CN107334469A - Non-contact more people's method for measuring heart rate and device based on SVMs - Google Patents
Non-contact more people's method for measuring heart rate and device based on SVMs Download PDFInfo
- Publication number
- CN107334469A CN107334469A CN201710608816.6A CN201710608816A CN107334469A CN 107334469 A CN107334469 A CN 107334469A CN 201710608816 A CN201710608816 A CN 201710608816A CN 107334469 A CN107334469 A CN 107334469A
- Authority
- CN
- China
- Prior art keywords
- time
- domain signal
- heart rate
- signal
- interest
- 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
Links
- 238000000034 method Methods 0.000 title claims abstract description 47
- 238000012706 support-vector machine Methods 0.000 title claims abstract description 41
- 230000001815 facial effect Effects 0.000 claims abstract description 44
- 238000001514 detection method Methods 0.000 claims abstract description 25
- 210000001061 forehead Anatomy 0.000 claims abstract description 19
- 238000009532 heart rate measurement Methods 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 8
- 238000012360 testing method Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 6
- 108010076504 Protein Sorting Signals Proteins 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 claims description 2
- 210000003128 head Anatomy 0.000 claims description 2
- 230000000694 effects Effects 0.000 abstract description 6
- 238000005259 measurement Methods 0.000 abstract description 6
- 230000005622 photoelectricity Effects 0.000 description 5
- 230000010349 pulsation Effects 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 230000009022 nonlinear effect Effects 0.000 description 2
- 238000000513 principal component analysis Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 210000000887 face Anatomy 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
- A61B5/024—Detecting, measuring or recording pulse rate or heart rate
- A61B5/02416—Detecting, measuring or recording pulse rate or heart rate using photoplethysmograph signals, e.g. generated by infrared radiation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7203—Signal processing specially adapted for physiological signals or for diagnostic purposes for noise prevention, reduction or removal
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7225—Details of analog processing, e.g. isolation amplifier, gain or sensitivity adjustment, filtering, baseline or drift compensation
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
- A61B5/7235—Details of waveform analysis
- A61B5/7264—Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
Landscapes
- Health & Medical Sciences (AREA)
- Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Signal Processing (AREA)
- Biophysics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Veterinary Medicine (AREA)
- Public Health (AREA)
- Animal Behavior & Ethology (AREA)
- Biomedical Technology (AREA)
- Surgery (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Physiology (AREA)
- Pathology (AREA)
- Medical Informatics (AREA)
- Psychiatry (AREA)
- Theoretical Computer Science (AREA)
- Cardiology (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Biology (AREA)
- Power Engineering (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
The invention discloses a kind of non-contact more people's method for measuring heart rate and device based on SVMs.Methods described, including：Facial area-of-interest is detected by face detection algorithm, the facial area-of-interest includes forehead region and cheek region；The facial area-of-interest gray average is calculated, photoplethysmographic is obtained and traces time-domain signal；Time-domain signal is traced to the photoplethysmographic to pre-process, and obtains the first time-domain signal；First time-domain signal is classified using SVMs, obtains the second time-domain signal；Second time-domain signal is calculated, obtains heart rate value.The problem of overcoming light change and the not high measurement accuracy brought of motion artifact, realize the quick effect accurately measured of more people's hearts rate.
Description
Technical field
The present embodiments relate to measuring of human health and field of machine vision technology, and in particular to one kind based on support to
The non-contact more people's method for measuring heart rate and device of amount machine.
Background technology
Heart rate can fully reflect the physical condition of a people, be the important evidence that doctor is diagnosed to patient.
With the enhancing of people's self health monitoring consciousness and the popularization of intelligent terminal, retouched based on image photoplethysmographic
The contactless method for measuring heart rate of note (IPPG) technology is increasingly valued by people.But this measuring method is present
The problem of easily being disturbed by light change and motion artifact.2010, Massachusetts science and engineering pay understand philosophy and its team propose utilize
The method of principal component analysis (ICA) carries out the suppression of motion artifact, realizes the automatic measurement of heart rate signal.2014, Sydney
The research team of university proposes the prediction of the method KNN progress IPPG signals using principal component analysis combination machine learning, enters one
The precision for improving non-contact heart rate measurement of step.
But existing non-contact method for measuring heart rate does not consider that light change can cause Face datection algorithm easy
There is flase drop frame, and the problem of the IPPG signals by being obtained comprising human face region also contain substantial amounts of noise.Further, since the heart
Rate signal belongs to nonlinear properties, and KNN algorithms can not carry out preferable nonlinear regression prediction to nonlinear properties and easily by sample
The unbalanced influence of this capacity, amount of calculation are larger, it is impossible to realize the quick measurement of heart rate.
The content of the invention
The present invention provides a kind of non-contact more people's method for measuring heart rate and device based on SVMs, is overcome with realizing
The problem of measurement accuracy that light is disturbed and motion artifact is brought is not high, reaches raising detection speed, realizes more people's hearts rate
The quick purpose accurately measured.
In a first aspect, the embodiments of the invention provide a kind of non-contact more people heart rate measurement sides based on SVMs
Method, it is characterised in that including：
Facial area-of-interest is detected by face detection algorithm, the facial area-of-interest includes forehead region and face
Buccal region domain；
The facial area-of-interest gray average is calculated, photoplethysmographic is obtained and traces time-domain signal；
It is rich to the photoplethysmographic to trace time-domain signal and pre-process, obtain the first time-domain signal；
First time-domain signal is classified using SVMs, obtains the second time-domain signal；
Second time-domain signal is calculated, obtains heart rate value.
Preferably, it is described that facial area-of-interest is detected by face detection algorithm, including：Using multithreading.
Preferably, it is described facial area-of-interest is detected by face detection algorithm after, in addition to：Described in tracking
Facial area-of-interest；
Gray average is taken to facial area-of-interest described in every frame.
Preferably, the facial area-of-interest includes forehead region and cheek region, including：The forehead region height
The 20% of the facial area-of-interest is accounted for, the cheek region highly accounts for the 60% of the facial area-of-interest, the volume
Head region is identical with the facial area-of-interest with the width of the cheek region.
Preferably, it is described time-domain signal is traced to photoplethysmographic to pre-process, including：
Time domain is traced to the photoplethysmographic by the Wavelet-denoising Method of moving average filter method and variable thresholding
Signal carries out denoising.
Preferably, it is described that first time-domain signal is classified using SVMs, including：
Using the fractal characteristic of heart rate signal, the time-domain signal directly obtained to area-of-interest is classified, without entering
Onestep extraction signal characteristic value.
Preferably, it is described that first time-domain signal is classified using SVMs, including：
Training stage, face area-of-interest in first time-domain signal is believed with the time domain extracted by flase drop frame respectively
Number it is trained respectively as positive negative sample；
Test phase, tested using the SVMs trained, select second time-domain signal.
Preferably, the calculating heart rate signal, including：
Time frequency analysis is carried out to the second time-domain signal sequence using Short Time Fourier Transform method.
Second aspect, the embodiment of the present invention additionally provide a kind of non-contact more people's heart rate measurement dresses based on SVMs
Put, it is characterised in that including：
Detection module, for detecting facial area-of-interest, the facial area-of-interest bag by face detection algorithm
Include forehead region and cheek region；
Primary signal acquisition module, for calculating the facial area-of-interest gray average, obtain photoelectricity volume pulsation
Ripple traces time-domain signal；
Pretreatment module, pre-processed for tracing time-domain signal to the photoplethysmographic, when obtaining first
Domain signal；
Heart rate signal acquisition module, for classifying using SVMs to first time-domain signal, obtain the
Two time-domain signals；
Computing module, for calculating second time-domain signal, obtain heart rate value.
Preferably, it is described that first time-domain signal is classified using SVMs, including：
Training module, for respectively by first time-domain signal by the time-domain signal of face extraction with being carried by flase drop frame
The time-domain signal taken is trained respectively as positive negative sample；
Test module, tested with the SVMs trained, select the heart rate signal.
The present invention is gone forward side by side by face detection algorithm and track algorithm detection with tracking area-of-interest, extraction IPPG signals
Row pretreatment, is then classified using the fractal characteristic of heart rate by SVMs to signal, the letter to being categorized as heart rate
Number heart rate value is calculated, solve the problems, such as that measurement result easily by light and motion artifacts, realizes that the heart rate of more people is quickly accurate
The beneficial effect really measured.
Brief description of the drawings
Fig. 1 is a kind of flow of non-contact more people's method for measuring heart rate based on SVMs in the embodiment of the present invention one
Figure.
Fig. 2 is a kind of face of non-contact more people's method for measuring heart rate based on SVMs in the embodiment of the present invention one
Area-of-interest schematic diagram.
Fig. 3 is a kind of flow of non-contact more people's method for measuring heart rate based on SVMs in the embodiment of the present invention two
Figure.
Fig. 4 is a kind of structure of non-contact more people's heart rate measurement devices based on SVMs in the embodiment of the present invention three
Schematic diagram.
Fig. 5 is a kind of test of non-contact more people's heart rate measurement devices based on SVMs in the embodiment of the present invention three
Effect diagram.
Embodiment
The present invention is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining the present invention, rather than limitation of the invention.It also should be noted that in order to just
Part related to the present invention rather than entire infrastructure are illustrate only in description, accompanying drawing.
Embodiment one
Fig. 1 is the flow for non-contact more people's method for measuring heart rate based on SVMs that the embodiment of the present invention one provides
Figure, specifically comprises the following steps：
Step 100, facial area-of-interest is detected by face detection algorithm, the facial area-of-interest includes forehead
Region and cheek region；
Wherein, the detection of face is carried out using the Face datection algorithm integrated in opencv, and testing result is initialized
Compress track algorithm, the numbering of face and the parameter of detection block detected by algorithm return；
In this step, typically, as shown in Fig. 2 area-of-interest is located at the middle position in Face datection frame region, both
Magnitude relationship is：
Wherein,
hd, wdRepresent Face datection frame respectively is higher than width；hi, wiROI Gao Yukuan is represented respectively；
Further, ROI region is divided into forehead and cheek two parts, both width are identical with ROI, highly exist such as
Lower relation：
Wherein, h1Represent the height in forehead region, h2Represent the height of cheek region.
Step 200, the facial area-of-interest gray average is calculated, obtain photoplethysmographic and trace time domain letter
Number；
Wherein, using the method for compression tracking, the region that face detection algorithm detects is tracked, and by image face
The colour space is transformed into HSV by RGB, and the gray average of H passages, obtains IPPG signal sequences in calculating per frame grey image R OI regions
Row：
X (t)=0.3*x1(t)+0.7*x2(t) (3)
Wherein,
x1(t) forehead region H passage gray averages, x are represented2(t) cheek region H passage gray averages are represented,
Multithreading is realized using openMP in this step, Face tracking algorithm is accelerated, realizes the reality of multiple faces
When the tracking and acquisitions of IPPG signals.
Step 300, it is rich to the photoplethysmographic trace time-domain signal and pre-process, obtain the first time domain letter
Number；
Wherein, the random noise in primary signal is suppressed using the moving average filter that length is 10, then,
Further above-mentioned signal is carried out except processing of making an uproar using the Wavelet noise-eliminating method of variable thresholding；
In this step, typically, the heuristic threshold value of selection of threshold function (heursure)：It is optimum prediction variable threshold
Selection, is determined using unbiased possibility predication or hard -threshold by the signal to noise ratio of signal, selects to estimate without partial likelihood when noise is bigger
Meter method, conversely, selection hard threshold method；The calculation of signal to noise ratio is as follows：
Wherein, N represents the length of clock signal, yiRepresent primary signal, xiRepresent the estimation signal after processing, noise
It is better than bigger expression signal transacting effect.
Step 400, using SVMs first time-domain signal is classified, obtain the second time-domain signal；
Wherein, using the pretreated signal that length is 64, respectively using the signal comprising human face region as positive sample,
Not comprising human face region or only the signal comprising a part of human face region is used as negative sample, choose respectively 100 positive samples and
50 negative samples proceed by training；Then new signal is classified using the SVMs trained, selected just
True heart rate signal；
In this step, typically, the workflow diagram of example SVMs as shown in Figure 3, in the present embodiment parameter punish
Parameter g uses mesh parameter optimizing function Automatic Optimal in penalty factor c and kernel function, as can be seen from Figure, when taking c respectively
When=8, g=0.70711, classification accuracy rate is up to 80.5%.
Step 500, the heart rate signal is calculated, obtain heart rate value；
Wherein, the signal chosen is transformed into frequency domain using Short Time Fourier Transform, selection is in 0.5HZ-3HZ
Highest frequency value fHR；
In this step, typically, the calculation formula of heart rate value is as follows：
HR=fHR*60 (5)
Wherein, HR represents the number of heartbeat per minute.
The technical scheme of the present embodiment, by interested area division and multithreading is used, eye is overcome and blinks
The motion artifacts problem brought, has reached the effect quick and precisely measured.
It is described to detect that facial area-of-interest is preferred by face detection algorithm on the basis of above-mentioned each technical scheme
Multithreading can be used.Detect that facial area-of-interest so sets that be advantageous in that can be with by face detection algorithm
Realize the real-time acquisition of more people's heart rate signals.
On the basis of above-mentioned each technical scheme, the facial area-of-interest includes forehead region and cheek region, excellent
Choosing can set the forehead region height to account for the 20% of the facial area-of-interest, and the cheek region highly accounts for the face
The 60% of portion's area-of-interest, the width of the forehead region and the cheek region with the facial area-of-interest phase
Together.
Specifically, as shown in Fig. 2
The facial area-of-interest is so set including forehead region and cheek region is advantageous in that exclusion eye pair
The interference of IPPG signal extractions, improve the signal to noise ratio of signal.
It is described rich to the photoelectricity volume pulsation to trace time-domain signal and located in advance on the basis of above-mentioned each technical scheme
When reason can preferably be traced by the way that the Wavelet-denoising Method of moving average filter method or variable thresholding is rich to the photoelectricity volume pulsation
Domain signal carries out denoising.It is described it is rich to the photoelectricity volume pulsation trace time-domain signal pre-processed so set it is good
It is in the signal to noise ratio in the interference that can effectively remove the random noise in signal, improving signal.
Embodiment two
The embodiment of the present invention three provide non-contact more people's method for measuring heart rate based on SVMs, specifically include as
Lower step：
Step 100, facial area-of-interest is detected by face detection algorithm, the facial area-of-interest includes forehead
Region and cheek region.
Step 200, the facial area-of-interest gray average is calculated, obtain photoplethysmographic and trace time domain letter
Number.
Step 300, it is rich to the photoplethysmographic trace time-domain signal and pre-process, obtain the first time domain letter
Number.
Step 400, using SVMs first time-domain signal is classified, obtain heart rate signal；It is described right
First time-domain signal, which carries out classification, to be included：
Step 401, training stage, respectively by first time-domain signal by face extracting section time-domain signal with by
The time-domain signal of flase drop frame extraction is trained respectively as positive negative sample；
Step 402, test phase, tested using the SVMs trained, select the heart rate signal；
Wherein, the second time-domain signal obtained after the pretreatment is sent directly into SVM networks, and according to the result of classification
The operation of next step is carried out, if the signal is classified as heart rate signal, preserves the calculating that data are used for carrying out heart rate value, it is no
Then, by rejection of data.
Step 500, the heart rate signal is calculated, obtain heart rate value.
The technical scheme of the present embodiment, by SVMs, solve to the flase drop frame in Face datection algorithm and
The noise jamming problem that light change is brought, has reached the effect for improving heart rate detection precision.
Embodiment three
Fig. 4 is a kind of non-contact more people's heart rate measurement devices based on SVMs that the embodiment of the present invention three provides
Structural representation, the concrete structure for being somebody's turn to do non-contact more people's heart rate measurement devices based on SVMs are as follows：
Detection module, for detecting facial area-of-interest, the facial area-of-interest bag by face detection algorithm
Include forehead region and cheek region；
Primary signal acquisition module, for calculating the facial area-of-interest gray average, obtain photoelectricity volume pulsation
Ripple traces time-domain signal；
Pretreatment module, pre-processed for tracing time-domain signal to the photoplethysmographic, when obtaining first
Domain signal；
Heart rate signal acquisition module, for classifying using SVMs to first time-domain signal, obtain the heart
Rate signal；
Computing module, for calculating the heart rate signal, obtain heart rate value.
Wherein, the result of the real-time heart rate detection of more people is as shown in figure 5, the figure is in the present embodiment the use of resolution ratio to be 800*
The result tested under 600 common computer camera, ID represents the numbering of each test object in figure, and HRate represents corresponding and compiled
The heart rate value of number tester, the number on the right 64 represent that the Mean Time Measurement of a frame in video is 64ms.
The course of work of the device：Before measured is located at computer, facial video image is gathered using common camera, and in fact
When export measured heart rate value.
The technical scheme of the present embodiment, by computer camera and SVMs, solve more people's heart rate measurements and forbidden
True problem, the effect that more people's hearts rate quick and precisely measure is reached.
The said goods can perform the method that any embodiment of the present invention is provided, and possess the corresponding functional module of execution method
And beneficial effect.The said goods can perform the method that any embodiment of the present invention is provided, and possess the corresponding function of execution method
Module and beneficial effect.
Pay attention to, above are only presently preferred embodiments of the present invention and institute's application technology principle.It will be appreciated by those skilled in the art that
The invention is not restricted to specific embodiment described here, can carry out for a person skilled in the art various obvious changes,
Readjust and substitute without departing from protection scope of the present invention.Therefore, although being carried out by above example to the present invention
It is described in further detail, but the present invention is not limited only to above example, without departing from the inventive concept, also
Other more equivalent embodiments can be included, and the scope of the present invention is determined by scope of the appended claims.
Claims (10)
- A kind of 1. non-contact more people's method for measuring heart rate based on SVMs, it is characterised in that including：Facial area-of-interest is detected by face detection algorithm, the facial area-of-interest includes forehead region and cheek area Domain；The facial area-of-interest gray average is calculated, face-image photoplethysmographic is obtained and traces (IPPG) time domain letter Number；Time-domain signal is traced to the photoplethysmographic to pre-process, and obtains the first time-domain signal；First time-domain signal is classified using SVMs, obtains the second time-domain signal；Second time-domain signal is calculated, obtains heart rate value.
- 2. according to the method for claim 1, it is characterised in that described to detect that face is interested by face detection algorithm Region, including：Using multithreading.
- 3. according to the method for claim 1, it is characterised in that detect that face sense is emerging by face detection algorithm described After interesting region, in addition to：Track the facial area-of-interest；Gray average is taken to facial area-of-interest described in every frame.
- 4. according to the method for claim 1, it is characterised in that the facial area-of-interest includes forehead region and cheek Region, including：The forehead region height accounts for the 20% of the facial area-of-interest, and it is emerging that the cheek region highly accounts for the face sense The 60% of interesting region, the forehead region are identical with the facial area-of-interest with the width of the cheek region.
- 5. according to the method for claim 1, it is characterised in that described that time-domain signal progress is traced to photoplethysmographic Pretreatment, including：Time domain letter is traced to the photoplethysmographic by the Wavelet-denoising Method of moving average filter method and variable thresholding Number carry out denoising.
- 6. according to the method for claim 1, described to be classified using SVMs to the first time-domain signal, its feature It is, including：Using the fractal characteristic of heart rate signal, the time-domain signal directly obtained to area-of-interest is classified, without further Extract signal characteristic value.
- 7. according to the method for claim 1, described to be classified using SVMs to the first time-domain signal, its feature It is, including：Training stage, respectively by first time-domain signal by the time-domain signal of region of interesting extraction with being extracted by flase drop frame Time-domain signal be trained respectively as positive negative sample；Test phase, tested using the SVMs trained, select the heart rate signal.
- 8. according to the method for claim 1, it is characterised in that the calculating heart rate signal, including：Time frequency analysis is carried out to the heart rate signal sequence using Short Time Fourier Transform method.
- A kind of 9. non-contact more people's heart rate measurement devices based on SVMs, it is characterised in that including：Detection module, for detecting facial area-of-interest by face detection algorithm, the facial area-of-interest includes volume Head region and cheek region；Primary signal acquisition module, for calculating the facial area-of-interest gray average, obtain photoplethysmographic and retouch Remember time-domain signal；Pretreatment module, pre-processed for tracing time-domain signal to the photoplethysmographic, obtain the first time domain letter Number；Heart rate signal acquisition module, for being classified using SVMs to first time-domain signal, when obtaining second Domain signal；Computing module, for calculating the heart rate signal, obtain heart rate value.
- 10. device according to claim 9, described that first time-domain signal is classified using SVMs, It is characterised in that it includes：Training module, for respectively by the time-domain signal of the region of interesting extraction in first time-domain signal and by flase drop frame The time-domain signal of extraction is trained respectively as positive negative sample；Test module, tested with the SVMs trained, select the heart rate signal.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710608816.6A CN107334469A (en) | 2017-07-24 | 2017-07-24 | Non-contact more people's method for measuring heart rate and device based on SVMs |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710608816.6A CN107334469A (en) | 2017-07-24 | 2017-07-24 | Non-contact more people's method for measuring heart rate and device based on SVMs |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107334469A true CN107334469A (en) | 2017-11-10 |
Family
ID=60217452
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710608816.6A Pending CN107334469A (en) | 2017-07-24 | 2017-07-24 | Non-contact more people's method for measuring heart rate and device based on SVMs |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107334469A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109512418A (en) * | 2018-10-22 | 2019-03-26 | 哈尔滨理工大学 | A kind of volumetric blood flow pulse imaging reduction motion artifacts method |
| CN109512416A (en) * | 2018-11-21 | 2019-03-26 | 哈尔滨理工大学 | A kind of volume pulsation wave extracting method and system |
| CN110866498A (en) * | 2019-11-15 | 2020-03-06 | 北京华宇信息技术有限公司 | Portable heart rate monitoring device and heart rate monitoring method thereof |
| CN110974196A (en) * | 2019-12-13 | 2020-04-10 | 福州大学 | Non-contact respiration and heart rate detection method in motion state |
| CN111127511A (en) * | 2018-12-18 | 2020-05-08 | 玄云子智能科技(深圳)有限责任公司 | Non-contact heart rate monitoring method |
| CN111670004A (en) * | 2018-03-07 | 2020-09-15 | 三星电子株式会社 | Electronic device and method for measuring heart rate |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2438849A1 (en) * | 2010-10-06 | 2012-04-11 | Latvijas Universitate | Device and method for an optical contactless monitoring of cardiac parameters |
| CN102499664A (en) * | 2011-10-24 | 2012-06-20 | 西双版纳大渡云海生物科技发展有限公司 | Video-image-based method and system for detecting non-contact vital sign |
| CN104657712A (en) * | 2015-02-09 | 2015-05-27 | 惠州学院 | Method for detecting masked person in monitoring video |
| CN105718860A (en) * | 2016-01-15 | 2016-06-29 | 武汉光庭科技有限公司 | Positioning method and system based on safe driving map and binocular recognition of traffic signs |
| CN105787475A (en) * | 2016-03-29 | 2016-07-20 | 西南交通大学 | Traffic sign detection and identification method under complex environment |
| CN105893941A (en) * | 2016-03-28 | 2016-08-24 | 电子科技大学 | Facial expression identifying method based on regional images |
| CN106264505A (en) * | 2016-07-21 | 2017-01-04 | 浙江师范大学 | A kind of heart rate spectral peak system of selection based on support vector machine |
| CN106491117A (en) * | 2016-12-06 | 2017-03-15 | 上海斐讯数据通信技术有限公司 | A kind of signal processing method and device based on PPG heart rate measurement technology |
| CN106778695A (en) * | 2017-01-19 | 2017-05-31 | 北京理工大学 | A kind of many people's examing heartbeat fastly methods based on video |
-
2017
- 2017-07-24 CN CN201710608816.6A patent/CN107334469A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP2438849A1 (en) * | 2010-10-06 | 2012-04-11 | Latvijas Universitate | Device and method for an optical contactless monitoring of cardiac parameters |
| CN102499664A (en) * | 2011-10-24 | 2012-06-20 | 西双版纳大渡云海生物科技发展有限公司 | Video-image-based method and system for detecting non-contact vital sign |
| CN104657712A (en) * | 2015-02-09 | 2015-05-27 | 惠州学院 | Method for detecting masked person in monitoring video |
| CN105718860A (en) * | 2016-01-15 | 2016-06-29 | 武汉光庭科技有限公司 | Positioning method and system based on safe driving map and binocular recognition of traffic signs |
| CN105893941A (en) * | 2016-03-28 | 2016-08-24 | 电子科技大学 | Facial expression identifying method based on regional images |
| CN105787475A (en) * | 2016-03-29 | 2016-07-20 | 西南交通大学 | Traffic sign detection and identification method under complex environment |
| CN106264505A (en) * | 2016-07-21 | 2017-01-04 | 浙江师范大学 | A kind of heart rate spectral peak system of selection based on support vector machine |
| CN106491117A (en) * | 2016-12-06 | 2017-03-15 | 上海斐讯数据通信技术有限公司 | A kind of signal processing method and device based on PPG heart rate measurement technology |
| CN106778695A (en) * | 2017-01-19 | 2017-05-31 | 北京理工大学 | A kind of many people's examing heartbeat fastly methods based on video |
Non-Patent Citations (2)
| Title |
|---|
| 孔令琴: "非接触式生理信号检测关键技术研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
| 范九伦: "《模式识别导论》", 31 May 2012 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111670004A (en) * | 2018-03-07 | 2020-09-15 | 三星电子株式会社 | Electronic device and method for measuring heart rate |
| CN109512418A (en) * | 2018-10-22 | 2019-03-26 | 哈尔滨理工大学 | A kind of volumetric blood flow pulse imaging reduction motion artifacts method |
| CN109512416A (en) * | 2018-11-21 | 2019-03-26 | 哈尔滨理工大学 | A kind of volume pulsation wave extracting method and system |
| CN111127511A (en) * | 2018-12-18 | 2020-05-08 | 玄云子智能科技(深圳)有限责任公司 | Non-contact heart rate monitoring method |
| CN111127511B (en) * | 2018-12-18 | 2022-03-29 | 玄云子智能科技(深圳)有限责任公司 | Non-contact heart rate monitoring method |
| CN110866498A (en) * | 2019-11-15 | 2020-03-06 | 北京华宇信息技术有限公司 | Portable heart rate monitoring device and heart rate monitoring method thereof |
| CN110974196A (en) * | 2019-12-13 | 2020-04-10 | 福州大学 | Non-contact respiration and heart rate detection method in motion state |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107334469A (en) | Non-contact more people's method for measuring heart rate and device based on SVMs | |
| Qiu et al. | EVM-CNN: Real-time contactless heart rate estimation from facial video | |
| CN105147274B (en) | A kind of method that heart rate is extracted in the face video signal from visible spectrum | |
| Hsu et al. | Learning-based heart rate detection from remote photoplethysmography features | |
| CN110097034A (en) | A kind of identification and appraisal procedure of Intelligent human-face health degree | |
| CN109350030B (en) | System and method for processing human face video heart rate signal based on phase amplification | |
| Casado et al. | Face2PPG: An unsupervised pipeline for blood volume pulse extraction from faces | |
| CN103827916A (en) | Wrinkle detection method, wrinkle detection device and wrinkle detection program, as well as wrinkle evaluation method, wrinkle evaluation device and wrinkle evaluation program | |
| Feng et al. | Motion artifacts suppression for remote imaging photoplethysmography | |
| CN105844612A (en) | Device and method for analyzing biophysiological periodic data | |
| JP2017093760A (en) | Device and method for measuring periodic variation interlocking with heart beat | |
| CN104424632B (en) | A kind of video contrast's method for detecting abnormality and device | |
| JP6727469B1 (en) | Information processing apparatus, program, and information processing method | |
| CN110236511A (en) | A kind of noninvasive method for measuring heart rate based on video | |
| Jayanthi et al. | Automatic diagnosis of retinal diseases from color retinal images | |
| EP3393347B1 (en) | Apparatus and method for measuring the quality of an extracted signal | |
| Tschirren et al. | Automated analysis of Doppler ultrasound velocity flow diagrams | |
| Szankin et al. | Long distance vital signs monitoring with person identification for smart home solutions | |
| CN104679967A (en) | Method for judging reliability of psychological test | |
| CN114140496A (en) | Non-contact respiration detection method and device | |
| Rouast et al. | Remote photoplethysmography: Evaluation of contactless heart rate measurement in an information systems setting | |
| Zheng et al. | Shielding facial physiological information in video | |
| Slapnicar et al. | Contact-free monitoring of physiological parameters in people with profound intellectual and multiple disabilities | |
| CN104688199A (en) | Non-contact type pulse measurement method based on skin pigment concentration difference | |
| CN109524109A (en) | A kind of contactless fatigue monitoring method based on muscle pressure state |
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: 20171110 |
|
| WD01 | Invention patent application deemed withdrawn after publication |