Granted Patent B2

US 12,572,687 · App. 18/360,346 · Granted Mar 10, 2026

Inventors: Md. Imtiaz Hossain (Dacca, BD); Tahasin Mahmud (Dacca, BD); Aba Kowser (Dacca, BD)

Assignee: Samsung Electronics Co., Ltd.

G06F21/6245G06T11/60G06V10/25G06V10/44

Abstract

Claims (73)

1 . A method, performed by an electronic device, of removing sensitive information from image data, the method comprising:

obtaining first image data;

determining whether the first image data includes at least one piece of sensitive information, by using a sensitive information detection model configured to detect sensitive information, wherein the sensitive information comprises disease information;

in case that the first image data is determined as including the at least one piece of sensitive information, identifying a first area related to the at least one piece of sensitive information in the first image data based on the first image data and the at least one piece of sensitive information; and

generating second image data based on data in the first area reconstructed based on an image reconstruction model, and

wherein the identifying of the first area comprises:

providing the first image data and the at least one piece of sensitive information to an explainable artificial intelligence model to identify at least one feature point that contributes to detection of the at least one piece of sensitive information from the first image data, and

wherein the first area is determined based on the identified at least one feature point.

2 . The method of claim 1 , wherein the generating of the second image data comprises:

masking the first area in the first image data; and

generating the second image data based on the first image data in which the first area is masked, by using the image reconstruction model.

3 . The method of claim 1 , further comprising:

outputting at least one of contents of the at least one piece of sensitive information or the first image data in which the first area is marked.

4 . The method of claim 1 , further comprising:

receiving a request for removal of the at least one piece of sensitive information from the first image data,

wherein the generating of the second image data comprises generating the second image data in response to the request for removal of the at least one piece of sensitive information.

5 . The method of claim 1 , further comprising:

determining whether the second image data includes the at least one piece of sensitive information by using the sensitive information detection model;

when the second image data is determined as including the at least one piece of sensitive information, identifying a second area related to the at least one piece of sensitive information in the second image data based on the second image data and the sensitive information detection model; and

generating third image data by modifying the second area by using the image reconstruction model.

6 . The method of claim 1 , further comprising:

receiving a request for sharing or storage of the first image data or the second image data;

determining whether the second image data includes the at least one piece of sensitive information by using the sensitive information detection model; and

when the second image data is determined as not including the at least one piece of sensitive information, performing sharing or storage of the second image data based on the request for sharing or storage.

7 . The method of claim 1 ,

wherein the first image data is a first frame included in a plurality of frames constituting first video data, and

wherein the second image data is a second frame generated by modifying the first area by using the image reconstruction model.

8 . The method of claim 7 ,

wherein, among the plurality of frames, at least some frames which are determined as including sensitive information are emphasized, by using an emphasis mark to indicate the at least some frames, and output on a display, and

wherein the at least some frames include the first frame.

9 . The method of claim 7 , further comprising:

determining whether the second frame includes the at least one piece of sensitive information by using the sensitive information detection model;

when the second frame is determined as including the at least one piece of sensitive information, identifying a second area related to the at least one piece of sensitive information in the second frame based on the second frame and the sensitive information detection model; and

generating a third frame by modifying the second area by using the image reconstruction model.

10 . The method of claim 7 , further comprising:

determining whether the second frame includes the at least one piece of sensitive information, by using the sensitive information detection model; and

when the second frame is determined as not including the at least one piece of sensitive information, generating second video data including the second frame.

11 . The method of claim 10 , further comprising:

receiving a request for sharing or storage of the first video data or the second video data; and

performing sharing or storage of the second video data based on the request for sharing or storage.

12 . An electronic device configured to remove sensitive information from image data, the electronic device comprising:

memory, comprising one or more storage media, storing at least one instruction; and

at least one processor communicatively coupled to the memory and configured to execute the at least one instruction stored in the memory,

wherein, by executing the at least one instruction, the at least one processor is further configured to:

obtain first image data,

determine whether the first image data includes at least one piece of sensitive information, by using a sensitive information detection model configured to detect sensitive information, wherein the sensitive information comprises disease information,

in case that the first image data is determined as including the at least one piece of sensitive information, identify a first area related to the at least one piece of sensitive information in the first image data based on the first image data and the at least one piece of sensitive information, and

generate second image data based on data in the first area reconstructed based on an image reconstruction model, and

wherein, by executing the at least one instruction, the at least one processor is further configured to in case that the first image data is determined as including the at least one piece of sensitive information, provide the first image data and the at least one piece of sensitive information to an explainable artificial intelligence model to identify at least one feature point that contributes to detection of the at least one piece of sensitive information from the first image data, and

wherein the first area is determined based on the identified at least one feature point.

13 . The electronic device of claim 12 , wherein, by executing the at least one instruction, the at least one processor is further configured to:

mask the first area in the first image data; and

generate the second image data based on the first image data in which the first area is masked, by using the image reconstruction model.

14 . The electronic device of claim 12 , wherein, by executing the at least one instruction, the at least one processor is further configured to output at least one of contents of the at least one piece of sensitive information or the first image data in which the first area is marked.

15 . The electronic device of claim 12 , wherein, by executing the at least one instruction, the at least one processor is further configured to:

receive a request for removal of the at least one piece of sensitive information from the first image data; and

generate the second image data based on the request for removal of the at least one piece of sensitive information.

16 . The electronic device of claim 12 , wherein, by executing the at least one instruction, the at least one processor is further configured to:

determine whether the second image data includes the at least one piece of sensitive information by using the sensitive information detection model;

when the second image data is determined as including the at least one piece of sensitive information, identify a second area related to the at least one piece of sensitive information in the second image data based on the second image data and the sensitive information detection model; and

generate third image data by modifying the second area by using the image reconstruction model.

17 . The electronic device of claim 12 , wherein, by executing the at least one instruction, the at least one processor is further configured to:

receive a request for sharing or storage of the first image data or the second image data;

determine whether the second image data includes the at least one piece of sensitive information by using the sensitive information detection model; and

when the second image data is determined as not including the at least one piece of sensitive information, perform sharing or storage of the second image data based on the request for sharing or storage.

18 . At least one non-transitory computer-readable recording medium having recorded thereon a program including computer-executable instructions that, when executed by one or more processors of an electronic device, cause the electronic device to perform operations, the operations comprising:

obtaining first image data;

determining whether the first image data includes at least one piece of sensitive information, by using a sensitive information detection model configured to detect sensitive information, wherein the sensitive information comprises disease information;

in case that the first image data is determined as including the at least one piece of sensitive information, identifying a first area related to the at least one piece of sensitive information in the first image data based on the first image data and the at least one piece of sensitive information; and

generating second image data based on data in the first area reconstructed based on an image reconstruction model, and

wherein the operations further comprises:

in case that the first image data is determined as including the at least one piece of sensitive information, providing the first image data and the at least one piece of sensitive information to an explainable artificial intelligence model to identify at least one feature point that contributes to detection of the at least one piece of sensitive information from the first image data, and

wherein the first area is determined based on the identified at least one feature point.

Assignments (1)

Priority Claims (1)

Continuity (1)

References Cited (23)