CN109820499A - The anti-interference heart rate detection method of height, electronic equipment and storage medium based on video - Google Patents
The anti-interference heart rate detection method of height, electronic equipment and storage medium based on video Download PDFInfo
- Publication number
- CN109820499A CN109820499A CN201811583917.3A CN201811583917A CN109820499A CN 109820499 A CN109820499 A CN 109820499A CN 201811583917 A CN201811583917 A CN 201811583917A CN 109820499 A CN109820499 A CN 109820499A
- Authority
- CN
- China
- Prior art keywords
- heart rate
- signal
- original signal
- treated
- video
- 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.)
- Granted
Links
Landscapes
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
The invention discloses the anti-interference heart rate detection method of height based on video, include the following steps: to acquire illuminance, and enter corresponding light pattern according to the illuminance；Video data is acquired in real time under light pattern, and face is tracked in video data and extracts face's video information；Face's video information is divided frame by frame, extracts the original signal of each pixel in the image of effective coverage；Average value processing is carried out to original signal, original signal stores by treated；To treated, original signal analyzed obtains heart rate signal, and carries out the noise Processing for removing heart rate signal that obtains that treated to heart rate signal；By treated, heart rate signal carries out being converted to frequency-region signal, and calculates the signal-to-noise ratio of frequency-region signal.The present invention integrates existing algorithm under conditions of enough illumination, and can detect heart rate in low illumination and dark surrounds, overcomes the prior art to can not be to the defect that heart rate is detected under brightness environment.
Description
Technical field
The present invention relates to a kind of heart rate detection technical field more particularly to a kind of anti-interference heart rate detections of height based on video
Method, electronic equipment and storage medium.
Background technique
Currently, remotely the main method of monitoring cardiovascular activity is exactly long-range body of light inspection variation scanning figure method (rPPG).
The photosensitive sensor (RGB) that rPPG passes through conventional multi-wavelength channel detects the small color in application on human skin surface as caused by beat pulse
Coloured silk variation, to achieve the purpose that long-range contactless monitoring cardiovascular activity.Technology based on core rPPG can be used for detecting people
Body-centered rate, respiratory rate, the parameter of the cardiovascular activities such as the oxygen content of blood (SpO2) and blood pressure, and the mental status of detection people.This
Kind of remote detecting method technology possesses that hardware device is small and exquisite cheap, and detection process is simple and fast, no infringement contactless to human body, together
When also possess enough accuracy features.Its inconvenience for compensating for conventional contact detection method, has expanded application scenarios, together
When have economic serviceability.There are great application prospect and economic value in artificial intelligence machine vision and medical field.
The core of rPPG is the algorithm for extracting pulse (heart rate) information.It is suggested in recent years there are many algorithm.Include:
Blind source analytic approach (BBS), typically there is independent component analysis method ICA and Principal Component Analysis PCA.It is exactly RGB
Signal intensity track is separated into different independent signal sources, take wherein the maximum signal source of periodic intensity as pulse signal.
This method is done well under the conditions of relative quiescent, but lacks the robustness changed to movement, illumination condition.Such as in difference
Head movement under illuminance, accuracy of measurement obviously glides when hurriedly breathing after strenuous exercise.
Color difference analysis method (Chrome) is changed by the coloration of analysis image information, to extract beat pulse signal.With
Blind source analytic approach is different, and Chrome is the analysis method based on model, is not analyzed under " blind " scene, to light source,
Reflection, skin and sensor have mathematical description and estimation error.So in exercise motion and during postexercise recovery
Color difference analysis is substantially better than blind source analytic approach, is also an advantage over BBS's on overall performance.
Pulse blood volume analytic approach (PBV) passes through the mutation analysis of blood volume by calculating the blood volume in graphical information
Pulse signal.It can be distinguished as this method and be changed as caused by pulse with blood volume caused by the movement of head, so can filter
Some noises as caused by moving.PBV is also based on the algorithm of model, to being to be better than Chrome on the robust performance of movement
With BBS's.But overall performance is each has something to recommend him at last with Chrome.
Subspace Rotation analytic approach (2SR) is a kind of analysis method based on photograph pixel data, by effective to face
Pixel carries out spatial decomposition, extracts the pulse signal included in pixel space.This method focuses on each pixel
Rgb color vector, rather than the average value of all pixels point.Overall performance is better than above-mentioned all methods, but to video quality demands
It is relatively high, and its principle determines that it will be unable to processing single channel camera video.
But currently existing scheme has the following deficiencies:
But there are many technical problem needs to overcome in the complicated light environment of practical application for rPPG technology.Such as people
The colour of skin, illumination condition in movement and postexercise recovery process etc., can all cause from each convenient noise, interference is effective
Signal is to reduce accuracy of measurement.The algorithm above is all based on visible light section multichannel RGB camera, and both for
Distorted signals caused by moving carries out a degree of improvement.But to illumination condition, such as in low illumination, monochromatic source or black
Secretly equal extreme environments are almost blank, however this scene is much in practice.As it can be seen that rPPG detection technique is to illumination ring
The raising of condition robustness is necessary.
Summary of the invention
For overcome the deficiencies in the prior art, one of the objects of the present invention is to provide high anti-interference heart rate detection method,
It can solve the technical issues of prior art cannot carry out heart rate detection to dark surrounds.
The second object of the present invention is to provide a kind of electronic equipment, can solve the prior art cannot to dark surrounds into
The technical issues of row heart rate detection.
The third object of the present invention is to provide a kind of computer readable storage medium, and can solve the prior art cannot be right
Dark surrounds carries out the technical issues of heart rate detection.
An object of the present invention adopts the following technical scheme that realization:
The anti-interference heart rate detection method of height based on video, which comprises the steps of:
Illumination determination step: acquisition illuminance, and corresponding light pattern is entered according to the illuminance；
Video acquisition step: acquiring video data under corresponding light pattern in real time, in video data to face into
Row is tracked and extracts face's video information；
Face's segmentation step: dividing face's video image frame by frame, extracts each pixel in the image of effective coverage
The original signal of point；
Average value processing step: average value processing is carried out to the original signal, original signal stores by treated；
Analysis filters out step: to treated, original signal is analyzed to obtain heart rate signal, and to the heart rate signal
Carry out the noise Processing for removing heart rate signal that obtains that treated；
Frequency domain switch process: by treated, heart rate signal carries out being converted to frequency-region signal, and calculates the letter of frequency-region signal
It makes an uproar ratio.
Further, the light pattern includes bright mode and brightness mode.
Further, it in video acquisition step, when light pattern is brightness mode, opens infrared light supply and adopts in real time again
Collect frequency evidence.
Further, when light pattern is bright mode, the original signal in face's segmentation step is that RGB is original
Signal carries out at mean value the original signal in three channels in the RGB original signal in average value processing step respectively
Reason；When light pattern is brightness mode, the original signal in face's segmentation step is infrared original signal, at mean value
It manages in step, average value processing is carried out to the infrared original signal.
Further, when the light pattern is the bright mode, by the RGB of the triple channel after the average value processing
Original signal analysis processing respectively obtain corresponding a variety of heart rate signals, in the frequency domain switch process, respectively to processing after
A variety of heart rate signals be converted to corresponding frequency-region signal, and calculate the signal-to-noise ratio of corresponding frequency-region signal.
Further, the analysis under bright mode filters out in step, using Space Rotating method, pulse blood volumetric method and color
Poor analytic approach to treated, analyzed obtains three treated heart rate signal by original signal.
Further, the analysis under brightness mode filters out in step, specifically includes following sub-step:
It obtains sub-step: choosing storage treated in original signal the infrared original signal in preset duration recently,
Defining the infrared original signal is f (t)=Sir (t), t ∈ [0,30s]；
Initial subslep: intrinsic signals, the centre frequency of intrinsic signals, intrinsic signals to the infrared original signal
Lagrange factor initialized, and determine constant value: the number of iterations upper limit N, the intrinsic signals predetermined number resolved into
K, limits parameter alpha and iteration step value τ and value ε is defined in convergence；
Operation sub-step: being iterated operation by formula one to three, and as n >=N or When stop interative computation；Formula one:It is public
Formula two:Formula three:
WhereinFor the intrinsic signals of infrared original signal,For the centre frequency of intrinsic signals,It is bright for the glug of intrinsic signals
Day factor；For k-th of intrinsic signals estimated value of nth iteration,For k-th of intrinsic signals of nth iteration
Center frequency estimation value,For the Lagrange factor estimated value of nth iteration；
Decompose sub-step: decomposition obtains K intrinsic signals, removes trend intrinsic signals, passes through remaining K-1 intrinsic signals
Obtain initial heart rate signal；
It filters out sub-step: noise being carried out by Butterworth bandpass filter to initial heart rate and eliminates to obtain treated the heart
Rate signal.
The second object of the present invention adopts the following technical scheme that realization:
A kind of electronic equipment including memory, processor and stores the meter that can be run on a memory and in processor
Calculation machine program, the processor perform the steps of illumination determination step when executing the computer program: acquisition illuminance,
And corresponding light pattern is entered according to the illuminance；
Video acquisition step: acquiring video data under corresponding light pattern in real time, in video data to face into
Row is tracked and extracts face's video information；
Face's segmentation step: dividing face's video image frame by frame, extracts each in effective coverage (ROI) image
The original signal of pixel；
Average value processing step: average value processing is carried out to the original signal, original signal stores by treated；
Analysis filters out step: to treated, original signal is analyzed to obtain heart rate signal, and to the heart rate signal
Carry out the noise Processing for removing heart rate signal that obtains that treated；
Frequency domain switch process: by treated, heart rate signal carries out being converted to frequency-region signal, and calculates the letter of frequency-region signal
It makes an uproar ratio.
Further, the light pattern includes bright mode and brightness mode.
The third object of the present invention adopts the following technical scheme that realization:
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
Method described in the above-mentioned any one of row.
Compared with prior art, the beneficial effects of the present invention are
The present invention integrates existing algorithm under conditions of enough illumination, is learnt from other's strong points to offset one's weaknesses by judging, is improved comprehensive
Close anti-interference ability；And heart rate can be detected in low illumination and dark surrounds based on single channel sensor, it overcomes existing
Technology is to can not be to the defect that heart rate is detected under brightness environment.
Detailed description of the invention
Fig. 1 is the flow chart of the height anti-interference heart rate detection method of the invention based on video.
Specific embodiment
In the following, being described further in conjunction with attached drawing and specific embodiment to the present invention, it should be noted that not
Under the premise of conflicting, new implementation can be formed between various embodiments described below or between each technical characteristic in any combination
Example.
Embodiment one
As shown in Figure 1, the present invention provides the anti-interference heart rate detection method of height based on video, this method passes through following hard
Part structure is realized: a general network camera (RGB), and preferred pixel is not less than 300,000, and frame per second is not less than 20fps.One close
Infrared camera, pixel are not less than 1,000,000, and frame per second is not less than 20fps.One LED near-infrared light source, spectrum in 960nm or so,
The illuminance of spherical surface is not less than 200Lux at one meter of radius.One computer, memory are not less than 8G, and CUP dominant frequency is not less than 2.8GHz.
RGB camera and near-infrared camera are connect by USB3.0 with computer.Tester regards at 0.5-1.2 rice before camera
Frequency typing.Method includes the following steps:
S1: acquisition illuminance, and corresponding light pattern is entered according to the illuminance；
This step determines illuminance judge whether illumination is enough thus enters corresponding light by RGB camera
Ray mode.Light pattern includes brightness and bright mode.When light pattern is brightness mode, opens infrared light supply and adopt in real time again
Collect frequency evidence.Illuminance then enters brightness mode lower than 100Lux, opens infrared light supply, and infrared camera starts to enroll video.
Enter bright mode higher than 100Lux, close infrared equipment, RGB camera starts to enroll video.According to experimental result, work as illumination
When degree is lower than 100Lux (the facial illuminance from 1 meter of camera or so place), heart rate measurement accuracy sharply declines.It activates at this time
Infrared equipment, naked eyes are almost invisible when due near infrared light, will not generate change to visual field environment, therefore will not influence people
The activity of progress.Infrared video recording is carried out using infrared camera, so that heart rate analysis can also be carried out in dark surrounds.Note
Rgb video signal is Sr_org (t, x, y), Sg_org (t, x, y), Sb_org (t, x, y), Infrared Video Signal Sir_org
(t,x,y).Wherein t is the time, and x and y represent each pixel coordinate of each frame video pictures.
S2: acquiring video data under corresponding light pattern in real time, and face is tracked and is mentioned in video data
Take face's video information；
The present invention carries out face extraction to video with tracking features method and tracks, and tracking features method is the prior art.
S3: dividing face's video image frame by frame, extracts the original of each pixel in effective coverage (ROI) image
Beginning signal；
This step extracts original signal Sr (t, x0, y0), Sg (t, x0, y0), the Sb (t, x0, y0) of effective coverage (ROI)
Or Sir (t, x0, y0).It is also pulse original signal, wherein x0 and y0 represents useful area pixel point coordinate.Honorable portion only has
Baring skin is effective area, eyes, mouth, eyebrow, and fringe etc. will will remove, only remaining effective coverage (ROI) area.Pass through
Complete bashful grid is divided, according to the gray value difference of pixel point group in lattice, removes invalid grid pixel point group, specifically
After doing gray proces to face, hair, eyes, the grid pixel group point gray value of the colors such as eyebrow dark place will remote low effective skin,
The default grid value gray value that be averaged be inactive area (0 be black 255 be white) less than 50.
S4: average value processing is carried out to the original signal, original signal stores by treated；
When light pattern is bright mode, the original signal in face's segmentation step is RGB original signal, equal
Be worth in processing step, frame by frame to the original signal Sr (t, x0, y0) in three channels at the RGB original signal end, Sg (t, x0,
Y0), Sb (t, x0, y0) carries out average value processing respectively；The original when light pattern is brightness mode, in face's segmentation step
Beginning signal is infrared original signal, in average value processing step, is carried out frame by frame to the infrared original signal Sir (t, x0, y0)
Average value processing.Inside message buffer, buffer can deposit 30s to be believed each group information data storage handled well for 900 frames
Data are ceased, the data information of update in latter every three seconds is filled with.
S5: to treated, original signal is analyzed to obtain heart rate signal, and carries out noise to the heart rate signal and disappear
Except the processing heart rate signal that obtains that treated；
Analysis under bright mode filters out in step, using Space Rotating method, pulse blood volumetric method and color difference analysis method
To treated, original signal analyzed obtains three treated heart rate signal.Going bail for, there are newest in signal buffer
RGB multi channel signals Sr (t) in 30s, Sg (t) and Sb (t), using Subspace Rotation method (2SR), pulse blood volumetric method
(PVB) and color difference analysis method (Chrom) analyzes heart rate signal S1_raw (t), S2_raw (t) and S3_raw (t).It will be filtered
Signal S1_raw (t), S2_raw (t) and S3_raw (t) the Butterworth bandpass filter of wave processing, carry out noise elimination,
Obtain heart rate signal S1 (t), S2 (t) and S3 (t).It is right under the frame per second of 30fps since human heart rate's range is 50-120bpm
Frequency 0.75-2Hz is answered, so removing the noise of signal using Butterworth bandpass filtering.
Under brightness mode, S5 step is specifically:
The infrared original signal in preset duration recently is chosen in storage treated original signal, is defined described infrared
Original signal is f (t)=Sir (t), t ∈ [0,30s]；Preset duration is actually to choose in buffer in newest 30s
Effective infrared original signal Sir (t), and variant patterns decomposition and separation is carried out by following step and goes out heart rate signal S_raw
(t)。
To the intrinsic signals of the infrared original signal, the centre frequency of intrinsic signals, intrinsic signals it is Lagrangian because
Son is initialized, and determines constant value: the number of iterations upper limit N, the intrinsic signals predetermined number K resolved into, limits parameter alpha,
And value ε is defined in iteration step value τ and convergence；
Operation sub-step: being iterated operation by formula one to three, and as n >=N or When stop interative computation；Formula one:It is public
Formula two:Formula three:
WhereinFor the intrinsic signals of infrared original signal,For the centre frequency of intrinsic signals,It is bright for the glug of intrinsic signals
Day factor,For k-th of intrinsic signals estimated value of nth iteration,For k-th of intrinsic signals of nth iteration
Center frequency estimation value,For the Lagrange factor estimated value of nth iteration.
Decompose sub-step: decomposition obtains K intrinsic signals { u1(ω),u2(ω),…,uk(ω) }, remove this reference of trend
Number u1(ω) obtains initial heart rate signal
It filters out sub-step: noise being carried out by Butterworth bandpass filter to initial heart rate and eliminates to obtain treated the heart
Rate signal.By without the signal S_raw (t) of filtering processing Butterworth bandpass filter, noise elimination is carried out, the heart is obtained
Rate signal S (t).Since human heart rate's range is 50-120bpm, the respective frequencies 0.75-2Hz under the frame per second of 30fps, so adopting
With Butterworth bandpass filtering, the noise of signal is removed.
S6: by treated, heart rate signal carries out being converted to frequency-region signal, and calculates the signal-to-noise ratio of frequency-region signal.
Under brightness mode, by signal S (t) translation bit frequency-region signal S (f).It is average heart rate HB hertz at frequency spectrum maximum value,
It is converted into heartbeat per minute: HB*60bmp.Under bright mode, by signal S1 (t), S2 (t) and S3 (t) translation bit frequency-region signal S1
(f), S2 (f) and S3 (f).It is average heart rate frequency HB1, HB2 and HB3 at frequency spectrum maximum value.It is converted into heartbeat per minute: HB*
60bmp.To different gained signal spectrum S1 (f), S2 (f) and S3 (f) calculate separately Signal to Noise Ratio (SNR), take SNR maximum value corresponding
Signal be final heart rate signal S (t) and heart rate frequency HB.
In bright mode, after rgb signal pretreatment, with Subspace Rotation method, chromatism method and the analysis of pulse blood volumetric method
Method carries out heart rate analysis parallel, is then judged in evaluation system, and the result to a kind of method is all in acquisition time 30s
Interior heart rate signal does spectrum analysis to each consequential signal, the judgment criteria according to the signal-to-noise ratio of heart rate frequency spectrum as foundation.
Fourier transform first is carried out to signal, obtains corresponding frequency spectrum.It is heart rate at spectral frequencies maximum value, first calculates heart rate in 0.75-
Irrelevance in 2Hz critical field takes and deviates reasonable method.If then being carried out to distinct methods all in reasonable irrelevance
Signal-to-noise ratio compares, the HR values of optimum selecting SNR big method.If SNR is not much different, heart rate deviation is more than a certain range
Then it is averaged.The resistance to extraneous various interference is substantially increased in this way, improves heart rate accuracy rate.The present invention can choose
2SR analytic approach.
In brightness mode, after IR signal and processing, it is decomposed with variant patterns decomposition method, resolves into different narrow
The sub- intrinsic signals of spectral domain.To resolve into that how many subsignal can just isolate heart rate signal information since VMD not can solve, I
Using circulation increase resolution model quantity method (empirically beginning default 4), find optimal mode decompose number.One IR letter
Number 4 subsignals are broken down into, spectrogram is marked with different colours.The signal of purple curves has maximum value in log-spectral domain,
And signal-to-noise ratio is also met the requirements, so being heart rate signal, stopping continues to decompose.Under static state, error rate is less than 2%.
Embodiment two
Embodiment two discloses a kind of electronic equipment, including memory, processor and storage are on a memory and can be
The computer program of processor operation, the processor perform the steps of illumination and determine step when executing the computer program
It is rapid: acquisition illuminance, and corresponding light pattern is entered according to the illuminance；
Video acquisition step: acquiring video data under corresponding light pattern in real time, in video data to face into
Row is tracked and extracts face's video information；
Face's segmentation step: dividing face's video image frame by frame, extracts each pixel in the image of effective coverage
The original signal of point；
Average value processing step: average value processing is carried out to the original signal, original signal stores by treated；
Analysis filters out step: to treated, original signal is analyzed to obtain heart rate signal, and to the heart rate signal
Carry out the noise Processing for removing heart rate signal that obtains that treated；
Frequency domain switch process: by treated, heart rate signal carries out being converted to frequency-region signal, and calculates the letter of frequency-region signal
It makes an uproar ratio.
The anti-interference heart rate detection method of height disclosed in the specific implementation principle and the present invention of above-mentioned steps based on video
The principle of detailed description is completely the same, and details are not described herein.
Embodiment three
Embodiment three discloses a kind of readable computer storage medium, which is somebody's turn to do for storing program
When program is executed by processor, the anti-interference heart rate detection method of height based on video of embodiment one is realized.
The above embodiment is only the preferred embodiment of the present invention, and the scope of protection of the present invention is not limited thereto,
The variation and replacement for any unsubstantiality that those skilled in the art is done on the basis of the present invention belong to institute of the present invention
Claimed range.
Claims (10)
1. the anti-interference heart rate detection method of height based on video, which comprises the steps of:
Illumination determination step: acquisition illuminance, and corresponding light pattern is entered according to the illuminance；
Video acquisition step: video data is acquired in real time under the corresponding light pattern, to people in the video data
Face is tracked and extracts face's video information；
Face's segmentation step: dividing face's video image frame by frame, extracts each pixel in the image of effective coverage
The original signal of point；
Average value processing step: average value processing is carried out to the original signal, the original signal stores by treated；
Analysis filters out step: to treated, the original signal is analyzed to obtain heart rate signal, and to the heart rate signal
Carry out the noise Processing for removing heart rate signal that obtains that treated；
Frequency domain switch process: by treated, the heart rate signal carries out being converted to frequency-region signal, and calculates the frequency-region signal
Signal-to-noise ratio.
2. high anti-interference heart rate detection method as described in claim 1, which is characterized in that the light pattern includes bright mould
Formula and brightness mode.
3. high anti-interference heart rate detection method as claimed in claim 2, which is characterized in that in the video acquisition step,
When the light pattern is the brightness mode, opens infrared light supply and acquire the video data in real time again.
4. high anti-interference heart rate detection method as claimed in claim 3, which is characterized in that when the light pattern is the light
When bright mode, the original signal in face's segmentation step is RGB original signal, in the average value processing step,
Average value processing is carried out respectively to the original signal in three channels in the RGB original signal；When the light pattern is described
When brightness mode, the original signal in face's segmentation step is infrared original signal, in the average value processing step
In, average value processing is carried out to the infrared original signal.
5. high anti-interference heart rate detection method as claimed in claim 4, which is characterized in that when the light pattern is the light
When bright mode, the RGB original signal analysis processing of the triple channel after the average value processing is respectively obtained into corresponding a variety of hearts rate
Signal, in the frequency domain switch process, to treated, a variety of heart rate signals are converted to corresponding frequency-region signal respectively,
And calculate the signal-to-noise ratio of corresponding frequency-region signal.
6. high anti-interference heart rate detection method as claimed in claim 5, which is characterized in that described under the bright mode
Analysis filters out in step, and using Space Rotating method, pulse blood volumetric method and color difference analysis method, to treated, original signal is carried out
Analysis obtains three treated heart rate signal.
7. high anti-interference heart rate detection method as claimed in claim 4, which is characterized in that described under the brightness mode
Analysis filters out in step, specifically includes following sub-step:
It obtains sub-step: choosing storage treated in original signal the infrared original signal in preset duration recently, definition
The infrared original signal is f (t)=Sir (t)；
Initial subslep: the drawing to the intrinsic signals, the centre frequency of intrinsic signals, intrinsic signals of the infrared original signal
The Ge Lang factor is initialized, and determines constant value: the number of iterations upper limit N, the intrinsic signals predetermined number K resolved into, limit
Value ε is defined in parameter alpha and iteration step value τ processed and convergence；
Operation sub-step: being iterated operation by formula one to three, and as n >=N or When stop interative computation；Formula one:It is public
Formula two:Formula three:
WhereinFor the intrinsic signals of infrared original signal,For the centre frequency of intrinsic signals,It is bright for the glug of intrinsic signals
Day factor；For k-th of intrinsic signals estimated value of nth iteration,For k-th of intrinsic signals of nth iteration
Center frequency estimation value,For the Lagrange factor estimated value of nth iteration；
Decompose sub-step: decomposition obtains K intrinsic signals, removes trend intrinsic signals, is obtained by remaining K-1 intrinsic signals
Initial heart rate signal；
It filters out sub-step: noise being carried out by Butterworth bandpass filter to the initial heart rate and eliminates to obtain treated the heart
Rate signal.
8. a kind of electronic equipment including memory, processor and stores the calculating that can be run on a memory and in processor
Machine program, which is characterized in that the processor performs the steps of when executing the computer program
Illumination determination step: acquisition illuminance, and corresponding light pattern is entered according to the illuminance；
Video acquisition step: video data is acquired in real time under corresponding institute's light pattern, to face in the video data
It is tracked and extracts face's video information；
Face's segmentation step: dividing face's video image frame by frame, extracts each pixel in the image of effective coverage
The original signal of point；
Average value processing step: average value processing is carried out to the original signal, the original signal stores by treated；
Analysis filters out step: to treated, the original signal is analyzed to obtain heart rate signal, and to the heart rate signal
Carry out the noise Processing for removing heart rate signal that obtains that treated；
Frequency domain switch process: by treated, heart rate signal carries out being converted to frequency-region signal, and calculates the letter of the frequency-region signal
It makes an uproar ratio.
9. electronic equipment as claimed in claim 8, which is characterized in that the light pattern includes bright mode and brightness mould
Formula.
10. a kind of computer readable storage medium, is stored thereon with computer program, it is characterised in that: the computer program
The height anti-interference heart rate detection method such as claim 1-7 any one is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811583917.3A CN109820499B (en) | 2018-12-24 | 2018-12-24 | High anti-interference heart rate detection method based on video, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811583917.3A CN109820499B (en) | 2018-12-24 | 2018-12-24 | High anti-interference heart rate detection method based on video, electronic equipment and storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109820499A true CN109820499A (en) | 2019-05-31 |
| CN109820499B CN109820499B (en) | 2022-10-28 |
Family
ID=66861048
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811583917.3A Active CN109820499B (en) | 2018-12-24 | 2018-12-24 | High anti-interference heart rate detection method based on video, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109820499B (en) |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111243739A (en) * | 2020-01-07 | 2020-06-05 | 四川大学 | Anti-interference physiological parameter telemetering method and system |
| CN111297339A (en) * | 2020-02-21 | 2020-06-19 | 乐普（北京）医疗器械股份有限公司 | Method and device for generating signal of photoplethysmography |
| CN111297347A (en) * | 2020-02-21 | 2020-06-19 | 乐普（北京）医疗器械股份有限公司 | Method and apparatus for generating photoplethysmography signals |
| CN111297344A (en) * | 2020-03-24 | 2020-06-19 | 陈恬慧 | Pulse-taking visual array acquisition method |
| CN111429345A (en) * | 2020-03-03 | 2020-07-17 | 贵阳像树岭科技有限公司 | Method for visually calculating heart rate and heart rate variability with ultra-low power consumption |
| CN112043254A (en) * | 2020-08-12 | 2020-12-08 | 厦门大学 | Prawn heart rate detection method and system based on video image |
| CN112669954A (en) * | 2020-04-22 | 2021-04-16 | 中国科学院心理研究所 | Remote heart rate detection system and method thereof |
| CN113712526A (en) * | 2021-09-30 | 2021-11-30 | 四川大学 | Pulse wave extraction method and device, electronic equipment and storage medium |
| WO2022076776A1 (en) * | 2020-10-08 | 2022-04-14 | Fresenius Medical Care Holdings, Inc. | Techniques for determining characteristics of dialysis access sites using image information |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150124067A1 (en) * | 2013-11-04 | 2015-05-07 | Xerox Corporation | Physiological measurement obtained from video images captured by a camera of a handheld device |
| CN106017926A (en) * | 2016-05-13 | 2016-10-12 | 山东理工大学 | Rolling bearing fault diagnosis method based on variational mode decomposition |
| US20160324432A1 (en) * | 2015-05-07 | 2016-11-10 | Whoop, Inc. | Heart rate detection using ambient light |
| CN107361762A (en) * | 2017-08-04 | 2017-11-21 | 山东理工大学 | ECG baseline drift bearing calibration based on variation mode decomposition |
| CN108056773A (en) * | 2017-12-11 | 2018-05-22 | 重庆邮电大学 | Based on the algorithms of QRS complexes detection in electrocardiogram signal for improving variation mode decomposition |
| CN108135487A (en) * | 2015-10-08 | 2018-06-08 | 皇家飞利浦有限公司 | For obtaining the equipment, system and method for the vital sign information of object |
| CN108272448A (en) * | 2018-03-29 | 2018-07-13 | 合肥工业大学 | A kind of contactless baby's physiological parameter monitoring method round the clock |
| CN108614259A (en) * | 2018-05-02 | 2018-10-02 | 电子科技大学 | A kind of heartbeat respiratory characteristic monitoring method based on ultra-wideband radar sensors |
| CN108888249A (en) * | 2018-06-07 | 2018-11-27 | 北京邮电大学 | A kind of method and device of the more people's vital sign monitorings of contactless car |
-
2018
- 2018-12-24 CN CN201811583917.3A patent/CN109820499B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20150124067A1 (en) * | 2013-11-04 | 2015-05-07 | Xerox Corporation | Physiological measurement obtained from video images captured by a camera of a handheld device |
| US20160324432A1 (en) * | 2015-05-07 | 2016-11-10 | Whoop, Inc. | Heart rate detection using ambient light |
| CN108135487A (en) * | 2015-10-08 | 2018-06-08 | 皇家飞利浦有限公司 | For obtaining the equipment, system and method for the vital sign information of object |
| CN106017926A (en) * | 2016-05-13 | 2016-10-12 | 山东理工大学 | Rolling bearing fault diagnosis method based on variational mode decomposition |
| CN107361762A (en) * | 2017-08-04 | 2017-11-21 | 山东理工大学 | ECG baseline drift bearing calibration based on variation mode decomposition |
| CN108056773A (en) * | 2017-12-11 | 2018-05-22 | 重庆邮电大学 | Based on the algorithms of QRS complexes detection in electrocardiogram signal for improving variation mode decomposition |
| CN108272448A (en) * | 2018-03-29 | 2018-07-13 | 合肥工业大学 | A kind of contactless baby's physiological parameter monitoring method round the clock |
| CN108614259A (en) * | 2018-05-02 | 2018-10-02 | 电子科技大学 | A kind of heartbeat respiratory characteristic monitoring method based on ultra-wideband radar sensors |
| CN108888249A (en) * | 2018-06-07 | 2018-11-27 | 北京邮电大学 | A kind of method and device of the more people's vital sign monitorings of contactless car |
Non-Patent Citations (2)
| Title |
|---|
| WENJIN WANG等: "《A Novel Algorithm for Remote Photoplethysmography- Spatial Subspace Rotation》", 《TRANSACTIONS ON BIOMEDICAL ENGINEERING》 * |
| 雷莹: "《基于变分模态分解的神经网络心电信号预测方法研究》", 《医药卫生科技辑》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111243739A (en) * | 2020-01-07 | 2020-06-05 | 四川大学 | Anti-interference physiological parameter telemetering method and system |
| CN111297339A (en) * | 2020-02-21 | 2020-06-19 | 乐普（北京）医疗器械股份有限公司 | Method and device for generating signal of photoplethysmography |
| CN111297347A (en) * | 2020-02-21 | 2020-06-19 | 乐普（北京）医疗器械股份有限公司 | Method and apparatus for generating photoplethysmography signals |
| WO2021164350A1 (en) * | 2020-02-21 | 2021-08-26 | 乐普（北京）医疗器械股份有限公司 | Method and device for generating photoplethysmography signal |
| CN111297347B (en) * | 2020-02-21 | 2022-07-29 | 乐普（北京）医疗器械股份有限公司 | Method and apparatus for generating photoplethysmography signals |
| CN111297339B (en) * | 2020-02-21 | 2022-07-29 | 乐普（北京）医疗器械股份有限公司 | Method and device for generating signal of photoplethysmography |
| CN111429345A (en) * | 2020-03-03 | 2020-07-17 | 贵阳像树岭科技有限公司 | Method for visually calculating heart rate and heart rate variability with ultra-low power consumption |
| CN111297344A (en) * | 2020-03-24 | 2020-06-19 | 陈恬慧 | Pulse-taking visual array acquisition method |
| CN112669954A (en) * | 2020-04-22 | 2021-04-16 | 中国科学院心理研究所 | Remote heart rate detection system and method thereof |
| CN112043254A (en) * | 2020-08-12 | 2020-12-08 | 厦门大学 | Prawn heart rate detection method and system based on video image |
| WO2022076776A1 (en) * | 2020-10-08 | 2022-04-14 | Fresenius Medical Care Holdings, Inc. | Techniques for determining characteristics of dialysis access sites using image information |
| CN113712526A (en) * | 2021-09-30 | 2021-11-30 | 四川大学 | Pulse wave extraction method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109820499B (en) | 2022-10-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109820499A (en) | The anti-interference heart rate detection method of height, electronic equipment and storage medium based on video | |
| CN106778695B (en) | Multi-person rapid heart rate detection method based on video | |
| US11229372B2 (en) | Systems and methods for computer monitoring of remote photoplethysmography based on chromaticity in a converted color space | |
| US8805019B2 (en) | Processing images of at least one living being | |
| CN105451646B (en) | Device, system and method for extracting physiological information | |
| EP3664704B1 (en) | Device, system and method for determining a physiological parameter of a subject | |
| EP3440991A1 (en) | Device, system and method for determining a physiological parameter of a subject | |
| Subramaniam et al. | Estimation of the Cardiac Pulse from Facial Video in Realistic Conditions. | |
| Chen et al. | RealSense= real heart rate: Illumination invariant heart rate estimation from videos | |
| Tang et al. | Non-contact heart rate monitoring by combining convolutional neural network skin detection and remote photoplethysmography via a low-cost camera | |
| Bobbia et al. | Remote photoplethysmography based on implicit living skin tissue segmentation | |
| CN105701806B (en) | Parkinson's tremor motion feature detection method based on depth image and system | |
| US20200311388A1 (en) | Human body physiological parameter monitoring method based on face recognition for workstation | |
| CN111281367A (en) | Anti-interference non-contact heart rate detection method based on face video | |
| CN111050638B (en) | Computer-implemented method and system for contact photoplethysmography (PPG) | |
| US11701015B2 (en) | Computer-implemented method and system for direct photoplethysmography (PPG) with multiple sensors | |
| CN113361480B (en) | Human body pulse wave acquisition method based on face video | |
| CN113693573B (en) | Video-based non-contact multi-physiological-parameter monitoring system and method | |
| CN110095109A (en) | Attitude detecting method based on hoistable platform | |
| CN116509359A (en) | Multi-target heart rate monitoring method and device for construction site operation | |
| Zarándy et al. | Multi-Level Optimization for Enabling Life Critical Visual Inspections of Infants in Resource Limited Environment | |
| CN117678998A (en) | Non-contact heart rate detection method based on self-adaptive projection plane and feature screening | |
| CN116994292A (en) | Non-contact heart rate extraction method and device | |
| CN117556195A (en) | Video heart rate measurement method based on time-frequency domain wiener filtering |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |