Check patentability & draft patents in minutes with Patsnap Eureka AI!

Remote heart rate monitoring algorithm based on neural network model

A neural network model and heart rate monitoring technology, applied in the field of computer vision, can solve the problems of ignoring background information, affecting the accuracy and stability of the algorithm, and separating heart rate related information, so as to achieve the effect of improving accuracy

Pending Publication Date: 2022-04-12
FUDAN UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, studies have shown that there are still many problems in the existing remote heart rate monitoring, such as only considering the facial information and ignoring the background information, or thinking that there is only a simple linear relationship between the background information and the facial information, etc., which makes the existing methods in from It becomes difficult to separate heart rate-related information from facial information, which greatly affects the accuracy and stability of the algorithm

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Remote heart rate monitoring algorithm based on neural network model
  • Remote heart rate monitoring algorithm based on neural network model
  • Remote heart rate monitoring algorithm based on neural network model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0039] (1) Obtain a video containing a human face, and preprocess the video. figure 1For the overall process of video preprocessing, the face and background area are first extracted based on the 81 feature points of the face, and then the face and background area are respectively embedded to obtain the corresponding features. The embedding refers to mapping the data from the original image sequence to a high-dimensional time series . The preprocessing algorithm steps are as follows:

[0040] (1.1) Use the existing open source algorithm to detect the 81 key points of the face in each frame of the video; here, the key point detection uses the DLIB [11] open source software library, and the distribution of the detected 81 key points is as follows figure 2 shown;

[0041] (1.2) Use the following key points of 81 human face key points to represent the face area in a rectangle surrounded by the following rules: use (x 1 ,y 1 ) and (x 2 ,y 2 ) represents the upper-left and low...

Embodiment 2

[0052] (1) same as embodiment 1;

[0053] (1.1) is identical with embodiment 1;

[0054] (1.2) is identical with embodiment 1;

[0055] (1.3) According to the facial area obtained in (1.2), cut out a background area with a certain width and height around it; Combined with the facial area of ​​width w and height h obtained in (1.2) to obtain the area containing the background, such as image 3 The rectangular area formed by the union of the green box and the red box shown;

[0056] (1.4) is identical with embodiment 1;

[0057] (1.5)Feat f The calculation is the same as in Example 1, Feat b The calculation of is performed on the background area obtained in (1.3) and calculates Feat f same operation.

[0058] (2) same as embodiment 1;

[0059] (2.1) is identical with embodiment 1;

[0060] (2.2) The background flow network that uses the characteristics of the region obtained in (1.3) as the input, its output minus the output of the foreground flow network, is used to re...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention belongs to the technical field of computer vision, and particularly relates to a remote heart rate monitoring algorithm based on a multi-stream neural network model. According to the method, the double-flow Transform neural network is adopted, the face information and the background information can be processed at the same time, the background information is used for guiding separation of a face information center rate related signal and a noise signal, and the stability of remote heart rate monitoring can be remarkably improved; wherein the multi-stream neural network is formed by connecting more than two paths of feature extraction networks in series with a result regression network; the feature extraction network is composed of more than two mutually independent networks with the same structure; the result regression network is composed of a neural network, and the output is the predicted heart rate; according to the method, the heart rate is calculated through face video analysis, the face and background information is considered at the same time, inhibition of non-physiological signals in the face information is guided through the background information, and therefore the accuracy of a remote heart rate monitoring algorithm is remarkably improved.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and in particular relates to a remote heart rate monitoring algorithm. Background technique [0002] The heart is one of the most important organs in the human body, and the activities of almost all cells in the human body are inseparable from its support. Therefore, having a healthy heart has become a prerequisite for a person's health. The latest cardiovascular disease report released by the National Center for Cardiovascular Diseases counts the incidence of cardiovascular diseases from 1990 to 2017. The results show that the mortality rate of cardiovascular diseases is much higher than that of other types of diseases, and the mortality rate shows a slow upward trend. Among the deaths due to diseases in 2017, nearly half of the deaths were caused by cardiovascular diseases, far exceeding other types of diseases such as tumors and respiratory systems. In view of the high disability rate...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06V40/16G06V10/82G06N3/08A61B5/00A61B5/024
Inventor 康家琪杨夙
Owner FUDAN UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com