Method and apparatus with selective combined authentication
A method and apparatus with selective combined authentication performs a single authentication based on a first modality among plural modalities, and in response to the single authentication having failed, determines whether to perform a combined authentication by a combination of two or more of the plural modalities, and selectively, depending on a result of the determining of whether to perform the combined authentication, performs the combined authentication.
1 . A processor-implemented authentication method, the method comprising:
performing a single authentication of a first feature of a user by a first modality, among plural modalities, to determine whether the first feature satisfies a first condition determinative of success or failure of the single authentication, the first modality corresponding to a first biometric information of the user, the first feature being extracted from data sensed by first modality sensor at a single timing;
when the single authentication of the user has failed, determining whether to perform a combined authentication of the user, based on whether the first feature satisfies an entry condition prior to performing the combined authentication of the user for each combination of the plural modalities; and
selectively, depending on a result of the determining whether to perform the combined authentication of the user:
extracting a second feature of the user by a second modality corresponding to a combination of two or more of the plural modalities, the second modality corresponding to a second biometric information of the user different from the first biometric information, the second feature being extracted from data sensed by second modality sensor at a single timing;
generating a third feature of the user by fusing the first feature extracted by the first modality and the second feature of the user extracted by the second modality; and
performing the combined authentication of the user using the third feature to determine whether the third feature satisfies a third condition determinative of success or failure of the combined authentication,
wherein the entry condition and the first condition are determined based on a first feature space of the first feature,
wherein a threshold of the entry condition is less strict than a threshold of the first condition, and
wherein the third condition is determined corresponding to a third feature space of the third feature, the third feature corresponds to a fusion of the first feature and the second feature.
2 . The method of claim 1 , wherein the first modality is an iris image modality.
3 . The method of claim 1 , wherein the second modality is a combination of two or more of a face, a fingerprint, an iris, a vein, a palmprint, a signature, a voice, a gait, and a DNA structure modality.
4 . The method of claim 1 , wherein the combined authentication is applied to a payment application.
5 . A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform the control method of claim 1 .
6 . A processor-implemented authentication method, the method comprising:
performing a first single authentication of a first feature of a user by a first modality, among plural modalities, to determine whether the first feature satisfies a first condition determinative of success or failure of the first single authentication, the first modality corresponding to a first biometric information of the user, the first feature being extracted from data sensed by first modality sensor at a single timing;
performing a second single authentication of a second feature of the user by a second modality, among the plural modalities, based on whether the second feature satisfies a second condition determinative of success or failure of the second single authentication, the second modality corresponding to a second biometric information of the user different from the first biometric information, the second feature being extracted from data sensed by second modality sensor at a single timing;
determining whether to perform a combined authentication of the user, if both of the first single authentication and the second single authentication, which have been performed, are determined to be a failure, based on whether the first feature satisfies an entry condition, prior to performing the combined authentication of the user for each combination of the plural modalities;
selectively, depending on a result of the determining whether to perform the combined authentication of the user,
generating a third feature of the user by a third modality corresponding to a combination of two or more of the plural modalities by fusing the first feature extracted by the first modality and the second feature of the user extracted by the second modality; and
performing the combined authentication of the user based on the third feature used to determine whether the third feature satisfies a third condition determinative of success or failure of the combined authentication,
wherein the entry condition and the first condition is determined based on a first feature space of the first feature,
wherein a threshold of the entry condition is less strict than a threshold of the first condition, and
wherein the third condition is determined corresponding to a third feature space of the third feature, the third feature corresponds to a fusion of the first feature and the second feature.
7 . The method of claim 6 , wherein the first modality is a face image modality and
wherein the second modality is an iris image modality.
8 . The method of claim 6 , wherein the third modality is a combination of two or more of a face image modality, a fingerprint image modality, an iris image modality, a vein image modality, a palmprint image modality, a signature modality, a voice modality, a gait modality, and a DNA structure modality.
9 . The method of claim 6 , wherein the combined authentication is applied to an unlock function of a smart phone.
10 . A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform the control method of claim 6 .
11 . A processor-implemented authentication method, the method comprising:
performing a single authentication of a first feature of a user by a first modality, among plural modalities, to determine whether the first feature satisfies a first condition determinative of success or failure of the single authentication, the first modality corresponding to a first biometric information of the user, the first feature being extracted from data sensed by first modality sensor at a single timing;
when the single authentication of the user has failed, determining whether to perform a combined authentication of the user, based on whether an eye region is detected in a first image of the first modality, which is determined based on whether the first feature satisfies an entry condition, prior to performing the combined authentication of the user for each combination of the plural modalities; and
selectively, depending on a result of the determining whether to perform the combined authentication of the user,
extracting a second feature of the user by a second modality corresponding to a combination of two or more of the plural modalities, the second modality corresponding to a second biometric information of the user different from the first biometric information, the second feature being extracted from data sensed by second modality sensor at a single timing;
generating a third feature of the user by fusing the first feature extracted by the first modality and the second feature of the user extracted by the second modality; and
performing the combined authentication of the user based on the third feature to determine whether the third feature satisfies a third condition determinative of success or failure of the combined authentication,
wherein the entry condition and the first condition is determined based on a first feature space of the first feature,
wherein a threshold of the entry condition is less strict than a threshold of the first condition, and
wherein the third condition is determined corresponding to a third feature space of the third feature, the third feature corresponds to a fusion of the first feature and the second feature.
12 . The method of claim 11 , wherein the first modality is a face image modality.
13 . The method of claim 11 , wherein the second modality is a combination of two or more of a face image modality, a fingerprint image modality, an iris image modality, a vein image modality, a palmprint image modality, a signature modality, a voice modality, a gait modality, and a DNA structure modality.
14 . The method of claim 11 , wherein the first image is a face image.
15 . The method of claim 11 , wherein the first image comprises a color image, an IR image, or a depth image.
16 . The method of claim 11 , wherein the second image comprises a color image, an IR image, or a depth image.
17 . A non-transitory computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform the control method of claim 11 .