Granted Patent B2

US 12,575,745 · App. 18/452,862 · Granted Mar 17, 2026

Inventor: Ki Bum Park (Daegu, KR)

Assignee: GB SOFT CO., LTD.

A61B5/021A61B5/1032A61B5/7207G06T5/20G06T5/70G06T7/0016G06T7/90G16H50/30G06T2207/10024G06T2207/20056G06T2207/20084G06T2207/30088

Abstract

Claims (41)

1 . A method of measuring a biometric index of a subject using an artificial neural network in a non-contact manner, performed by one or more processors, the method comprising:

obtaining a plurality of image frames of the subject;

obtaining a first color channel value, a second color channel value, and a third color channel value for at least one image frame included in the plurality of image frames;

calculating a first difference value and a second difference value based on the first color channel value, the second color channel value, and the third color channel value for at least one image frame included in the plurality of image frames, wherein the first difference value is a difference value between the first color channel value and the second color channel value for a same image frame, and the second difference value indicates a difference value between the first color channel value and the third color channel value for the same image frame;

obtaining a first characteristic value based on the first difference value for at least one image frame included in a first image frame group for a first image frame group obtained for a first predetermined time;

obtaining a second characteristic value based on the second difference value for at least one image frame included in the first image frame group;

obtaining time series data based on the first characteristic value and the second characteristic value; obtaining biosignal characteristic data based on the time series data; and

obtaining a biometric index of the subject using a biometric index obtaining model based on the biosignal characteristic data.

2 . The method of claim 1 , wherein obtaining biosignal characteristic data based on the time series data comprises:

filtering the time series data using a band pass filter; and

obtaining the biosignal characteristic data using the filtered time series data.

3 . The method of claim 2 , wherein filtering the time series data using the band pass filter comprises:

filtering the time series data using a band pass filter having a frequency band corresponding to a heart rate of the subject.

4 . The method of claim 1 , wherein filtering the time series data using a band pass filter corresponding to a heart rate of the subject comprises:

performing Fourier transform on the time series data to extract a frequency band corresponding to the heart rate of the subject.

5 . The method of claim 1 , wherein obtaining biosignal characteristic data based on the time series data comprises:

dividing the time series data into a plurality of detailed time series data; and

extracting target time series data based on the divided time series data, wherein the target time series data comprises a first maximum value and a second maximum value smaller than the first maximum value.

6 . The method of claim 5 , wherein the biosignal characteristic data is generated based on at least one of a value at a start time point of the target time series data, the first maximum value, and a value at an end time point of the target time series data.

7 . The method of claim 5 , wherein the biosignal characteristic data is generated based on time series data obtained by first differentiating the target time series data and time series data obtained by second differentiating.

8 . The method of claim 1 , wherein the biometric index of the subject comprises at least one of a lowest blood pressure or a highest blood pressure of the subject.

9 . The method of claim 1 , wherein the first, second, and third color channels are color channels according to an RGB color space.

10 . The method of claim 9 , wherein the first color channel is set to a Green channel, the second color channel is set to a Red channel, and the third color channel is set to a Blue channel.

11 . The method of claim 1 ,

wherein the first characteristic value is obtained based on a first deviation value of the first difference value for at least one image frame included in the first image frame group,

wherein the second characteristic value is obtained based on a second deviation value of the second difference value for at least one image frame included in the first image frame group,

wherein the first deviation value is calculated based on a first difference value for the at least one image frame and an average value of the first difference value for the first image frame group, and

wherein the second deviation value is calculated based on a second difference value for the at least one image frame and an average value of the second difference value for the first image frame group.

12 . The method of claim 11 , wherein the first characteristic value and the second characteristic value are normalized values.

13 . The method of claim 12 ,

wherein the first characteristic value is a value normalized by a first standard deviation value, and the second characteristic value is a value normalized by a second standard deviation value,

wherein the first standard deviation value is a standard deviation value of the first difference value for the first image frame group, and the second standard deviation value is a standard deviation value of the second difference value for the first image frame group.

14 . The method of claim 1 , wherein the time series data is obtained based on third characteristic data obtained as a sum of the first characteristic value and the second characteristic value.

15 . A non-transitory computer-readable storage medium storing instructions thereon, the instructions, when executed by a processor cause the processor to:

obtaining a plurality of image frames of a subject;

obtain a first color channel value, a second color channel value, and a third color channel value for at least one image frame included in the plurality of image frames;

calculate a first difference value and a second difference value based on the first color channel value, the second color channel value, and the third color channel value for at least one image frame included in the plurality of image frames, wherein the first difference value is a difference value between the first color channel value and the second color channel value for a same image frame, and the second difference value indicates a difference value between the first color channel value and the third color channel value for the same image frame;

obtain a first characteristic value based on the first difference value for at least one image frame included in a first image frame group for a first image frame group obtained for a first predetermined time;

obtain a second characteristic value based on the second difference value for at least one image frame included in the first image frame group;

obtain time series data based on the first characteristic value and the second characteristic value; obtaining biosignal characteristic data based on the time series data; and

obtain a biometric index of the subject using a biometric index obtaining model based on the biosignal characteristic data.

Assignments (1)

Priority Claims (1)

Continuity (2)

References Cited (21)