CN109770885A - A kind of examing heartbeat fastly method based on preview frame - Google Patents

A kind of examing heartbeat fastly method based on preview frame Download PDF

Info

Publication number
CN109770885A
CN109770885A CN201910226962.1A CN201910226962A CN109770885A CN 109770885 A CN109770885 A CN 109770885A CN 201910226962 A CN201910226962 A CN 201910226962A CN 109770885 A CN109770885 A CN 109770885A
Authority
CN
China
Prior art keywords
frame
heart rate
face
camera
bvp
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN201910226962.1A
Other languages
Chinese (zh)
Inventor
杨学志
高冠群
王定良
刘雪南
金兢
方帅
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hefei University of Technology
Original Assignee
Hefei University of Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Hefei University of Technology filed Critical Hefei University of Technology
Priority to CN201910226962.1A priority Critical patent/CN109770885A/en
Publication of CN109770885A publication Critical patent/CN109770885A/en
Pending legal-status Critical Current

Links

Abstract

The examing heartbeat fastly method based on preview frame that the invention discloses a kind of, this method respectively detects face or finger tip using mobile phone camera, when when using face datection, open equipment, call front camera, make face's face mobile phone camera, it is ensured that include face information in sequence of frames of video collected;When using finger tip detection, equipment is opened, rear camera is called, finger tip is covered on rear camera;In the one-dimension array for being 500 by the frame sequence deposit length of camera acquisition;While circle collection is to continuous preview frame, processing data are simultaneously saved, and start to calculate when reaching 256 frame;Start continuous heart rate detection.The present invention fetters problem for the complicated for operation and limbs of contact heart rate measurement, this method is without using electrode or sensor contacts human body, only it can be detected heart rate by regular handset, detection efficiency and the usage experience of testee are effectively improved, prolonged rhythm of the heart and disease prevention are suitable for.

Description

A kind of examing heartbeat fastly method based on preview frame
Technical field
It is specifically a kind of based on the fast of preview frame the present invention relates to video image processing application and heart rate detection field Fast heart rate detection method.
Background technique
With the development of computer technology and universal, more and more computer theories are related to medical domain, are applied to The links such as medical diagnosis and routine health monitoring, have played powerful booster action for the progress of medicine.Heart rate is reflection One of index of human health status, and judge most basic one of the physical signs of cardiovascular disease.The clinical heart traditionally Rate detection means needs multiple positions of the 12 lead line contact human body using electrocardiograph, trivial operations, the degree of automation It is not high, there are higher professional knowledge requirement, the heart rate detection not being suitable under common scenarios to user.
Photoplethysmography (Photoplethysmography, PPG) is to carry out the heart using computer vision technique Rate detects most basic method, it passes through light emitting diode subcutaneously tissue emissions feux rouges, and feux rouges is by subcutaneous capillary network In hemoglobin absorption, reflect or be transmitted to the photistor of the other end, signal is after treatment and in arterial blood Amount of hemoglobin be positively correlated, by measure intensity of reflected light, trace volumetric blood pulse (Blood volume Pulse, BVP) after signal, it can directly calculate heart rate.
There is a kind of contactless heart rate detection method using general network camera at present, this method utilizes independent element Three average color traces are separated into three Ji Yuan letters by analysis (Independent Component Analysis, ICA) Number, by the power Spectral Estimation heart rate for analyzing second base source signal.However, second Ji Yuan of ICA output can not table Show PPG signal.
The above method can accurately measure heart rate value by contactless video method under certain scene, still Calculating process is excessively very long, and for tested personnel, excessively high time cost will be with non-contact detection method bring Convenience offsets, and impracticable.There are many more other heart rate detection methods, such as acquire data using video method, first By shooting video, video is reprocessed, BVP waveform is obtained and carries out calculating heart rate.But handle recording and the processing time of video It is longer, it is not nearly enough quick, while also needing to will receive image or video acquisition field using USB camera or IP Camera The limitation of scape.
Summary of the invention
The examing heartbeat fastly method based on preview frame that the technical problem to be solved in the present invention is to provide a kind of, this method benefit Video acquisition is carried out with smart phone, can simply, efficiently obtain heart rate Value Data.
Examing heartbeat fastly method based on preview frame of the invention includes the following steps:
The acquisition of step 1) video frame, takes Smartphone device, is carried out respectively to face or finger tip using mobile phone camera Detection calls front camera, makes face's face mobile phone camera, it is ensured that acquired when using face datection, opening equipment Sequence of frames of video in include face information;When using finger tip detection, equipment is opened, rear camera is called, finger tip is covered It covers on rear camera;
The frame sequence of camera acquisition is stored in the one-dimension array that length is 500 by step 2) data prediction;Circulation While collecting continuous preview frame, handles data and save, start to calculate when reaching 256 frame;
Step 3) rate calculation starts continuous heart rate detection, specifically includes:
The input of step 3.1) data, in the one-dimension array that length is 500, circle collection is same to continuous preview frame When, it handles data and saves, start to calculate when reaching 256 frame;20 frame of each counting period, present frame are expressed as cnt, array It is taken as { cnt -255, cnt } and substitutes into calculating;
Selected, the face recognition algorithms provided using Google of step 3.2) ROI region, carry out face inspection to each frame It surveys, obtains region vector Rect (x, y, w, h) only comprising face information, as ROI region, wherein Rect indicates a matrix Region, the initial vertex of x and y representing matrix, the length and width of w and h representing matrix select green logical when doing pixel averagely Road is calculated;
Step 3.3) adds image rendering, when from camera readjustment to preview frame, adds one layer of rendering automatically using algorithm, The rendering that a pale red (#FF5151) is added when face, improves the pixel value of RGB channel, while reducing background interference;Finger tip Mono- pale red (#FFB5B5) rendering of Shi Tianjia, reduces the pixel value in red green channel, prevents from being saturated;
Step 3.4) spectra calculation heart rate obtains heart rate value by calculating BVP [x] [y] [i];
Step 3.5) takes average calculating heart rate: by calculating average heart rate array, obtaining heart rate value.
In step 1), the preposition peak of currently used equipment, 25fps frame per second are set by the parameter of mobile phone camera And RGB color domain, take G path computation.
Step 3.4) specifically includes:
Step 3.4.1) the BVP value of all pixels point of the i-th frame (i=1,2,3 ... S) is summed, obtain one-dimensional BVP Semaphore B [i], the semaphore can be with secondary indication BVP signals;
Step 3.4.2) Fast Fourier Transform (FFT) is carried out to B, obtain its power spectrum PBvp:
F (t)=fft (B (t));
PBvp(t)=| F (t) |2
Wherein, fft is Fast Fourier Transform (FFT) function;
Step 3.4.3) heart rate value HRCalculating:
T=max { PBvp(t)};
Wherein, fpsThe frame per second of video.
Step 3.5) specifically includes:
Step 3.5.1) all real-time heart rate value additions are averaged;
Step 3.5.2) array is traversed, the real-time heart rate value that will be greater than average value 10 is left out, and the reality of average value 10 will be less than When heart rate value leave out;
Step 3.5.3) average calculating heart rate value is taken again.
The principle of the present invention and advantageous effect are analyzed below.
Android phone equipment can be used in the present invention, and camera is controlled using Android program and carries out adopting for preview frame Collection.It is divided into both direction in acquisition, one is calling Front camera to acquire face information, carry out face can be synchronized in collection process Detection, it is ensured that include face information in sequence of frames of video collected;Another kind is to call rearmounted camera acquisition finger tip information.It adopts Collect 500 frame informations to be calculated.The number of statistics frame, also can be dynamic to reject in the detection process according to the variation of pixel value Fall the frame (finger tip or face shake) of interference after collecting 256 frame preview frames, carries out rate calculation.Every 20 frame calculates Primary heart rate in real time simultaneously saves.Finally obtain a heart rate value sequence.Calculate average value.And weed out higher than average value 10 and low 10 value (being considered interference information value, it may be possible to as caused by the shake of hand or face).Recalculate heart rate average value.Both It obtains a result.
The present invention proposes a kind of based on preview frame for the complicated for operation and limbs constraint problem of contact heart rate measurement Examing heartbeat fastly method.The technology only can be detected by regular handset without using electrode or sensor contacts human body Heart rate.Detection efficiency and the usage experience of testee are effectively improved, prolonged rhythm of the heart and disease prevention are suitable for.
The present invention provides two kinds of detection modes, face and finger tip.It is calculated using preview frame, does not need recorded video, saved Time, improve calculating speed.
Present invention employs multi-thread design structures, introduce the thought of concurrent operation, use in the readjustment part of camera Sub thread, effectively raises arithmetic speed, so that contactless heart rate detection technology is had better usage experience, improves The practical significance of the technology.
Detailed description of the invention
Invention is further described in detail with reference to the accompanying drawings and detailed description:
Fig. 1 is the schematic process flow diagram of the method for the present invention.
Specific embodiment
Referring to attached drawing, the examing heartbeat fastly method of the invention based on preview frame includes the following steps:
The acquisition of step 1) video frame, takes Smartphone device, is carried out respectively to face or finger tip using mobile phone camera Detection calls front camera, makes face's face mobile phone camera, it is ensured that acquired when using face datection, opening equipment Sequence of frames of video in include face information;When using finger tip detection, equipment is opened, rear camera is called, finger tip is covered It covers on rear camera;
The frame sequence of camera acquisition is stored in the one-dimension array that length is 500 by step 2) data prediction;Circulation While collecting continuous preview frame, handles data and save, start to calculate when reaching 256 frame;
Step 3) rate calculation starts continuous heart rate detection, specifically includes:
The input of step 3.1) data, in the one-dimension array that length is 500, circle collection is same to continuous preview frame When, it handles data and saves, start to calculate when reaching 256 frame;20 frame of each counting period, present frame are expressed as cnt, array It is taken as { cnt -255, cnt } and substitutes into calculating;
Selected, the face recognition algorithms provided using Google of step 3.2) ROI region, carry out face inspection to each frame It surveys, obtains region vector Rect (x, y, w, h) only comprising face information, (Region Of Interest feels emerging as ROI Interesting region) region, wherein Rect indicates a matrix area, the initial vertex of x and y representing matrix, the length of w and h representing matrix And width, select green channel to be calculated when doing pixel averagely;
Step 3.3) adds image rendering, when from camera readjustment to preview frame, adds one layer of rendering automatically using algorithm, The rendering that a pale red (#FF5151) is added when face, improves the pixel value of RGB channel, while reducing background interference;Finger tip Mono- pale red (#FFB5B5) rendering of Shi Tianjia, reduces the pixel value in red green channel, prevents from being saturated;
Step 3.4) spectra calculation heart rate obtains heart rate value by calculating BVP [x] [y] [i];
Step 3.5) takes average calculating heart rate: by calculating average heart rate array, obtaining heart rate value.
In step 1), the preposition peak of currently used equipment, 25fps frame per second are set by the parameter of mobile phone camera And RGB color domain, take G path computation.
Step 3.4) specifically includes:
Step 3.4.1) the BVP value of all pixels point of the i-th frame (i=1,2,3 ... S) is summed, obtain one-dimensional BVP Semaphore B [i], the semaphore can be with secondary indication BVP signals;
Step 3.4.2) Fast Fourier Transform (FFT) is carried out to B, obtain its power spectrum PBvp:
F (t)=fft (B (t));
PBvp(t)=| F (t) |2
Wherein, fft is Fast Fourier Transform (FFT) function;
Step 3.4.3) heart rate value HRCalculating:
T=max { PBvp(t)};
Wherein, fpsThe frame per second of video.
Step 3.5) specifically includes:
Step 3.5.1) all real-time heart rate value additions are averaged;
Step 3.5.2) array is traversed, the real-time heart rate value that will be greater than average value 10 is left out, and the reality of average value 10 will be less than When heart rate value leave out;
Step 3.5.3) average calculating heart rate value is taken again.
The principle of the present invention and advantageous effect are analyzed below.
Android phone equipment can be used in the present invention, and camera is controlled using Android program and carries out adopting for preview frame Collection.It is divided into both direction in acquisition, one is calling Front camera to acquire face information, carry out face can be synchronized in collection process Detection, it is ensured that include face information in sequence of frames of video collected;Another kind is to call rearmounted camera acquisition finger tip information.It adopts Collect 500 frame informations to be calculated.The number of statistics frame, also can be dynamic to reject in the detection process according to the variation of pixel value Fall the frame (finger tip or face shake) of interference after collecting 256 frame preview frames, carries out rate calculation.Every 20 frame calculates Primary heart rate in real time simultaneously saves.Finally obtain a heart rate value sequence.Calculate average value.And weed out higher than average value 10 and low 10 value (being considered interference information value, it may be possible to as caused by the shake of hand or face).Recalculate heart rate average value.Both It obtains a result.
The present invention proposes a kind of based on preview frame for the complicated for operation and limbs constraint problem of contact heart rate measurement Examing heartbeat fastly method.The technology only can be detected by regular handset without using electrode or sensor contacts human body Heart rate.Detection efficiency and the usage experience of testee are effectively improved, prolonged rhythm of the heart and disease prevention are suitable for.
The present invention provides two kinds of detection modes, face and finger tip.It is calculated using preview frame, does not need recorded video, saved Time, improve calculating speed.
Present invention employs multi-thread design structures, introduce the thought of concurrent operation, use in the readjustment part of camera Sub thread, effectively raises arithmetic speed, so that contactless heart rate detection technology is had better usage experience, improves The practical significance of the technology.
In conclusion the present invention is not limited to above-mentioned specific embodiments.Those skilled in the art are not departing from skill of the present invention Under the premise of art scheme, several changes or modification can be done, above-mentioned change or modification each fall within protection scope of the present invention.

Claims (4)

1. a kind of examing heartbeat fastly method based on preview frame, it is characterized in that including the following steps:
The acquisition of step 1) video frame, takes Smartphone device, is detected respectively to face or finger tip using mobile phone camera, When using face datection, opening equipment, front camera is called, makes face's face mobile phone camera, it is ensured that view collected It include face information in frequency frame sequence;When using finger tip detection, equipment is opened, rear camera is called, finger tip is covered on On rear camera;
The frame sequence of camera acquisition is stored in the one-dimension array that length is 500 by step 2) data prediction;Circle collection While to continuous preview frame, processing data are simultaneously saved, and start to calculate when reaching 256 frame;
Step 3) rate calculation starts continuous heart rate detection, specifically includes:
The input of step 3.1) data, in the one-dimension array that length is 500, while circle collection is to continuous preview frame, Processing data simultaneously save, and start to calculate when reaching 256 frame;20 frame of each counting period, present frame are expressed as cnt, and array takes It substitutes into and calculates for { cnt -255, cnt };
Selected, the face recognition algorithms provided using Google of step 3.2) ROI region, are carried out Face datection to each frame, obtained To the region vector Rect (x, y, w, h) for only including face information, as ROI region, wherein Rect indicates a matrix area, The initial vertex of x and y representing matrix, the length and width of w and h representing matrix select green channel when doing pixel averagely It is calculated;
Step 3.3) adds image rendering, when from camera readjustment to preview frame, adds one layer of rendering, face automatically using algorithm The rendering of mono- pale red (#FF5151) of Shi Tianjia, improves the pixel value of RGB channel, while reducing background interference;Add when finger tip Add pale red (#FFB5B5) rendering, reduces the pixel value in red green channel, prevent from being saturated;
Step 3.4) spectra calculation heart rate obtains heart rate value by calculating BVP [x] [y] [i];
Step 3.5) takes average calculating heart rate: by calculating average heart rate array, obtaining heart rate value.
2. the examing heartbeat fastly method based on preview frame as described in claim 1, it is characterized in that in step 1), by mobile phone The parameter of camera is set as the preposition peak of currently used equipment, 25fps frame per second and RGB color domain, takes the channel G meter It calculates.
3. the examing heartbeat fastly method based on preview frame as described in claim 1, it is characterized in that the step 3.4) is specific Include:
Step 3.4.1) the BVP value of all pixels point of the i-th frame (i=1,2,3 ... S) is summed, obtain one-dimensional BVP signal It measures B [i], which can be with secondary indication BVP signal;
Step 3.4.2) Fast Fourier Transform (FFT) is carried out to B, obtain its power spectrum PBvp:
F (t)=fft (B (t)) (3)
PBvp(t)=| F (t) |2 (4)
Wherein, fft is Fast Fourier Transform (FFT) function;
Step 3.4.3) heart rate value HRCalculating:
T=max { PBvp(t)} (5)
Wherein, fpsThe frame per second of video.
4. the examing heartbeat fastly method based on preview frame as described in claim 1, it is characterized in that the step 3.5) is specific Include:
Step 3.5.1) all real-time heart rate value additions are averaged;
Step 3.5.2) array is traversed, the real-time heart rate value that will be greater than average value 10 is left out, and the real-time heart of average value 10 will be less than Rate value is left out;
Step 3.5.3) average calculating heart rate value is taken again.
CN201910226962.1A 2019-03-25 2019-03-25 A kind of examing heartbeat fastly method based on preview frame Pending CN109770885A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201910226962.1A CN109770885A (en) 2019-03-25 2019-03-25 A kind of examing heartbeat fastly method based on preview frame

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201910226962.1A CN109770885A (en) 2019-03-25 2019-03-25 A kind of examing heartbeat fastly method based on preview frame

Publications (1)

Publication Number Publication Date
CN109770885A true CN109770885A (en) 2019-05-21

Family

ID=66490416

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201910226962.1A Pending CN109770885A (en) 2019-03-25 2019-03-25 A kind of examing heartbeat fastly method based on preview frame

Country Status (1)

Country Link
CN (1) CN109770885A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113397519A (en) * 2021-08-05 2021-09-17 季华实验室 Cardiovascular health state detection device

Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2014064575A2 (en) * 2012-10-23 2014-05-01 Koninklijke Philips N.V. Device and method for obtaining vital sign information of a living being
CN103826532A (en) * 2011-08-22 2014-05-28 Isis创新有限公司 Remote monitoring of vital signs
CN105989357A (en) * 2016-01-18 2016-10-05 合肥工业大学 Human face video processing-based heart rate detection method
CN106491114A (en) * 2016-10-25 2017-03-15 Tcl集团股份有限公司 A kind of heart rate detection method and device
CN107211165A (en) * 2015-01-09 2017-09-26 艾斯适配有限公司 Devices, systems, and methods for automatically delaying video display
CN107330945A (en) * 2017-07-05 2017-11-07 合肥工业大学 A kind of examing heartbeat fastly method based on video
CN107392177A (en) * 2017-08-05 2017-11-24 江西中医药大学 A kind of human body identification verification method and its device
US10076250B2 (en) * 2015-06-14 2018-09-18 Facense Ltd. Detecting physiological responses based on multispectral data from head-mounted cameras
CN108883294A (en) * 2015-10-15 2018-11-23 医科达有限公司 System and method for monitoring structure motion in entire radiotherapy
CN108932783A (en) * 2018-09-19 2018-12-04 南京邮电大学 A kind of access control system towards big flow scene based on two-dimension human face identification
CN109480808A (en) * 2018-09-27 2019-03-19 深圳市君利信达科技有限公司 A kind of heart rate detection method based on PPG, system, equipment and storage medium

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103826532A (en) * 2011-08-22 2014-05-28 Isis创新有限公司 Remote monitoring of vital signs
WO2014064575A2 (en) * 2012-10-23 2014-05-01 Koninklijke Philips N.V. Device and method for obtaining vital sign information of a living being
CN107211165A (en) * 2015-01-09 2017-09-26 艾斯适配有限公司 Devices, systems, and methods for automatically delaying video display
US10076250B2 (en) * 2015-06-14 2018-09-18 Facense Ltd. Detecting physiological responses based on multispectral data from head-mounted cameras
CN108883294A (en) * 2015-10-15 2018-11-23 医科达有限公司 System and method for monitoring structure motion in entire radiotherapy
CN105989357A (en) * 2016-01-18 2016-10-05 合肥工业大学 Human face video processing-based heart rate detection method
CN106491114A (en) * 2016-10-25 2017-03-15 Tcl集团股份有限公司 A kind of heart rate detection method and device
CN107330945A (en) * 2017-07-05 2017-11-07 合肥工业大学 A kind of examing heartbeat fastly method based on video
CN107392177A (en) * 2017-08-05 2017-11-24 江西中医药大学 A kind of human body identification verification method and its device
CN108932783A (en) * 2018-09-19 2018-12-04 南京邮电大学 A kind of access control system towards big flow scene based on two-dimension human face identification
CN109480808A (en) * 2018-09-27 2019-03-19 深圳市君利信达科技有限公司 A kind of heart rate detection method based on PPG, system, equipment and storage medium

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
王盛波: "基于视频实时持续心率检测及可视化", 《中国优秀硕士学位论文全文数据库(医药卫生科技辑)》 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113397519A (en) * 2021-08-05 2021-09-17 季华实验室 Cardiovascular health state detection device

Similar Documents

Publication Publication Date Title
Li et al. The obf database: A large face video database for remote physiological signal measurement and atrial fibrillation detection
US10004410B2 (en) System and methods for measuring physiological parameters
Wang et al. A comparative survey of methods for remote heart rate detection from frontal face videos
Fan et al. Robust blood pressure estimation using an RGB camera
CN106580301B (en) A kind of monitoring method of physiological parameter, device and handheld device
CN110384491A (en) A kind of heart rate detection method based on common camera
CN106073729B (en) The acquisition method of photoplethysmographic signal
CN104545864B (en) Psychological regulation method and apparatus
US20130096439A1 (en) Method and system for contact-free heart rate measurement
CN105989357A (en) Human face video processing-based heart rate detection method
KR101738278B1 (en) Emotion recognition method based on image
Gudi et al. Efficient real-time camera based estimation of heart rate and its variability
Qiao et al. Measuring heart rate and heart rate variability with smartphone camera
CN106793962A (en) Method and apparatus for continuously estimating human blood-pressure using video image
CN107330945A (en) A kind of examing heartbeat fastly method based on video
CN109497977A (en) Human heart rate and method for detecting blood oxygen saturation and device
WO2021164350A1 (en) Method and device for generating photoplethysmography signal
CN111938622B (en) Heart rate detection method, device and system and readable storage medium
WO2019173283A1 (en) Method and apparatus for non-invasive hemoglobin level prediction
Malasinghe et al. A comparative study of common steps in video-based remote heart rate detection methods
CN109770885A (en) A kind of examing heartbeat fastly method based on preview frame
WO2021111436A1 (en) System and method for physiological measurements from optical data
CN112294282A (en) Self-calibration method of emotion detection device based on RPPG
US11324406B1 (en) Contactless photoplethysmography for physiological parameter measurement
Zhuang et al. Remote blood pressure measurement via spatiotemporal mapping of a short-time facial video

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20190521

WD01 Invention patent application deemed withdrawn after publication