CN109512416A - A kind of volume pulsation wave extracting method and system - Google Patents
A kind of volume pulsation wave extracting method and system Download PDFInfo
- Publication number
- CN109512416A CN109512416A CN201811388463.4A CN201811388463A CN109512416A CN 109512416 A CN109512416 A CN 109512416A CN 201811388463 A CN201811388463 A CN 201811388463A CN 109512416 A CN109512416 A CN 109512416A
- Authority
- CN
- China
- Prior art keywords
- signal
- skin area
- sef
- volume pulsation
- pulsation wave
- 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 66
- 230000010349 pulsation Effects 0.000 title claims abstract description 61
- 238000001914 filtration Methods 0.000 claims description 23
- 238000000605 extraction Methods 0.000 claims description 16
- 230000003044 adaptive effect Effects 0.000 claims description 12
- 238000010276 construction Methods 0.000 claims description 11
- 230000009466 transformation Effects 0.000 claims description 9
- 230000036387 respiratory rate Effects 0.000 claims description 7
- 229910052739 hydrogen Inorganic materials 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000005192 partition Methods 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 3
- 239000000284 extract Substances 0.000 abstract description 9
- 238000010586 diagram Methods 0.000 description 8
- 210000004369 blood Anatomy 0.000 description 4
- 239000008280 blood Substances 0.000 description 4
- 210000001519 tissue Anatomy 0.000 description 3
- 230000000694 effects Effects 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 210000005259 peripheral blood Anatomy 0.000 description 2
- 239000011886 peripheral blood Substances 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000017531 blood circulation Effects 0.000 description 1
- 230000002500 effect on skin Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 230000010247 heart contraction Effects 0.000 description 1
- 230000031700 light absorption Effects 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 210000003491 skin Anatomy 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
- A61B5/026—Measuring blood flow
- A61B5/0295—Measuring blood flow using plethysmography, i.e. measuring the variations in the volume of a body part as modified by the circulation of blood therethrough, e.g. impedance plethysmography
-
- 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
- A61B5/026—Measuring blood flow
- A61B5/0261—Measuring blood flow using optical means, e.g. infrared light
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Medical Informatics (AREA)
- Physics & Mathematics (AREA)
- Cardiology (AREA)
- Biophysics (AREA)
- Pathology (AREA)
- Hematology (AREA)
- Biomedical Technology (AREA)
- Heart & Thoracic Surgery (AREA)
- Physiology (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of volume pulsation wave extracting method and systems.This method comprises: obtaining human skin video;By in the image of video human body skin area and background area distinguish;The green area image pixel value of every width human body skin area is averaged to obtain skin area pixel sequence;The blue region image pixel value of every width background area is averaged to obtain background area pixels sequence;Using skin area pixel sequence as the desired signal of sef-adapting filter, using background area pixels sequence as input signal, the output signal and error signal at each moment are calculated;The coefficient vector of sef-adapting filter when seeking the root mean square minimum for making error signal using lowest mean square root method, obtains filtered skin area pixel sequence;Volume pulsation wave is extracted to filtered skin area pixel sequence using volume pulsation wave extractive technique.This method and system of the invention can effectively inhibit ambient light noise, improve the accuracy that volume pulsation wave extracts.
Description
Technical field
The present invention relates to medical domains, more particularly to a kind of volume pulsation wave extracting method and system.
Background technique
Volume pulsation wave can be extracted using camera under environment light condition, method particularly includes: by volume pulsation at
As (Photoplethysmographic imaging, PPGi) detects volumetric blood variation in living tissue.In living tissue
In, skin, muscle and tissue etc. are to maintain invariable, and the blood in skin to the absorption of light in entire blood circulation
Volume changes under heart contraction diastole effect in pulsating nature.Peripheral blood vessel blood volume is most when the heart contracts, absorbing amount
Also maximum, the light intensity of skin surface is also minimum;And peripheral blood vessel blood volume is minimum when diastole, absorbing amount is also minimum, skin
The light intensity on skin surface is also maximum.Camera passes through human body surface dermal imaging, brightness of image as light intensity is in pulsating nature variation
So that 2D signal is converted into one-dimensional signal to image pixel and is obtained with volume pulsation wave.
However since volume pulsation wave is extracted under environment light condition, include in the image of the skin surface of acquisition
There are the noises such as environment light, to cause greatly to interfere to the extraction of volume pulsation wave.
Summary of the invention
The object of the present invention is to provide a kind of volume pulsation wave extracting method and systems, can effectively inhibit environment light to make an uproar
Sound reduces the influence that ambient light noise extracts volume pulsation wave, improves the accuracy that volume pulsation wave extracts.
To achieve the above object, the present invention provides following schemes:
A kind of volume pulsation wave extracting method, comprising:
It obtains camera and human body keeps the human skin video acquired when pre-determined distance;
Using image partition method by the human body skin area in the image at each moment in the human skin video
It is distinguished with background area;
The green area image pixel value of the human body skin area described in every width is averaged, and skin area pixel sequence is obtained
Column;
The blue region image pixel value of the background area described in every width is averaged, and background area pixels sequence is obtained;
Using the skin area pixel sequence as the desired signal of sef-adapting filter, with the background area pixels sequence
The input signal as sef-adapting filter is arranged, the output signal and error corresponding with the output signal at each moment are calculated
Signal;
The coefficient of sef-adapting filter when seeking the root mean square minimum for making the error signal using lowest mean square root method
Vector obtains filtered skin area pixel sequence so as to adjust the coefficient vector of the sef-adapting filter;
Volume pulsation wave is extracted to the filtered skin area pixel sequence using volume pulsation wave extractive technique.
Optionally, the desired signal using the skin area pixel sequence as sef-adapting filter, with the back
Input signal of the scene area pixel sequence as sef-adapting filter calculates the output signal at each moment and believes with the output
Number corresponding error signal, specifically includes:
Calculate the formula of the output signal at each moment are as follows:
X ' (n)=xT(n)W(n)
The formula of error signal are as follows:
E (n)=d (n)-x ' (n)=d (n)-xT(n)W(n)
Wherein, x ' (n) is the output signal at n moment, and W (n) is the coefficient vector of adaptive-filtering, W (n)=[w1(n),w2
(n),…,wH(n)]T, x (n) is the input signal at n moment, x (n)=[u (n-1), u (n-2) ..., u (n-H)]T, H is adaptive
Answer filter coefficient quantity;D (n) is desired signal, and e (n) is the error signal at n moment.
Optionally, adaptive filter when seeking the root mean square minimum for making the error signal using lowest mean square root method
The coefficient vector of wave device obtains filtered skin area pixel sequence so as to adjust the coefficient vector of the sef-adapting filter
Column, specifically include:
Establish the square root equation of error signal:
J (n)=E [e (n)2]=E [(d (n)-xT(n)W(n))2]
J (n) is the root mean square of error signal;
To make square root equation obtain minimum value, W (n) need to be updated according to following formula:
W (n+1)=W (n)+1/2* μ e (n) x (n)
μ is step factor, 0 < μ <, 1/ λmax, wherein λmaxFor the maximum eigenvalue of input signal autocorrelation matrix;
Output signal corresponding to the coefficient vector of sef-adapting filter when using the root mean square minimum of error signal as
Filtered skin area pixel sequence.
Optionally, the filtered skin area pixel sequence is extracted using volume pulsation wave extractive technique described
After volume pulsation wave, further includes:
Fourier transformation is carried out to the filtered skin area pixel sequence, obtains transformed signal;
Peak value of the transformed signal within the scope of 0~0.5Hz is sought, respiratory rate is obtained;
Peak value of the transformed signal within the scope of 0.7~1.5Hz is sought, heart rate is obtained.
Invention additionally discloses a kind of volume pulsation wave extraction systems, comprising:
Module is obtained, keeps the human skin video acquired when pre-determined distance for obtaining camera and human body;
Divide module, for utilizing image partition method will be in the image at each moment in the human skin video
Human body skin area and background area distinguish;
First averages module, and the green area image pixel value for the human body skin area described in every width is averaging
Value, obtains skin area pixel sequence;
Second averages module, and the blue region image pixel value for the background area described in every width is averaged, obtained
To background area pixels sequence;
Adaptive-filtering constructs module, for the expectation letter using the skin area pixel sequence as sef-adapting filter
Number, using the background area pixels sequence as the input signal of sef-adapting filter, calculate each moment output signal and
Error signal corresponding with the output signal;
Filter parameter adjusts module, for seeking making the root mean square of the error signal minimum using lowest mean square root method
When the coefficient vector of sef-adapting filter obtain filtered skin so as to adjust the coefficient vector of the sef-adapting filter
Skin area pixel sequence;
Extraction module, for being extracted using volume pulsation wave extractive technique to the filtered skin area pixel sequence
Volume pulsation wave.
Optionally, the adaptive-filtering constructs module, specifically includes:
Output signal construction unit, the formula of the output signal for calculating each moment are as follows:
X ' (n)=xT(n)W(n)
Error signal construction unit, the formula for error signal are as follows:
E (n)=d (n)-x ' (n)=d (n)-xT(n)W(n)
Wherein, x ' (n) is the output signal at n moment, and W (n) is the coefficient vector of adaptive-filtering, W (n)=[w1(n),w2
(n),…,wH(n)]T, x (n) is the input signal at n moment, x (n)=[u (n-1), u (n-2) ..., u (n-H)]T, H is adaptive
Answer filter coefficient quantity;D (n) is desired signal, and e (n) is the error signal at n moment.
Optionally, the filter parameter adjusts module, specifically includes:
Square root equation construction unit, for establishing the square root equation of error signal:
J (n)=E [e (n)2]=E [(d (n)-xT(n)W(n))2]
J (n) is the root mean square of error signal;
To make square root equation obtain minimum value, W (n) need to be updated according to following formula:
W (n+1)=W (n)+1/2* μ e (n) x (n)
μ is step factor, 0 < μ <, 1/ λmax, wherein λmaxFor the maximum eigenvalue of input signal autocorrelation matrix;
Filter unit, corresponding to the coefficient vector of sef-adapting filter when for by the root mean square minimum of error signal
Output signal is as filtered skin area pixel sequence.
Optionally, the volume pulsation wave extraction system further include:
Fourier transformation module is obtained for carrying out Fourier transformation to the filtered skin area pixel sequence
Transformed signal;
Respiratory rate computing module is exhaled for seeking peak value of the transformed signal within the scope of 0~0.5Hz
Suction rate;
Rate calculation module obtains the heart for seeking peak value of the transformed signal within the scope of 0.7~1.5Hz
Rate.
The specific embodiment provided according to the present invention, the invention discloses following technical effects: presently disclosed volume
Pulse wave extracting method and system realize the noise remove of the image of skin area using adaptive filter method, effectively press down
Ambient light noise has been made, the influence that ambient light noise extracts volume pulsation wave has been reduced, improves the standard of volume pulsation wave extraction
Exactness.And it, can be according to ambient light noise real-time optimization by adjusting the parameter of sef-adapting filter in this method and system
The parameter of filter, the ability for making sef-adapting filter have self-teaching and tracking, substantially increases the real-time of filtering.
Detailed description of the invention
It in order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, below will be to institute in embodiment
Attached drawing to be used is needed to be briefly described, it should be apparent that, the accompanying drawings in the following description is only some implementations of the invention
Example, for those of ordinary skill in the art, without any creative labor, can also be according to these attached drawings
Obtain other attached drawings.
Fig. 1 is the method flow diagram of volume pulsation wave extracting method embodiment of the present invention.
Fig. 2 is the Image Acquisition schematic diagram of volume pulsation wave extracting method embodiment of the present invention.
Fig. 3 is the human body skin area and background area schematic diagram of volume pulsation wave extracting method embodiment of the present invention.
Fig. 4 is the schematic diagram of the adaptive-filtering of volume pulsation wave extracting method embodiment of the present invention.
Fig. 5 is the system construction drawing of volume pulsation wave extraction system embodiment of the present invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
The object of the present invention is to provide a kind of volume pulsation wave extracting method and systems, can effectively inhibit environment light to make an uproar
Sound reduces the influence that ambient light noise extracts volume pulsation wave, improves the accuracy that volume pulsation wave extracts.
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing and specific real
Applying mode, the present invention is described in further detail.
Fig. 1 is the method flow diagram of volume pulsation wave extracting method embodiment of the present invention.
A kind of volume pulsation wave extracting method, comprising:
Step 101: obtaining camera and human body keeps the human skin video acquired when pre-determined distance.The video of acquisition is
Color video.
It is illustrated for acquiring human face region image in the present invention.
Fig. 2 is the Image Acquisition schematic diagram of volume pulsation wave extracting method embodiment of the present invention.
Referring to fig. 2, using extraction human face region image in camera and 0.5 meter~1 meter distance range of face distance.Every width
Both include facial image in image, also includes background image.
Step 102: using image partition method by the human body in the image at each moment in the human skin video
Skin area and background area distinguish.
Fig. 3 is the human body skin area and background area schematic diagram of volume pulsation wave extracting method embodiment of the present invention.
Referring to Fig. 3, a-quadrant is human body skin area, i.e. human face region, and B area is background area.
Step 103: the green area image pixel value of the human body skin area described in every width is averaged, and skin region is obtained
Domain pixel sequence.
Step 104: the blue region image pixel value of the background area described in every width is averaged, and background area picture is obtained
Prime sequences.
One width color image may be considered the superposition of red green and yellow image, so in human body skin area
Green area image pixel value average and be converted into a numerical value, when the video of such human body skin area can be converted into
Between signal.Similarly, time signal is also converted to the image/video of background area, difference is not to be to be turned with green image
It changes, but is converted with blue image.
Step 105: using the skin area pixel sequence as the desired signal of sef-adapting filter, with the background area
Input signal of the domain pixel sequence as sef-adapting filter, calculate each moment output signal and with the output signal pair
The error signal answered.
Fig. 4 is the schematic diagram of the adaptive-filtering of volume pulsation wave extracting method embodiment of the present invention.
Referring to fig. 4, the purpose of adaptive-filtering is really and accurately desired signal to be extracted from output signal, therefore be related to
To signal include desired signal, input signal and output signal.D (n) is desired signal, and representative includes ambient light noise
Skin area pixel sequence, x (n) are the input signal at n moment, indicate background area pixels sequence.According to background in the present invention
Area pixel sequence determines the size of the ambient light noise in skin area pixel sequence, x (n)=[u (n-1), u (n-
2),…,u(n-H)]T, W (n) is the coefficient vector of adaptive-filtering, W (n)=[w1(n),w2(n),…,wH(n)]T, H is adaptive
Answer filter coefficient quantity.U (n-1), u (n-2) ..., u (n-H) are the input value at each moment;w1(n),w2(n) ..., wH
It (n) is each coefficient in the coefficient vector of adaptive-filtering.
The formula of the output signal at each moment are as follows:
X ' (n)=xT(n)W(n)
The formula of error signal are as follows:
E (n)=d (n)-x ' (n)=d (n)-xT(n)W(n)
Wherein, x ' (n) is the output signal at n moment, and e (n) is the error signal at n moment, when the coefficient of adaptive-filtering
After vector is adjusted, e (n) is the filtered skin area pixel sequence extracted.
Step 106: the adaptive-filtering when root mean square minimum for making the error signal is sought using lowest mean square root method
The coefficient vector of device obtains filtered skin area pixel sequence so as to adjust the coefficient vector of the sef-adapting filter.
The variation of the coefficient vector W (n) of adaptive digital filter is related to error signal e (n), big according to the value of e (n)
Small and adjust automatically W (n) is allowed to be suitble to the input u (n+1) of subsequent time, to make to export x ' (n+1) close to desired
Reference signal.There are many method for adjusting W (n), are such as based on lowest mean square method for root, least square method, transform domain method, grip ladder altogether
Degree method etc..The present invention uses lowest mean square method for root:
Lowest mean square root adaptive filter algorithm principle is to make filtering signal and expectation by adjusting filter coefficient W (n)
The mean-square value of error is minimum between signal.The calculating of filter coefficient W (n) raises entire filter wave in gradient direction by error amount
Device coefficient, and fast convergence makes the mean-square value of error for minimum.Its algorithm characteristic is that complexity is low, does not need largely to count
It calculates, and convergence rate block.Lowest mean square root method objective function is E [e (n)2] when reaching minimum, that is, realize optimal filter
Wave.The square root equation of the error signal of foundation are as follows:
J (n)=E [e (n)2]=E [(d (n)-xT(n)W(n))2]
J (n) is the root mean square of error signal;
It in order to obtain the mean-square value J (n) of minimal error, optimizes, leads on gradient direction in filter coefficient W (n)
Update W (n) is crossed, reaches fast convergence, it is as follows that W (n) optimizes formula:
W (n+1)=W (n)+1/2* μ e (n) x (n)
In formula, μ is step factor.
The condition of lowest mean square root algorithmic statement are as follows: 0 < μ <, 1/ λmax, λmaxIt is the maximum of input signal autocorrelation matrix
Characteristic value.
After determining coefficient vector W (n), the coefficient vector of sef-adapting filter when by the root mean square minimum of error signal
Corresponding output signal is as filtered skin area pixel sequence.
Step 107: volume being extracted to the filtered skin area pixel sequence using volume pulsation wave extractive technique
Pulse wave.
Optionally, after step 107, further includes:
Fourier transformation is carried out to the filtered skin area pixel sequence, obtains transformed signal;
Peak value of the transformed signal within the scope of 0~0.5Hz is sought, respiratory rate is obtained;
Peak value of the transformed signal within the scope of 0.7~1.5Hz is sought, heart rate is obtained.
Fig. 5 is the system construction drawing of volume pulsation wave extraction system embodiment of the present invention.
Referring to Fig. 5, the volume pulsation wave extraction system, comprising:
Module 501 is obtained, keeps the human skin video acquired when pre-determined distance for obtaining camera and human body;
Divide module 502, for utilizing image partition method by the image at each moment in the human skin video
In human body skin area and background area distinguish;
First averages module 503, and the green area image pixel value for the human body skin area described in every width asks flat
Mean value obtains skin area pixel sequence;
Second averages module 504, and the blue region image pixel value for the background area described in every width is averaged,
Obtain background area pixels sequence;
Adaptive-filtering constructs module 505, for the phase using the skin area pixel sequence as sef-adapting filter
It hopes signal, using the background area pixels sequence as the input signal of sef-adapting filter, calculates the output letter at each moment
Number and error signal corresponding with the output signal.
The adaptive-filtering constructs module 505, specifically includes:
Output signal construction unit, the formula of the output signal for calculating each moment are as follows:
X ' (n)=xT(n)W(n)
Error signal construction unit, the formula for error signal are as follows:
E (n)=d (n)-x ' (n)=d (n)-xT(n)W(n)
Wherein, x ' (n) is the output signal at n moment, and W (n) is the coefficient vector of adaptive-filtering, W (n)=[w1(n),w2
(n),…,wH(n)]T, x (n) is the input signal at n moment, x (n)=[u (n-1), u (n-2) ..., u (n-H)]T, H is adaptive
Answer filter coefficient quantity;E (n) is the error signal at n moment.
Filter parameter adjusts module 506, the root mean square for seeking making the error signal using lowest mean square root method
The coefficient vector of sef-adapting filter when minimum, so as to adjust the coefficient vector of the sef-adapting filter, after obtaining filtering
Skin area pixel sequence.
The filter parameter adjusts module 506, specifically includes:
Square root equation construction unit, for establishing the square root equation of error signal:
J (n)=E [e (n)2]=E [(d (n)-xT(n)W(n))2]
J (n) is the root mean square of error signal;
To make square root equation obtain minimum value, W (n) need to be updated according to following formula:
W (n+1)=w (n)+1/2* μ e (n) x (n)
μ is step factor, 0 < μ <, 1/ λmax, wherein λmaxFor the maximum eigenvalue of input signal autocorrelation matrix;
Filter unit, corresponding to the coefficient vector of sef-adapting filter when for by the root mean square minimum of error signal
Output signal is as filtered skin area pixel sequence.
Extraction module 507, for utilizing volume pulsation wave extractive technique to the filtered skin area pixel sequence
Extract volume pulsation wave.
The volume pulsation wave extraction system further include:
Fourier transformation module is obtained for carrying out Fourier transformation to the filtered skin area pixel sequence
Transformed signal;
Respiratory rate computing module is exhaled for seeking peak value of the transformed signal within the scope of 0~0.5Hz
Suction rate;
Rate calculation module obtains the heart for seeking peak value of the transformed signal within the scope of 0.7~1.5Hz
Rate.
Presently disclosed volume pulsation wave extracting method and system, realize skin area using adaptive filter method
Image noise remove, effectively inhibit ambient light noise, reduce the influence that ambient light noise extracts volume pulsation wave, mention
The accuracy that high volume pulsation wave extracts.And it, can by adjusting the parameter of sef-adapting filter in this method and system
According to the parameter of ambient light noise real-time optimization filter, the ability for making sef-adapting filter that there is self-teaching and tracking, greatly
The real-time of filtering is improved greatly.
For the system disclosed in the embodiment, since it is corresponded to the methods disclosed in the examples, so the ratio of description
Relatively simple, reference may be made to the description of the method.
Used herein a specific example illustrates the principle and implementation of the invention, and above embodiments are said
It is bright to be merely used to help understand method and its core concept of the invention;At the same time, for those skilled in the art, foundation
Thought of the invention, there will be changes in the specific implementation manner and application range.In conclusion the content of the present specification is not
It is interpreted as limitation of the present invention.
Claims (8)
1. a kind of volume pulsation wave extracting method characterized by comprising
It obtains camera and human body keeps the human skin video acquired when pre-determined distance;
Using image partition method by the human body skin area and back in the image at each moment in the human skin video
Scene area distinguishes;
The green area image pixel value of the human body skin area described in every width is averaged, and skin area pixel sequence is obtained;
The blue region image pixel value of the background area described in every width is averaged, and background area pixels sequence is obtained;
Using the skin area pixel sequence as the desired signal of sef-adapting filter, with background area pixels sequence work
For the input signal of sef-adapting filter, the output signal and error corresponding with output signal letter at each moment are calculated
Number;
The coefficient vector of sef-adapting filter when seeking the root mean square minimum for making the error signal using lowest mean square root method,
So as to adjust the coefficient vector of the sef-adapting filter, filtered skin area pixel sequence is obtained;
Volume pulsation wave is extracted to the filtered skin area pixel sequence using volume pulsation wave extractive technique.
2. a kind of volume pulsation wave extracting method according to claim 1, which is characterized in that described with the skin area
Pixel sequence as the desired signal of sef-adapting filter, using the background area pixels sequence as the defeated of sef-adapting filter
Enter signal, calculate the output signal and error signal corresponding with the output signal at each moment, specifically include:
Calculate the formula of the output signal at each moment are as follows:
X ' (n)=xT(n)W(n)
The formula of error signal are as follows:
E (n)=d (n)-x ' (n)=d (n)-xT(n)W(n)
Wherein, x ' (n) is the output signal at n moment, and W (n) is the coefficient vector of adaptive-filtering, W (n)=[w1(n),w2
(n),…,wH(n)]T, x (n) is the input signal at n moment, x (n)=[u (n-1), u (n-2) ..., u (n-H)]T, H is adaptive
Answer filter coefficient quantity;D (n) is desired signal, and e (n) is the error signal at n moment.
3. a kind of volume pulsation wave extracting method according to claim 2, which is characterized in that described to utilize lowest mean square root
The coefficient vector of sef-adapting filter when method seeks the root mean square minimum for making the error signal, so as to adjust described adaptive
The coefficient vector of filter obtains filtered skin area pixel sequence, specifically includes:
Establish the square root equation of error signal:
J (n)=E [e (n)2]=E [(d (n)-xT(n)W(n))2]
J (n) is the root mean square of error signal;
To make square root equation obtain minimum value, W (n) need to be updated according to following formula:
W (n+1)=W (n)+1/2* μ e (n) x (n)
μ is step factor, 0 < μ <, 1/ λmax, wherein λmaxFor the maximum eigenvalue of input signal autocorrelation matrix;
Output signal corresponding to the coefficient vector of sef-adapting filter when using the root mean square minimum of error signal is as filtering
Skin area pixel sequence afterwards.
4. a kind of volume pulsation wave extracting method according to claim 1, which is characterized in that utilize volume pulsation described
Wave extractive technique is extracted the filtered skin area pixel sequence after volume pulsation wave, further includes:
Fourier transformation is carried out to the filtered skin area pixel sequence, obtains transformed signal;
Peak value of the transformed signal within the scope of 0~0.5Hz is sought, respiratory rate is obtained;
Peak value of the transformed signal within the scope of 0.7~1.5Hz is sought, heart rate is obtained.
5. a kind of volume pulsation wave extraction system characterized by comprising
Module is obtained, keeps the human skin video acquired when pre-determined distance for obtaining camera and human body;
Divide module, for utilizing image partition method by the human body in the image at each moment in the human skin video
Skin area and background area distinguish;
First averages module, and the green area image pixel value for the human body skin area described in every width is averaged, obtained
To skin area pixel sequence;
Second averages module, and the blue region image pixel value for the background area described in every width is averaged, carried on the back
Scene area pixel sequence;
Adaptive-filtering construct module, for using the skin area pixel sequence as sef-adapting filter desired signal,
Using the background area pixels sequence as the input signal of sef-adapting filter, calculate each moment output signal and with institute
State the corresponding error signal of output signal;
Filter parameter adjusts module, when for seeking the root mean square minimum for making the error signal using lowest mean square root method
The coefficient vector of sef-adapting filter obtains filtered skin region so as to adjust the coefficient vector of the sef-adapting filter
Domain pixel sequence;
Extraction module, for extracting volume to the filtered skin area pixel sequence using volume pulsation wave extractive technique
Pulse wave.
6. a kind of volume pulsation wave extraction system according to claim 5, which is characterized in that the adaptive-filtering building
Module specifically includes:
Output signal construction unit, the formula of the output signal for calculating each moment are as follows:
X ' (n)=xT(n)W(n)
Error signal construction unit, the formula for error signal are as follows:
E (n)=d (n)-x ' (n)=d (n)-xT(n)W(n)
Wherein, x ' (n) is the output signal at n moment, and W (n) is the coefficient vector of adaptive-filtering, W (n)=[w1(n),w2
(n),…,wH(n)]T, x (n) is the input signal at n moment, x (n)=[u (n-1), u (n-2) ..., u (n-H)]T, H is adaptive
Answer filter coefficient quantity;D (n) is desired signal, and e (n) is the error signal at n moment.
7. a kind of volume pulsation wave extraction system according to claim 6, which is characterized in that the filter parameter adjustment
Module specifically includes:
Square root equation construction unit, for establishing the square root equation of error signal:
J (n)=E [e (n)2]=E [(d (n)-xT(n)W(n))2]
J (n) is the root mean square of error signal;
To make square root equation obtain minimum value, W (n) need to be updated according to following formula:
W (n+1)=W (n)+1/2* μ e (n) x (n)
μ is step factor, 0 < μ <, 1/ λmax, wherein λmaxFor the maximum eigenvalue of input signal autocorrelation matrix;
Filter unit, output corresponding to the coefficient vector of sef-adapting filter when for by the root mean square minimum of error signal
Signal is as filtered skin area pixel sequence.
8. a kind of volume pulsation wave extraction system according to claim 5, which is characterized in that further include:
Fourier transformation module is converted for carrying out Fourier transformation to the filtered skin area pixel sequence
Signal afterwards;
Respiratory rate computing module obtains respiratory rate for seeking peak value of the transformed signal within the scope of 0~0.5Hz;
Rate calculation module obtains heart rate for seeking peak value of the transformed signal within the scope of 0.7~1.5Hz.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811388463.4A CN109512416A (en) | 2018-11-21 | 2018-11-21 | A kind of volume pulsation wave extracting method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811388463.4A CN109512416A (en) | 2018-11-21 | 2018-11-21 | A kind of volume pulsation wave extracting method and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109512416A true CN109512416A (en) | 2019-03-26 |
Family
ID=65778225
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811388463.4A Pending CN109512416A (en) | 2018-11-21 | 2018-11-21 | A kind of volume pulsation wave extracting method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109512416A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113397497A (en) * | 2021-06-10 | 2021-09-17 | 维沃移动通信有限公司 | Signal processing method and device and electronic equipment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8185357B1 (en) * | 2002-08-23 | 2012-05-22 | The United States Of America As Represented By The Secretary Of The Navy | Nonlinear blind demixing of single pixel underlying radiation sources and digital spectrum local thermometer |
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 |
CN107334469A (en) * | 2017-07-24 | 2017-11-10 | 北京理工大学 | Non-contact more people's method for measuring heart rate and device based on SVMs |
CN108652609A (en) * | 2018-05-17 | 2018-10-16 | 歌尔科技有限公司 | A kind of heart rate acquisition methods, system and wearable device |
-
2018
- 2018-11-21 CN CN201811388463.4A patent/CN109512416A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US8185357B1 (en) * | 2002-08-23 | 2012-05-22 | The United States Of America As Represented By The Secretary Of The Navy | Nonlinear blind demixing of single pixel underlying radiation sources and digital spectrum local thermometer |
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 |
CN107334469A (en) * | 2017-07-24 | 2017-11-10 | 北京理工大学 | Non-contact more people's method for measuring heart rate and device based on SVMs |
CN108652609A (en) * | 2018-05-17 | 2018-10-16 | 歌尔科技有限公司 | A kind of heart rate acquisition methods, system and wearable device |
Non-Patent Citations (3)
Title |
---|
XIAOBAI LI 等: "Remote Heart Rate Measurement From Face Videos Under Realistic Situations", 《2014 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION》 * |
刘赫: "容积脉搏成像特征提取方法研究及在生理信号检测的应用", 《中国博士学位论文全文数据库 信息科技辑》 * |
王亿勤 等: "《中医诊断学研究思路与方法》", 30 September 2008 * |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113397497A (en) * | 2021-06-10 | 2021-09-17 | 维沃移动通信有限公司 | Signal processing method and device and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP5984841B2 (en) | Apparatus and method for extracting information from characteristic signals | |
EP3405105B1 (en) | Method and apparatus for estimating heart rate | |
CN103702014B (en) | Non-contact physiological parameter detection method, system and device | |
JP2016221264A5 (en) | ||
CN105989357A (en) | Human face video processing-based heart rate detection method | |
CN112233813A (en) | Non-contact non-invasive heart rate and respiration measurement method and system based on PPG | |
WO2014091977A1 (en) | Image-processing device and program | |
CN111134650A (en) | Heart rate information acquisition method and device, computer equipment and storage medium | |
EP2760338A1 (en) | Anisotropic processing of laser speckle images | |
DE102011056489A1 (en) | Adaptive time domain filtering to improve blood pressure estimation | |
JP6784403B2 (en) | Heart rate variability estimation method, heart rate variability estimation program and heart rate variability estimation system | |
Van Doorn et al. | Detectability of velocity gradients in moving random-dot patterns | |
CN109782414A (en) | A kind of automatic focusing method based on no reference configuration clarity | |
US20180070887A1 (en) | Blood pressure measuring device and blood pressure measuring method using the same | |
KR101798228B1 (en) | Pulse rate measuring method using image | |
CN106974617A (en) | The Signal Pre-Processing Method and signal wave crest detection method of a kind of efficiently and accurately | |
CN109512418A (en) | A kind of volumetric blood flow pulse imaging reduction motion artifacts method | |
CN109512416A (en) | A kind of volume pulsation wave extracting method and system | |
CN114693653A (en) | Method for accurately extracting pulse wave based on human superficial artery video | |
CN105809633B (en) | Remove the method and device of color noise | |
KR102284671B1 (en) | Method and apparatus for measuring blood pressure using skin images | |
US9943289B2 (en) | Color flow ultrasound imaging | |
KR102082292B1 (en) | Camera based photoplethysmogram estimation | |
CN111513701A (en) | Heart rate detection method and device, computer equipment and readable storage medium | |
CN114037626A (en) | Blood vessel imaging method, device, equipment and storage medium |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190326 |