Video display device and operating method therefor

Patent metadata

Field	Value
Publication number	US-11514150-B2
Application number	US-201816642154-A
Country	US
Kind code	B2
Filing date	Sep 5, 2018
Priority date	Sep 5, 2017
Publication date	Nov 29, 2022
Grant date	Nov 29, 2022

Abstract

Provided is an artificial intelligence (AI) system simulating functions of a human brain, such as recognition and decision, by using a machine learning algorithm, such as deep-learning.Image display apparatuses are more convenient for a user, by performing user authentication by using an authentication image set generated based on an object recognized from content viewed by the user.

First claim

The invention claimed is:

1. An image display apparatus comprising:
a display;
a memory storing at least one instruction;
a processor configured to execute the at least one instruction stored in the memory to:
recognize at least one object from first content output on the display and viewed by a user, by using a learning model using at least one neural network;
store an image indicating the at least one object recognized from the first content in a database as a true image;
in response to a user authentication request, select at least one first true image and at least one first false image respectively from a plurality of true images and a plurality of false images stored in the database, and control the display to output an authentication image set comprising the selected at least one first true image and at least one first false image; and
perform user authentication in response to a user input of selecting at least one image from the authentication image set.

2. The image display apparatus of claim 1 ,
wherein the first content is content viewed within a pre-set time from a point of time when the user authentication request is received, and
the image indicating the at least one object recognized from the first content includes at least one object extracted from the first content.

3. The image display apparatus of claim 1 ,
wherein the processor is further configured to execute the at least one instruction to:
recognize at least one object from second content viewed a pre-set time before a point of time when the user authentication request is received, by using the learning model using the at least one neural network; and
obtain an image indicating at least one object recognized from the second content as one of the plurality of false image.

4. The image display apparatus of claim 1 ,
wherein the processor is further configured to execute the at least one instruction to:
recognize at least one face from the first content by using the learning model using the at least one neural network; and
store, as the true image, an image indicating a person corresponding to each of the at least one face recognized from the first content in the database.

5. The image display apparatus of claim 1 ,
wherein the at least one object comprises
at least one of a person, a background, and a thing appearing in at least one frame among a plurality of frames forming the first content.

6. The image display apparatus of claim 1 ,
wherein the processor is further configured to execute the at least one instruction to:
recognize the at least one object from a plurality of frames forming the first content by using the learning model using the at least one neural network; and
store, as the true image, an image indicating an object included in the plurality of frames at least a pre-set number of times among the recognized at least one object in the database.

7. The image display apparatus of claim 1 , wherein the processor is further configured to execute the at least one instruction to:
generate at least one modified image by modifying at least one of the at least one first true image and at least one first false image via an adversarial example technique; and
control the display to output the authentication image set comprising the at least one modified image and the remaining of the at least one first true image and the at least one first false image, which are not modified.

8. The image display apparatus of claim 7 , wherein the processor is further configured to execute the at least one instruction to:
generate noise information in response to the user authentication request; and generate the at least one modified image by modifying at least one of the at least one first true image and the at least one first false image, based on the generated noise information.

9. The image display apparatus of claim 8 ,
wherein the noise information comprises information about at least one of a method of generating a noise signal to be applied to at least one image to be modified, a form of the noise signal, and an amount of the noise signal.

10. The image display apparatus of claim 1 , wherein the processor is further configured to execute the at least one instruction to:
generate at least one second true image by modifying at least one true image selected from the at least one first true image via an adversarial example technique; and control the display to output the authentication image set comprising the at least one second true image, remaining of the at least one first true image excluding the selected at least one true image, and the at least one first false image.

11. The image display apparatus of claim 1 , wherein the processor is further configured to execute the at least one instruction to:
generate at least one modified image by adding a noise signal to at least one of the at least one first true image and at least one first false image; and
control the display to output the authentication image set comprising the at least one modified image and remaining of the at least one first true image and at least one first false image, to which the noise signal is not added.

12. The image display apparatus of claim 1 ,
wherein the processor is further configured to execute the at least one instruction to
receive, from an external server, at least one of the plurality of false images stored in the database and the learning model.

13. The image display apparatus of claim 1 ,
wherein the processor is further configured to execute the at least one instruction to
determine a user who requested the user authentication as an authorized user in response to a user input of selecting all of the at least one first true image from the authentication image set.

14. The image display apparatus of claim 13 ,
wherein the processor is further configured to execute the at least one instruction to
control the display to output a new authentication image set in response to a user input that failed to select all of the at least one first true image from the authentication image set.

15. An operating method of an image display apparatus, the operating method comprising:
recognizing at least one object from first content output on a display of the image display apparatus and viewed by a user, by using a learning model using at least one neural network;
storing an image indicating the at least one object recognized from the first content in a database as a true image;
in response to a user authentication request, selecting at least one first true image and at least one first false image respectively from a plurality of true images and a plurality of false images stored in the database, and outputting an authentication image set comprising the selected at least one first true image and at least one first false image; and
performing user authentication in response to a user input of selecting at least one image from the authentication image set.

Assignees

• Samsung Electronics Co Ltd

Inventors

• Cho Eun-Ae
• Kim Jinhyun
• Park Gihoon
• Kwon Jae-Ook

Classifications

• G06N3/088

  Non-supervised learning, e.g. competitive learning · CPC title

• G06T2207/20081

  Training; Learning · CPC title

• G06F21/36 Primary

  by graphic or iconic representation · CPC title

• G06N3/084

  Backpropagation, e.g. using gradient descent · CPC title

• G06V40/172

  Classification, e.g. identification · CPC title