to transmit information of the transformation matrix to a server.
10. The electronic device of claim 1 , wherein the transformation matrix includes a rotation matrix.
11. A method for operating an electronic device encrypting biometric information, the method comprising:
obtaining a transformation matrix that maps a first feature value related to first biometric information to a first codeword;
generating a third feature value by applying the transformation matrix to a second feature value related to second biometric information;
comparing a cosine similarity between a second codeword most similar to the third feature value and the third feature value with a threshold;
mapping the second feature value to the second codeword based on the cosine similarity being greater than the threshold; and
generating a hashed second codeword based on the second codeword; and
transmitting or storing the hashed second codeword.
12. The method of claim 11 , further comprising:
obtaining a reference cosine similarity related to biometrics recognition of the second feature value;
determining the threshold based on the reference cosine similarity.
13. The method of claim 11 , further comprising, based on the cosine similarity being not greater than the threshold, determining encryption of the second biometric information as failed.
14. The method of claim 11 , wherein at least one of the first codeword or the second codeword is selected from an error correction code (ECC) including designated real-valued codewords.
15. The method of claim 14 , further comprising selecting the second codeword most similar in angular distance to the third feature value from among the codewords in the ECC.
16. The method of claim 11 , wherein at least one of the first biometric information or the second biometric information includes a facial image.
17. The method of claim 11 , wherein at least one of the first feature value or the second feature value is generated based on a facial feature extraction algorithm.
18. The method of claim 11 , further comprising:
extracting the first feature value from the first biometric information;
generating the transformation matrix that maps the first feature value to the first codeword; and
storing the transformation matrix.
19. The method of claim 18 , further comprising:
transmitting information of the transformation matrix to a server;
wherein the transformation matrix includes a rotation matrix.
20. One or more non-transitory computer-readable storage medium storing one or more programs, the one or more programs comprising instructions configured to, when executed by a processor of an electronic device, cause the electronic device to:
obtain a transformation matrix that maps a first feature value related to first biometric information to a first codeword;
generate a third feature value by applying the transformation matrix to a second feature value related to second biometric information;
compare a cosine similarity between a second codeword most similar to the third feature value and the third feature value with a threshold;
map the second feature value to the second codeword based on the cosine similarity being greater than the threshold;
generate a hashed second codeword based on the second codeword; and
transmit or store the hashed second codeword.