US 11,699,300 B2
Methods and apparatuses for updating user authentication data
Kyuhong Kim, Seoul (KR); Sungjoo Suh, Seoul (KR); Hyungsuk Kim, Seongnam-si (KR); Changkyu Choi, Seongnam-si (KR); Namjoon Kim, Anyang-si (KR); and Wonjun Kim, Hwaseong-si (KR)
Assigned to SAMSUNG ELECTRONICS CO., LTD., Gyeonggi-Do (KR)
Filed by Samsung Electronics Co., Ltd., Suwon-si (KR)
Filed on Apr. 6, 2020, as Appl. No. 16/840,782.
Application 16/840,782 is a division of application No. 15/078,148, filed on Mar. 23, 2016, granted, now 10,678,896.
Claims priority of application No. 10-2015-0093461 (KR), filed on Jun. 30, 2015; and application No. 10-2015-0152963 (KR), filed on Nov. 2, 2015.
Prior Publication US 2020/0233944 A1, Jul. 23, 2020
Int. Cl. G06V 40/12 (2022.01); G06F 16/23 (2019.01); G06F 16/51 (2019.01); G06F 21/32 (2013.01); G06V 10/98 (2022.01); G06V 10/26 (2022.01); G06V 10/50 (2022.01); G06V 10/75 (2022.01); G06T 1/00 (2006.01)
CPC G06V 40/1335 (2022.01) [G06F 16/23 (2019.01); G06F 16/51 (2019.01); G06F 21/32 (2013.01); G06T 1/0007 (2013.01); G06V 10/26 (2022.01); G06V 10/50 (2022.01); G06V 10/7515 (2022.01); G06V 10/98 (2022.01)] 17 Claims
OG exemplary drawing
 
1. A method for updating biometric authentication data in a database stored in a memory of a device, the method comprising:
authenticating an input image including biometric information of a biometric feature of a user based on enrolled images in the database stored in the memory;
identifying a portion of the input image that is not overlapped by any of the enrolled images;
determining a first similarity based on sizes of overlap regions between the input image and enrolled images in the database, the first similarity having a greatest value of similarity of the sizes of the overlap regions between the input image and the enrolled images in the database, the determining the first similarity including,
determining correlations in the overlap regions between the input image and the enrolled images,
determining the sizes of the overlap regions between the input image and the enrolled images, and
determining similarities between the input image and the enrolled images based on the correlations and the sizes of the overlap regions; and
adding the input image to the database based on whether the first similarity is below an upper threshold, the upper threshold based on the sizes of overlap regions between the enrolled images.