CN105962915A - Non-contact human body respiration rate and heart rate synchronous measurement method and system - Google Patents
Non-contact human body respiration rate and heart rate synchronous measurement method and system Download PDFInfo
- Publication number
- CN105962915A CN105962915A CN201610404234.1A CN201610404234A CN105962915A CN 105962915 A CN105962915 A CN 105962915A CN 201610404234 A CN201610404234 A CN 201610404234A CN 105962915 A CN105962915 A CN 105962915A
- Authority
- CN
- China
- Prior art keywords
- signal
- heart rate
- value
- breathing
- rate
- 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
- 230000029058 respiratory gaseous exchange Effects 0.000 title claims abstract description 92
- 230000001360 synchronised effect Effects 0.000 title claims abstract description 6
- 238000000691 measurement method Methods 0.000 title abstract description 4
- 238000000034 method Methods 0.000 claims abstract description 63
- 238000001914 filtration Methods 0.000 claims abstract description 25
- 238000012935 Averaging Methods 0.000 claims abstract description 7
- 230000001427 coherent effect Effects 0.000 claims abstract description 7
- 238000005259 measurement Methods 0.000 claims abstract description 7
- 238000000926 separation method Methods 0.000 claims abstract description 7
- 238000010606 normalization Methods 0.000 claims abstract description 6
- 238000001228 spectrum Methods 0.000 claims description 17
- 238000000605 extraction Methods 0.000 claims description 13
- 230000001235 sensitizing effect Effects 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 12
- 239000000284 extract Substances 0.000 claims description 10
- 238000013459 approach Methods 0.000 claims description 3
- 238000007405 data analysis Methods 0.000 claims description 3
- 230000007274 generation of a signal involved in cell-cell signaling Effects 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 206010008190 Cerebrovascular accident Diseases 0.000 claims 1
- 208000006011 Stroke Diseases 0.000 claims 1
- 230000008901 benefit Effects 0.000 abstract description 6
- 238000007781 pre-processing Methods 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 8
- 230000000694 effects Effects 0.000 description 8
- 230000008859 change Effects 0.000 description 7
- 238000011160 research Methods 0.000 description 7
- 230000033001 locomotion Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 238000002474 experimental method Methods 0.000 description 4
- 238000003384 imaging method Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000015572 biosynthetic process Effects 0.000 description 2
- 239000008280 blood Substances 0.000 description 2
- 210000004369 blood Anatomy 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000013461 design Methods 0.000 description 2
- 230000004069 differentiation Effects 0.000 description 2
- 230000001815 facial effect Effects 0.000 description 2
- 230000001788 irregular Effects 0.000 description 2
- 230000000750 progressive effect Effects 0.000 description 2
- 230000011664 signaling Effects 0.000 description 2
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000004438 eyesight Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009532 heart rate measurement Methods 0.000 description 1
- 239000010977 jade Substances 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000013186 photoplethysmography Methods 0.000 description 1
- 230000036391 respiratory frequency Effects 0.000 description 1
- 230000033764 rhythmic process Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 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
- A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
-
- 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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
- A61B5/0816—Measuring devices for examining respiratory frequency
-
- 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
-
- 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
- A61B5/7235—Details of waveform analysis
- A61B5/725—Details of waveform analysis using specific filters therefor, e.g. Kalman or adaptive filters
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Physiology (AREA)
- Public Health (AREA)
- Medical Informatics (AREA)
- Veterinary Medicine (AREA)
- Physics & Mathematics (AREA)
- General Health & Medical Sciences (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Signal Processing (AREA)
- Cardiology (AREA)
- Psychiatry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Artificial Intelligence (AREA)
- Pulmonology (AREA)
- Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)
- Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)
Abstract
The invention relates to a non-contact human respiration rate and heart rate synchronous measurement method and a system, wherein the measurement method comprises the following steps of obtaining human face video screen information, and selecting two sensitive areas from a video frame image; generating 2 groups of RGB observation signals by respectively using a coherent averaging method for the pixel value of the selected double-sensitive area of each frame in the video, and then sequentially carrying out high-pass filtering, trend removing, mean value removing and normalization preprocessing operations on the 2 groups of RGB observation signals; the device is used for carrying out 6-channel blind source separation on 2 groups of RGB observation signals generated based on the double sensitive areas, and separating out a respiration signal and a heart rate signal; the method is used for identifying a respiration signal and a heart rate signal from source signals after blind source separation, and extracting respiration and heart rate by combining a sliding window algorithm. The method has the advantages of high accuracy in synchronous measurement of the respiration rate and the heart rate, strong noise interference resistance, large application potential and the like.
Description
Technical field
The present invention relates to biological sign research applied technical field, be specifically related to a kind of contactless humanbody breathing rate and heart rate method for synchronously measuring and system.
Background technology
Heart rate and breathing rate are as important human body physical sign parameter, how by one of contactless metering system accurately focus measuring always Biomedical Engineering and Instruments area research realizing the two parameter.Imaging type photoplethaysmography (Imaging Photoplethysmography, IPPG) technology is a kind of noncontact physio-parameter detection technology grown up on the basis of PPG, this technology utilizes imaging device that the information comprising tested position is carried out video acquisition, by sensitizing range (the region of interest to video image, ROI) process, it is achieved a kind of biomedical detection method that the physiological parameters such as heart rate, breathing rate, blood oxygen saturation are extracted.IPPG technology have low cost, noncontact, safety, can measure continuously, the multiple advantage such as simple to operate, the research for contactless physiological signal measurements and remote medical monitoring provides a kind of new solution route and scheme.
In prior art, extracting the research aspect of human heart rate's index based on IPPG technology, the mainly extracting method of heart rate and breathing rate is directly applying or improving and optimizating of Research Thinking.The G passage of the observation signal predominantly generated from face video extracts heart rate, or by carrying out 3 passage blind source separating to extract heart rate based on classics ICA algorithm such as JADE, from the spectrum analysis of heart rate, extracts breathing rate further.Additionally, Part Methods has incorporated facial video tracking algorithm to overcome noise jamming problem.As a rule, when interference noise is less, the effect using G passage method is really fine, but the method existing defects during more interference noise, and, more noise source can affect the separating effect of 3 passage ICA algorithm.Although experimenter's facial movement problem can be solved above-mentioned to a certain extent based on face video tracking algorithm, but the noise jamming caused for the face local factor such as slight change and the faint change of light cannot accomplish to overcome well.Simultaneously, existing method separates opisthogenesis signalling channel for ICA and differentiates basic according to the power spectrum after FFT, the most also facing the interference problem of noise source, in the case of split tunnel is more, the discriminant accuracy of source channels seems particularly significant for the robustness of algorithm equally.Additionally, existing method is not capable of the simultaneous extraction of heart rate and breath signal, and then realize the synchro measure of heart rate and breathing.
Summary of the invention
It is an object of the invention to: provide a kind of contactless humanbody breathing rate to go out to breathe and heart rate signal with simultaneous extraction in heart rate method for synchronously measuring and system dignity portion video, it is possible to while overcoming noise jamming problem, it is achieved the synchro measure of two physiological parameters.
For achieving the above object, present method invention the technical scheme is that
Contactless humanbody breathing rate and heart rate method for synchronously measuring, this measuring method comprises the steps,
S100, acquisition human body face screen information, select sensitizing range at two from video frame images；
S101, pixel value to the selected Lazer sensillary area of frame each in video, use coherent averaging technique respectively, generates 2 groups of RGB observation signals, then 2 groups of RGB observation signals carry out high-pass filtering successively, goes trend, go average, normalization pretreatment operation；
S102, for 2 groups of RGB observation signals based on sensitive Area generation are carried out 6 passage blind source separating, isolate breath signal and heart rate signal；
S103, for source signal after blind source separating, identify breath signal and heart rate signal, extract breathing and heart rate in conjunction with sliding window algorithm.
Present method invention there is also following supplementary features:
In described step S103, specifically include following steps:
S201, isolated 6 channel source signals use 0.15Hz high-pass filterings and 8Hz low-pass filtering combine be filtered processing, remove residual noise interference；
S202, to 6 channel source signals of denoising in S201 step, calculate kurtosis value, and use K-means method that kurtosis value is carried out 3 cluster analyses, obtain the cluster that the kurtosis value of the cluster at breath signal place, i.e. signal is minimum；
S203, from cluster minimum described in step S202, determine breath signal, introduce breathing rate linear predictor, LPC linear prediction method is used to calculate current predicted value from 5 nearest breathing rate history value, and analyze in this cluster the Frequency point at all signals peak value in 0.2～0.8Hz frequency range, the nearest Frequency point of chosen distance predictive value is breathing rate candidate value；
Judging whether this candidate value exceeds the domain of walker of predictive value ± 0.3Hz, if not having, then obtaining breath signal and current breathing rate value；If beyond the domain of walker of predictive value, being then considered as outlier and abandon, this recognition failures, and return step S201；
S204, to step S203 is extracted remaining source signal after breath signal, use 0.8Hz high-pass filtering to eliminate low frequency component, calculate power spectrum kurtosis value again, and use K-means method to carry out 3 cluster analyses, obtain the cluster of the power spectrum kurtosis value maximum of the signal at heart rate signal place；
S205: determine heart rate signal from the maximum cluster described in step S204, introduce heart rate linear predictor, LPC linear prediction method is used to calculate current predicted value from 5 nearest heart rate history value, analyzing in this cluster Frequency point at all signals peak value in 0.8～2.3Hz frequency range, the nearest Frequency point of chosen distance predictive value is heart rate candidate value；
Judging whether this candidate value exceeds the domain of walker of predictive value ± 0.2Hz, if not having, then obtaining heart rate signal and Current heart rate value, if beyond the domain of walker of predictive value, being then considered as outlier and abandon, this recognition failures, and return step S201；
S206, to described face video use sliding window algorithm be analyzed, a length of 600 frames of sliding window, sliding step is 150 frames, data application approach described above in sliding window is obtained breathing rate and heart rate every time.
Introducing breathing rate described in described step S203 and step S205 and the linear predictor of heart rate, it is limited to after sliding window moves 5 times, in the sliding window of first 5 times in data analysis, kurtosis value reckling in source signal described in step S202 is appointed as breath signal, and power spectrum kurtosis value the maximum in the source signal after the 0.8Hz high-pass filtering described in step S204 is appointed as heart rate signal.
Blind source separation algorithm in described step S103 selects SOBI algorithm based on second-order statistic.
2 groups of RGB observation signals in described step S101 select the high-pass filtering of 0.15Hz.
For achieving the above object, native system invention the technical scheme is that
Contactless humanbody breathing rate and heart rate synchronized measurement system, it is characterised in that: system includes that face video acquisition and sensitive region selection module, observation signal generate and pretreatment module, multichannel blind source separating module, source signal identification and physiological parameter extraction module；
Face video acquisition is used for shooting human body face video with sensitive region selection module, then selects sensitizing range at two, respectively mouth region and bottleneck throat region from video frame images；
Observation signal generation and pretreatment module are used for the pixel value of the selected Lazer sensillary area to frame each in video, use coherent averaging technique respectively, generate 2 groups of RGB observation signals, again 2 groups of RGB observation signals are carried out successively 0.15Hz high-pass filtering, go trend, go average, normalization etc. 4 step pretreatment operation to suppress interference noise, it is thus achieved that comparatively ideal observation signal be used for subsequent analysis；
Multichannel blind source separating module is for carrying out 6 passage blind source separating, to isolate breath signal and heart rate signal to 2 groups of RGB observation signals based on sensitive Area generation；
Source signal identification identifies breath signal and heart rate signal with physiological parameter extraction module for the source signal after blind source separating, extracts breathing and heart rate in conjunction with sliding window algorithm.
Compared with prior art, the technique effect that the present invention possesses is: present invention achieves simultaneous extraction from face video and goes out human body respiration signal and heart rate signal, sensitive region based on face video, i.e. bottleneck throat region and mouth region.Wherein, bottleneck throat region has the characteristics of motion, and breathing rate strong correlation, and is generally difficult to be blocked by medicated clothing, is suitable as breathing rate sensitizing range；And there is irregular phenomenon in mouth region motion feature, weak to breathing rate relevant, mouth region is the same with whole face simultaneously, BVP signal intensity feature is obvious, can be as heart rate sensitizing range, therefore use this sensitive areas combine, achieve human body respiration signal and heart rate signal simultaneous extraction well；Additionally, present invention employs the technological means served as theme with blind source separating, 2 groups of RGB observation signals based on sensitive Area generation use 6 passage blind source separating process, by breath signal, heart rate signal, various noise source signal are separated well, solve various noise jamming problem, to the source signal after separating, kurtosis cluster and power spectrum kurtosis has been used to cluster the recognition methods combined with linear predictor, can identification of breathing signal exactly, heart rate signal, therefore the present invention has huge application potential, have breathing rate and heart rate synchro measure accuracy high, anti-noise jamming ability is stronger, the advantages such as application potential is big.
In addition to objects, features and advantages described above, the present invention also has other objects, features and advantages.Below with reference to figure, the present invention is further detailed explanation.
Accompanying drawing explanation
The Figure of description of the part constituting the application is used for providing a further understanding of the present invention, and the schematic description and description of the present invention is used for explaining the present invention, is not intended that inappropriate limitation of the present invention.In the accompanying drawings:
Fig. 1 is the logic diagram of contactless human body respiration rate and heart rate method for synchronously measuring；
Fig. 2 is the logic diagram of contactless human body respiration rate and heart rate synchronized measurement system；
Fig. 3 is contactless human body respiration rate and the logic diagram of the concrete grammar of step S103 in heart rate method for synchronously measuring；
Fig. 4 is contactless human body respiration rate and heart rate method for synchronously measuring flow chart of steps；
Fig. 5 is sensitizing range schematic diagram at two chosen in contactless human body respiration rate and heart rate method for synchronously measuring；
Fig. 6 is that selected video data carries out single region blind source separating design sketch respectively；
Fig. 7 is that data described in Fig. 6 carry out dual area 6 passage blind source separating design sketch；
Fig. 8 is the schematic diagram of identification of breathing signal from source signal described in Fig. 7；
Fig. 9 is the schematic diagram remaining from Fig. 8 and identifying heart rate signal in source signal；
Figure 10 is experimenter's breathing rate and heart rate measurement curve chart；
Figure 11 is experimenter's breathing rate and heart rate frequency distribution histogram.
Detailed description of the invention
It is further described in conjunction with Fig. 1 to Figure 11 is bright:
Contactless humanbody breathing rate and heart rate method for synchronously measuring, this measuring method comprises the steps,
S100, acquisition human body face screen information, select sensitizing range at two, respectively mouth region and bottleneck throat region from video frame images；
S101, pixel value to the selected Lazer sensillary area of frame each in video, use coherent averaging technique respectively, generates 2 groups of RGB observation signals, then 2 groups of RGB observation signals carry out high-pass filtering successively, goes trend, go average, normalization pretreatment operation；
S102, for 2 groups of RGB observation signals based on sensitive Area generation are carried out 6 passage blind source separating, isolate breath signal and heart rate signal；
S103, for source signal after blind source separating, identify breath signal and heart rate signal, extract breathing and heart rate in conjunction with sliding window algorithm.
First, present invention achieves simultaneous extraction from face video and go out human body respiration signal and heart rate signal, devise sensitive region based on face video, i.e. bottleneck throat region and mouth region.Wherein, bottleneck throat region has the characteristics of motion, and breathing rate strong correlation, and is generally difficult to be blocked by medicated clothing, is suitable as breathing rate sensitizing range；There is irregular phenomenon in mouth region motion feature, weak to breathing rate relevant, mouth region is the same with whole face simultaneously, BVP (blood volume change pulse) signal, signal intensity feature is obvious, can be as heart rate sensitizing range, compared with the prior art, use this sensitive areas combine, achieve human body respiration signal and heart rate signal simultaneous extraction well.
nullSecondly，Invention have employed the technological means served as theme with blind source separating，To 2 groups of RGB (R based on sensitive Area generation、G、B three primary colours passage is by red、Green、Change and their superpositions each other of blue three Color Channels obtain color miscellaneous，RGB be i.e. represent red、Green、The color of blue three passages，This standard almost include human eyesight can all colours of perception，It being to use at present one of the widest color system) observation signal uses 6 passage blind source separating to process，By breath signal、Heart rate signal、Various noise source signal are separated well，Solve the various noise jamming problems being commonly encountered in prior art，Simultaneously，To the source signal after separating，Kurtosis cluster and power spectrum kurtosis has creatively been used to cluster the recognition methods combined with linear predictor，Can identification of breathing signal exactly、Heart rate signal.
Additionally, contactless physiological parameter measurement method based on face video, the most gradually start to become research and the focus of application, correlation technique proposed by the invention possesses multiple innovative point, achieve the synchro measure of breathing rate and heart rate, the technological means served as theme with blind source separating used can solve noise jamming problem well, can further expand application, to extract more physiological parameter in this research field.Meanwhile, the method can be implemented in line function, and selected SOBI blind source separation algorithm possesses relatively low computational complexity, beneficially present invention graft application on nonidentity operation platform, and therefore the present invention has huge application potential.
In a word, a kind of contactless humanbody breathing rate of the present invention and heart rate method for synchronously measuring, there is, anti-noise jamming ability high to breathing rate and heart rate synchro measure accuracy compared with the advantage such as strong, application potential is big.
Preferably, shown in Fig. 3, in described step S103, for source signal after blind source separating, identify breath signal and heart rate signal, extract breathing and heart rate in conjunction with sliding window algorithm, specifically include following steps:
S201, isolated 6 channel source signals use 0.15Hz high-pass filterings and 8Hz low-pass filtering combine be filtered processing, remove residual noise interference；
S202, to 6 channel source signals of denoising in S201 step, calculate kurtosis value, and use K-means (hard clustering algorithm) method that kurtosis value is carried out 3 cluster analyses, obtain the cluster that the kurtosis value of the cluster at breath signal place, i.e. signal is minimum；
S203, from cluster minimum described in step S202, determine breath signal, introduce breathing rate linear predictor, LPC linear prediction method is used to calculate current predicted value from 5 nearest breathing rate history value, and analyze in this cluster the Frequency point at all signals peak value in 0.2～0.8Hz frequency range, the nearest Frequency point of chosen distance predictive value is breathing rate candidate value；
Judging whether this candidate value exceeds the domain of walker of predictive value ± 0.3Hz, if not having, then obtaining breath signal and current breathing rate value；If beyond the domain of walker of predictive value, being then considered as outlier and abandon, this recognition failures, and return step S201；
S204, to step S203 is extracted remaining source signal after breath signal, use 0.8Hz high-pass filtering to eliminate low frequency component, calculate power spectrum kurtosis value again, and use K-means method to carry out 3 cluster analyses, obtain the cluster of the power spectrum kurtosis value maximum of the signal at heart rate signal place；
S205: in order to determine heart rate signal from the maximum cluster described in step S204, introduce heart rate linear predictor, LPC linear prediction method is used to calculate current predicted value from 5 nearest heart rate history value, analyzing in this cluster Frequency point at all signals peak value in 0.8～2.3Hz frequency range, the nearest Frequency point of chosen distance predictive value is heart rate candidate value；
Judging whether this candidate value exceeds the domain of walker of predictive value ± 0.2Hz, if not having, then obtaining heart rate signal and Current heart rate value, if beyond the domain of walker of predictive value, being then considered as outlier and abandon, this recognition failures, and return step S201；
S206, to described face video use sliding window algorithm be analyzed, a length of 600 frames of sliding window, sliding step is 150 frames, data application approach described above in sliding window is obtained breathing rate and heart rate every time.
Further, introducing breathing rate described in described step S203 and step S205 and the linear predictor of heart rate, it is limited to after sliding window moves 5 times, in the sliding window of first 5 times in data analysis, kurtosis value reckling in source signal described in step S202 is appointed as breath signal, and power spectrum kurtosis value the maximum in the source signal after the 0.8Hz high-pass filtering described in step S204 is appointed as heart rate signal.
Further, the blind source separation algorithm in described step S103 selects SOBI algorithm based on second-order statistic.
For improving the accuracy of screen signals collecting, the height of the mouth region in described step S100 covers mouth, width close to face edge.
For improving signal filtering quality, 2 groups of RGB observation signals in described step S101 select the high-pass filtering of 0.15Hz.
Table 1 gives the measurement effect statistics of this specification institute extracting method, and with common commercial measuring instrument as standard reference data, in the case of user attitude is loosened, the detection accuracy of breathing rate and heart rate reaches 90% and more than 93%.
Table 1
It is introduced with heart rate combined detection system below for contactless nictation based on second-order blind identification:
Shown in Fig. 2, system includes that face video acquisition and sensitive region selection module 10, observation signal generate and pretreatment module 20, multichannel blind source separating module 30, source signal identification and physiological parameter extraction module 40；
Face video acquisition is used for shooting human body face video with sensitive region selection module 10, then selects sensitizing range at two, respectively mouth region and bottleneck throat region from video frame images；
Observation signal generation is used for the pixel value of the selected Lazer sensillary area to frame each in video with pretreatment module 20, use coherent averaging technique respectively, generate 2 groups of RGB observation signals, again 2 groups of RGB observation signals are carried out successively 0.15Hz high-pass filtering, go trend, go average, normalization etc. 4 step pretreatment operation to suppress interference noise, it is thus achieved that comparatively ideal observation signal be used for subsequent analysis；
Multichannel blind source separating module 30 is for carrying out 6 passage blind source separating, to isolate breath signal and heart rate signal to 2 groups of RGB observation signals based on sensitive Area generation；
Source signal identification identifies breath signal and heart rate signal with physiological parameter extraction module 40 for the source signal after blind source separating, extracts breathing and heart rate in conjunction with sliding window algorithm.
Below for accompanying drawing, the method and system of the present invention are carried out brief description:
In conjunction with Fig. 5, in the present embodiment, on the premise of without loss of generality, select the video data of one section of normal subjects, the RGB observation signal oscillogram generated by region, some places in analysis video frame, illustrate sensitizing range at two selected by the present invention, be mouth region and bottleneck throat region respectively, be respectively designated as region (I) and region (II).
In conjunction with Fig. 6, illustrate the 2 groups of RGB observation signals generated based on region (I) and region (II) are carried out respectively the effect of 3 passage blind source separating.RGB observation signal is in most cases and the improper physiological signal that is directly used in extracts, the commonly used the method for a lot of prior arts, such as, uses G passage to be used for extracting heart rate.In the present embodiment, on the premise of without loss of generality, selected experimental data has more interference noise, as shown in Fig. 6 (a) (c).Efficiently separating of physiological signal and noise signal can not be realized well by 3 passage blind source separating, as shown in Fig. 6 (b) (d), part prior art use single region 3 passage blind source separation method, faces similar defect equally.
In conjunction with Fig. 7, illustrating in the present embodiment to carry out 2 described in Fig. 3 group RGB observation signal the effect of 6 passage blind source separating, in terms of result, echo signal and each noise source are preferably separated, as shown in Fig. 7 (b).It is filtered removing residual noise, as shown in Fig. 7 (c) to it further.But it can be seen that passage 1 and passage 2 occur in that interference in the differentiation of breath signal from the spectrogram of Fig. 7 (c), therefore, only carry out source signal identification Shortcomings only in accordance with spectrogram, need further by new distinguished number.According to spectrogram identification source signal in part prior art, face similar problem equally.
In conjunction with Fig. 8, illustrate the present embodiment identifies from the source signal after 6 passage blind source separating described in Fig. 4 the schematic diagram of breath signal.The mode using source signal kurtosis value cluster and linear prediction to combine in the present invention carries out breathing source signal and differentiates, first calculate the kurtosis value of source signal, re-use K-means method, according to kurtosis value, source signal is carried out 3 cluster analyses, find out the cluster at breath signal place, the cluster that i.e. kurtosis value is minimum, comprise passage 1 and passage 2, it mostly is slow progressive formation in view of the change of human body respiration rate, further by breathing rate linear predictor subsidiary discriminant current breath signal passage, analyze in this cluster Frequency point at all signals peak value in 0.2～0.8Hz frequency range, the value range prediction value of passage 2 is nearest, elect breathing rate candidate value as, confirm the candidate value domain of walker without departing from predictive value ± 0.3Hz again, obtain breathing rate passage and breathing rate value, otherwise it is considered as outlier to abandon.
In conjunction with Fig. 9, illustrate the present embodiment identifies remaining source signal after the extraction breath signal described in Fig. 5 the schematic diagram of breath signal.nullThe present invention first use 0.8Hz high-pass filtering to eliminate low frequency component，The mode combined by source signal power spectrum kurtosis value cluster and linear prediction again carries out heart rate source signal differentiation，First meter spectrum calculates power spectrum kurtosis value，Re-use K-means method and carry out 3 cluster analyses according to power spectrum kurtosis value，Find out the cluster at heart rate signal place，The cluster that i.e. the power spectrum kurtosis value of signal is maximum，The present embodiment only comprises passage 3，It mostly is slow progressive formation in view of human heart rate's change，Further by heart rate linear predictor subsidiary discriminant Current heart rate signalling channel，Analyze in this cluster Frequency point at all signals peak value in 0.8～2.3Hz frequency range，The value range prediction value of passage 3 is nearest，Elect heart rate candidate value as，Confirm the candidate value domain of walker without departing from predictive value ± 0.2Hz again，Obtain heart rate passage and heart rate value，Otherwise it is considered as outlier to abandon.
In conjunction with Figure 10 and Figure 11, illustrate breathing rate and the measurement effect example of heart rate in the present embodiment, wherein " * " data are breathing rate measured by this method and heart rate (Estimated RR/HR), and " " data are data measured by business measuring instrument (Reference RR/HR).In order to effectively measure experimenter's breathing rate variation characteristic, in experiment, experimenter is interim consciously adjusts respiratory frequency, replaces successively by controlling mild breathing and quick breathing, manufactures 2 fluctuating waveforms on breathing, so that observation experiment effect, experiment shooting time is 4 to 6 minutes.In experiment, every secondary wave crest (the most quickly breathing) continues 45～60 seconds, then interval (breathing the most gently) 45～60 seconds.In terms of result, there are 2 secondary fluctuation waveforms in the breathing rate curve of experimenter, and wavy curve is the most perfect, the most identical with business gage data waveform, recording breathing rate situation of change well, meanwhile, this method the most accurately and is stablized for the rhythm of the heart of experimenter.Breathing rate and heart rate frequency distribution histogram show, in the range of breathing rate value and heart rate value are distributed in normal human's physiological parameter.
Claims (6)
1. contactless humanbody breathing rate and heart rate method for synchronously measuring, it is characterised in that: this measuring method includes as follows
Step,
S100, acquisition human body face screen information, select sensitizing range at two from video frame images；
S101, pixel value to the selected Lazer sensillary area of frame each in video, use coherent averaging technique respectively, generate 2
Group RGB observation signal, then 2 groups of RGB observation signals are carried out successively high-pass filtering, goes trend, go average, return
One changes pretreatment operation；
S102, for 2 groups of RGB observation signals based on sensitive Area generation are carried out 6 passage blind source separating, point
Separate out breath signal and heart rate signal；
S103, for source signal after blind source separating, identify breath signal and heart rate signal, calculate in conjunction with sliding window
Method extracts breathing and heart rate.
Contactless humanbody breathing rate the most according to claim 1 and heart rate method for synchronously measuring, it is characterised in that:
In described step S103, specifically include following steps:
S201,0.15Hz high-pass filterings and 8Hz low-pass filtering is used to combine and carry out isolated 6 channel source signals
Filtering Processing, removes residual noise interference；
S202, to 6 channel source signals of denoising in S201 step, calculate kurtosis value, and use K-means
Method carries out 3 cluster analyses to kurtosis value, obtains the cluster that the kurtosis value of the cluster at breath signal place, i.e. signal is minimum；
S203, from cluster minimum described in step S202, determine breath signal, introduce breathing rate linear predictor,
Use LPC linear prediction method to calculate current predicted value from 5 nearest breathing rate history value, and it is poly-to analyze this
Frequency point at apoplexy due to endogenous wind all signals peak value in 0.2～0.8Hz frequency range, the nearest Frequency point of chosen distance predictive value is
Breathing rate candidate value；
Judge this candidate value whether beyond the domain of walker of predictive value ± 0.3Hz, if not having, then obtain breath signal and
Current breathing rate value；If beyond the domain of walker of predictive value, being then considered as outlier and abandon, this recognition failures, and return
Step S201；
S204, to step S203 is extracted remaining source signal after breath signal, use 0.8Hz high-pass filtering to eliminate
Low frequency component, then calculate power spectrum kurtosis value, and use K-means method to carry out 3 cluster analyses, obtain heart rate
The cluster that the power spectrum kurtosis value of the signal at signal place is maximum；
S205: determine heart rate signal from the maximum cluster described in step S204, introduce heart rate linear predictor, make
From 5 nearest heart rate history value, calculate current predicted value with LPC linear prediction method, analyze institute in this cluster
Having Frequency point at signal peak value in 0.8～2.3Hz frequency range, the nearest Frequency point of chosen distance predictive value is heart rate candidate
Value；
Judge this candidate value whether beyond the domain of walker of predictive value ± 0.2Hz, if not having, then obtain heart rate signal and
Current heart rate value, if beyond the domain of walker of predictive value, being then considered as outlier and abandon, this recognition failures, and return step
Rapid S201；
S206, to described face video use sliding window algorithm be analyzed, a length of 600 frames of sliding window, sliding step
It is 150 frames, data application approach described above in sliding window is obtained breathing rate and heart rate every time.
Contactless humanbody breathing rate the most according to claim 2 and heart rate method for synchronously measuring, it is characterised in that:
Introducing breathing rate described in described step S203 and step S205 and the linear predictor of heart rate, be limited to sliding window and move
After dynamic 5 times, in the sliding window of first 5 times in data analysis, by kurtosis value in the source signal described in step S202
Little person is appointed as breath signal, and by power spectrum kurtosis in the source signal after the 0.8Hz high-pass filtering described in step S204
Value the maximum is appointed as heart rate signal.
Contactless humanbody breathing rate the most according to claim 1 and heart rate method for synchronously measuring, it is characterised in that:
Blind source separation algorithm in described step S103 selects SOBI algorithm based on second-order statistic.
Contactless humanbody breathing rate the most according to claim 1 and heart rate method for synchronously measuring, it is characterised in that:
2 groups of RGB observation signals in described step S101 select the high-pass filtering of 0.15Hz.
6. contactless humanbody breathing rate and heart rate synchronized measurement system, it is characterised in that: system includes that face video obtains
Take and generate and pretreatment module (20), multichannel blind source separating mould with sensitive region selection module (10), observation signal
Block (30), source signal identification and physiological parameter extraction module (40)；
Face video acquisition is used for shooting human body face video with sensitive region selection module (10), then from frame of video figure
Sensitizing range at two, respectively mouth region and bottleneck throat region is selected in Xiang；
Observation signal generation and pretreatment module (20) are used for the pixel value of the selected Lazer sensillary area to frame each in video,
Use coherent averaging technique respectively, generate 2 groups of RGB observation signals, then 2 groups of RGB observation signals are carried out 0.15Hz successively
High-pass filtering, go trend, go average, normalization etc. 4 step pretreatment operation to suppress interference noise, it is thus achieved that comparatively ideal
Observation signal is used for subsequent analysis；
Multichannel blind source separating module (30) is for carrying out 2 groups of RGB observation signals based on sensitive Area generation
6 passage blind source separating, to isolate breath signal and heart rate signal；
Source signal identification and physiological parameter extraction module (40) identify breathing letter for the source signal after blind source separating
Number and heart rate signal, in conjunction with sliding window algorithm extract breathing and heart rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610404234.1A CN105962915B (en) | 2016-06-02 | 2016-06-02 | Non-contact human body respiration rate and heart rate synchronous measurement method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610404234.1A CN105962915B (en) | 2016-06-02 | 2016-06-02 | Non-contact human body respiration rate and heart rate synchronous measurement method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105962915A true CN105962915A (en) | 2016-09-28 |
| CN105962915B CN105962915B (en) | 2019-04-05 |
Family
ID=57010353
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610404234.1A Active CN105962915B (en) | 2016-06-02 | 2016-06-02 | Non-contact human body respiration rate and heart rate synchronous measurement method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105962915B (en) |
Cited By (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106377226A (en) * | 2016-08-30 | 2017-02-08 | 苏州品诺维新医疗科技有限公司 | Cardiovascular state detection device and determining method for slope information |
| CN106580301A (en) * | 2016-12-21 | 2017-04-26 | 广州心与潮信息科技有限公司 | Physiological parameter monitoring method, device and hand-held device |
| CN106725410A (en) * | 2016-12-12 | 2017-05-31 | 努比亚技术有限公司 | A kind of heart rate detection method and terminal |
| CN107358220A (en) * | 2017-07-31 | 2017-11-17 | 江西中医药大学 | A kind of human heart rate and the contactless measurement of breathing |
| 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 |
| CN107967684A (en) * | 2017-12-11 | 2018-04-27 | 深圳大学 | Contactless sleep-respiratory detection method and device |
| CN109284672A (en) * | 2018-08-03 | 2019-01-29 | 金陵科技学院 | A kind of Mechanical Failure of HV Circuit Breaker diagnostic method based on PSO-Kmeans algorithm |
| CN109464147A (en) * | 2019-01-09 | 2019-03-15 | 浙江强脑科技有限公司 | Breath signal detection method, device and computer readable storage medium |
| CN109662727A (en) * | 2019-01-29 | 2019-04-23 | 广东工业大学 | Fetal heart rate detection method and system based on fuzzy clustering algorithm |
| GB2572961A (en) * | 2018-04-16 | 2019-10-23 | Clinicco Ltd | System for vital sign detection from a video stream |
| CN110367950A (en) * | 2019-07-22 | 2019-10-25 | 西安爱特眼动信息科技有限公司 | Contactless physiologic information detection method and system |
| CN111110211A (en) * | 2020-01-08 | 2020-05-08 | 中国农业大学 | Sow respiratory frequency and heart rate detection device and method |
| CN111444797A (en) * | 2020-03-16 | 2020-07-24 | 五邑大学 | Non-contact heart rate detection method |
| CN111568425A (en) * | 2020-06-08 | 2020-08-25 | 北京大学 | Non-contact type multi-person respiration detection method |
| CN113361526A (en) * | 2021-07-16 | 2021-09-07 | 合肥工业大学 | Non-contact respiration rate monitoring method fusing shoulder and chest area information |
| CN113693573A (en) * | 2021-08-27 | 2021-11-26 | 西安电子科技大学 | Video-based non-contact multi-physiological-parameter monitoring system and method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007136610A2 (en) * | 2006-05-17 | 2007-11-29 | University Of Hawaii | Determining presence and/or physiological motion of one or more subjects with multiple receiver doppler radar systems |
| CN101558997A (en) * | 2009-05-27 | 2009-10-21 | 江西蓝天学院 | Electroencephalograph signal identification method based on second-order blind identification |
| US20110251493A1 (en) * | 2010-03-22 | 2011-10-13 | Massachusetts Institute Of Technology | Method and system for measurement of physiological 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 |
| CN105046209A (en) * | 2015-06-30 | 2015-11-11 | 华侨大学 | Non-contact heart rate measurement method based on canonical correlation analysis |
| CN105147274A (en) * | 2015-08-04 | 2015-12-16 | 河北工业大学 | Method for extracting heart rate from visible spectrum section face video signal |
-
2016
- 2016-06-02 CN CN201610404234.1A patent/CN105962915B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007136610A2 (en) * | 2006-05-17 | 2007-11-29 | University Of Hawaii | Determining presence and/or physiological motion of one or more subjects with multiple receiver doppler radar systems |
| CN101558997A (en) * | 2009-05-27 | 2009-10-21 | 江西蓝天学院 | Electroencephalograph signal identification method based on second-order blind identification |
| US20110251493A1 (en) * | 2010-03-22 | 2011-10-13 | Massachusetts Institute Of Technology | Method and system for measurement of physiological 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 |
| CN105046209A (en) * | 2015-06-30 | 2015-11-11 | 华侨大学 | Non-contact heart rate measurement method based on canonical correlation analysis |
| CN105147274A (en) * | 2015-08-04 | 2015-12-16 | 河北工业大学 | Method for extracting heart rate from visible spectrum section face video signal |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106377226B (en) * | 2016-08-30 | 2019-05-03 | 苏州涵轩信息科技有限公司 | A kind of determination method of cardiovascular status detection device and slope information |
| CN106377226A (en) * | 2016-08-30 | 2017-02-08 | 苏州品诺维新医疗科技有限公司 | Cardiovascular state detection device and determining method for slope information |
| CN106725410A (en) * | 2016-12-12 | 2017-05-31 | 努比亚技术有限公司 | A kind of heart rate detection method and terminal |
| CN106580301A (en) * | 2016-12-21 | 2017-04-26 | 广州心与潮信息科技有限公司 | Physiological parameter monitoring method, device and hand-held device |
| CN107358220A (en) * | 2017-07-31 | 2017-11-17 | 江西中医药大学 | A kind of human heart rate and the contactless measurement of breathing |
| 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 |
| CN107967684A (en) * | 2017-12-11 | 2018-04-27 | 深圳大学 | Contactless sleep-respiratory detection method and device |
| GB2572961A (en) * | 2018-04-16 | 2019-10-23 | Clinicco Ltd | System for vital sign detection from a video stream |
| CN109284672A (en) * | 2018-08-03 | 2019-01-29 | 金陵科技学院 | A kind of Mechanical Failure of HV Circuit Breaker diagnostic method based on PSO-Kmeans algorithm |
| CN109464147A (en) * | 2019-01-09 | 2019-03-15 | 浙江强脑科技有限公司 | Breath signal detection method, device and computer readable storage medium |
| CN109662727A (en) * | 2019-01-29 | 2019-04-23 | 广东工业大学 | Fetal heart rate detection method and system based on fuzzy clustering algorithm |
| CN110367950A (en) * | 2019-07-22 | 2019-10-25 | 西安爱特眼动信息科技有限公司 | Contactless physiologic information detection method and system |
| CN111110211A (en) * | 2020-01-08 | 2020-05-08 | 中国农业大学 | Sow respiratory frequency and heart rate detection device and method |
| CN111444797A (en) * | 2020-03-16 | 2020-07-24 | 五邑大学 | Non-contact heart rate detection method |
| CN111444797B (en) * | 2020-03-16 | 2023-03-17 | 五邑大学 | Non-contact heart rate detection method |
| CN111568425A (en) * | 2020-06-08 | 2020-08-25 | 北京大学 | Non-contact type multi-person respiration detection method |
| CN111568425B (en) * | 2020-06-08 | 2021-08-03 | 北京大学 | Non-contact type multi-person respiration detection method |
| CN113361526A (en) * | 2021-07-16 | 2021-09-07 | 合肥工业大学 | Non-contact respiration rate monitoring method fusing shoulder and chest area information |
| CN113361526B (en) * | 2021-07-16 | 2022-09-27 | 合肥工业大学 | Non-contact respiration rate monitoring method fusing shoulder and chest area information |
| CN113693573A (en) * | 2021-08-27 | 2021-11-26 | 西安电子科技大学 | Video-based non-contact multi-physiological-parameter monitoring system and method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105962915B (en) | 2019-04-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105962915A (en) | Non-contact human body respiration rate and heart rate synchronous measurement method and system | |
| CN105147274B (en) | A kind of method that heart rate is extracted in the face video signal from visible spectrum | |
| JP6521845B2 (en) | Device and method for measuring periodic fluctuation linked to heart beat | |
| CN107529646B (en) | Non-contact heart rate measurement method and device based on Euler image amplification | |
| CN110619301B (en) | Emotion automatic identification method based on bimodal signals | |
| EP3697294B1 (en) | System and method for non-invasive hematological measurements | |
| Lewandowska et al. | Measuring pulse rate with a webcam | |
| CN109044322A (en) | A kind of contactless heart rate variability measurement method | |
| JP2015500722A (en) | Method and apparatus for detecting and quantifying skin symptoms in a skin zone | |
| JP5405994B2 (en) | Image processing apparatus, image processing method, image processing system, and skin evaluation method | |
| CN111259895B (en) | Emotion classification method and system based on facial blood flow distribution | |
| Qi et al. | Non-contact driver cardiac physiological monitoring using video data | |
| CN109350030A (en) | System and method based on phase enhanced processing face video heart rate signal | |
| Huang et al. | Emotion classification using deep neural networks and emotional patches | |
| CN111281367A (en) | Anti-interference non-contact heart rate detection method based on face video | |
| US11701015B2 (en) | Computer-implemented method and system for direct photoplethysmography (PPG) with multiple sensors | |
| Rajna et al. | Detection of short-term activity avalanches in human brain default mode network with ultrafast MR encephalography | |
| KR101436988B1 (en) | Method and Apparatus of Skin Pigmentation Detection Using Projection Transformed Block Coefficient | |
| CN111839492A (en) | Heart rate non-contact type measuring method based on face video sequence | |
| Kossack et al. | Local Remote Photoplethysmography Signal Analysis for Application in Presentation Attack Detection. | |
| Cui et al. | Non-contact time varying heart rate monitoring in exercise by video camera | |
| CN104573668B (en) | Facial color and luster adaptability automatic identifying method based on spectral reflectivity | |
| CN116091487B (en) | Skin wrinkle analysis comparison method and system based on image recognition | |
| CN112869737A (en) | Non-contact human body blood oxygen saturation detection method | |
| CN113591769B (en) | Non-contact heart rate detection method based on photoplethysmography |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |