CN111374646A - Non-hardware-dependent pulse condition information acquisition system and method based on smart phone - Google Patents
Non-hardware-dependent pulse condition information acquisition system and method based on smart phone Download PDFInfo
- Publication number
- CN111374646A CN111374646A CN201811628312.1A CN201811628312A CN111374646A CN 111374646 A CN111374646 A CN 111374646A CN 201811628312 A CN201811628312 A CN 201811628312A CN 111374646 A CN111374646 A CN 111374646A
- Authority
- CN
- China
- Prior art keywords
- pulse
- signal
- video
- signals
- period
- 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
Images
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
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0062—Arrangements for scanning
- A61B5/0064—Body surface scanning
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/0059—Measuring for diagnostic purposes; Identification of persons using light, e.g. diagnosis by transillumination, diascopy, fluorescence
- A61B5/0077—Devices for viewing the surface of the body, e.g. camera, magnifying lens
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4854—Diagnosis based on concepts of traditional oriental medicine
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/68—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient
- A61B5/6887—Arrangements of detecting, measuring or recording means, e.g. sensors, in relation to patient mounted on external non-worn devices, e.g. non-medical devices
- A61B5/6898—Portable consumer electronic devices, e.g. music players, telephones, tablet computers
Abstract
The invention discloses a hardware-independent pulse condition acquisition system based on a smart phone, which comprises: a pulse condition acquisition part and a pulse condition information analysis and extraction part; the acquisition part acquires pulse signals by utilizing a camera of the mobile phone to be attached to skin to shoot videos, wherein the shot positions comprise two positions of a wrist radial artery and a finger tip of an index finger; the analysis and extraction of the information acquisition comprises the steps of extracting and decomposing pulse signals from a shot video, and fusing processed waveforms and signals; the pulse signal acquisition device can realize the same pulse signal acquisition function of a common pulse diagnosis instrument without depending on hardware facilities such as a special pressurization system, a pressure sensor and the like, has an algorithm for processing and analyzing waveforms, improves the portability of pulse diagnosis operation and obtains more perfect pulse signals.
Description
Technical Field
The invention relates to a system and a method for acquiring hardware-independent pulse condition information based on a smart phone.
Background
The pulse diagnosis of traditional Chinese medicine is well known in China, and is called as the four diagnostic methods of traditional Chinese medicine in combination with inspection, inquiry and auscultation. The pulse diagnosis utilizes the pressure exerted by fingers on the radial artery of the human body to sense and detect the characteristics of the pulse, such as position, number, shape, potential and the like, and judges the functional state of the viscera through comprehensive analysis, thereby providing effective information for the diagnosis of the traditional Chinese medicine.
The pulse diagnosis in traditional Chinese medicine has certain subjectivity and is well known in the theory of 'easy to get in the heart and difficult to get in the lower limbs', so that the objectivity of the pulse diagnosis in traditional Chinese medicine is a research hotspot in the modernization of traditional Chinese medicine. The pulse condition instrument is a device which utilizes the modern measurement technology and simulates the traditional pulse feeling method to obtain the quantitative data of the pulse condition and analyze the data. Scholars in different fields develop various pulse instruments with different performances by using various technologies and methods, including MX-3C type, MX-811 type, ZM-III type, MXY-1 type, BYS-14 type four-lead pulse instruments, MTYA type pulse instrument, YGJ physician multifunctional syndrome differentiation instrument (integrating the functions of the pulse instruments) and the like. Chen et al (2015) proposed the use of linear laser Positioning and CMOS image Sensors to acquire Pulse signals (Chen Y, Chang R S, Jwo K W, et al. A Non-Contact Pulse Automatic Positioning measuring System for a proportional Chinese Medicine [ J ]. Sensors,2015,15(5): 9899. 9914.)
Disclosure of Invention
The invention provides a hardware-independent pulse condition information acquisition system based on a smart phone. The current very common smart phone (containing a camera and a flash lamp) is used for replacing a signal sensor, and hardware facilities such as a special pressurizing system, a pressure sensor and the like are not needed. The method comprises the steps of shooting and recording videos of the blood vessel beating process of two different positions of a radial artery and a finger tip of an index finger by using a camera and a flash lamp of a mobile phone, extracting and analyzing the videos by using a computer or an intelligent mobile phone and combining a matching algorithm provided by the inventor, and calculating and obtaining corresponding pulse condition information. The invention specifically comprises the following steps: the system comprises a video acquisition hardware platform, a pulse condition information acquisition part and a video analysis and extraction pulse condition information processing part.
According to one aspect of the invention, a pulse condition information acquisition system based on a smart phone is provided, which comprises:
the pulse condition acquisition part comprises:
the intelligent mobile phone camera is used for being attached to the body part of the pulse condition video to be shot, starting the flash lamp and the camera to shoot so as to obtain the video,
a video control part, when the mobile phone camera is tightly attached to the skin, the flash lamp is turned on and the brightness of the flash lamp is controlled, and simultaneously the video function is turned on,
a pulse information video analyzing and extracting part for processing the signals of the first video shot from the first part and the second video shot from the second part, filtering the noise, fusing the data and the like to obtain a single-cycle complete waveform,
the pulse condition information analyzing and extracting part comprises:
a signal extracting section for decomposing the first video and the second video into a series of frame pictures by frame, and analyzing R-channel information of each frame picture to obtain an unprocessed pulse signal containing a plurality of pulse periods and containing noise, an
A section for performing subsequent signal processing, comprising:
the pulse signals are screened according to the statistical information such as the amplitude intensity and the like of the pulse signals so as to be decomposed into a part of a complete signal set of each period,
a linear function is applied to the signal waveform of each cycle and normalized to eliminate a portion of the influence of zero point drift,
linear interpolation is performed on the processed signals of each period to eliminate discrete signals and portions of the signals of which the periods are not exactly the same in magnitude,
and the complete waveform acquisition part is used for calculating the mean variance of sample values at different sites of the waveform, performing linear weighted combination on signals with different shooting positions, namely different pulse information sources according to the minimum variance principle, and finally obtaining a complete waveform of a single pulse period which is subjected to signal processing such as denoising and the like.
According to another aspect of the invention, a pulse condition information acquisition method based on a smart phone is provided, which is characterized by comprising the following steps:
the pulse condition acquisition step comprises:
the camera of the intelligent mobile phone is used for clinging to the body part of the pulse condition video to be shot, the flash lamp and the camera are started for shooting, so as to obtain the video,
a video recording control step, when the mobile phone camera is tightly attached to the skin, the flash lamp is turned on, the brightness of the flash lamp is controlled, and the video recording function is turned on at the same time,
a pulse condition information video analysis and extraction step, signal processing is carried out on a first video shot from a first part and a second video shot from a second part, a single-cycle complete waveform is obtained through noise filtering, data fusion and the like,
the pulse condition information analyzing and extracting steps comprise:
performing signal extraction from the video, including decomposing the first video and the second video by frames to obtain a series of frame pictures, analyzing R channel information of each frame picture to obtain unprocessed pulse signal containing a plurality of pulse periods and containing noise, and
performing subsequent signal processing, including:
the pulse signals are screened according to the statistical information such as the amplitude intensity and the like of the pulse signals so as to be decomposed into a complete signal set of each period,
a linear function is applied to the signal waveform of each cycle and normalized to eliminate the influence of zero point drift,
a step of performing linear interpolation on the processed signals of each period to eliminate the discrete signals and the influence of different periods of different sizes,
and a complete waveform obtaining step, which comprises the steps of calculating the mean variance of sample values of different sites of the waveform, performing linear weighted combination on signals with different shooting positions, namely different pulse information sources according to the minimum variance principle, and finally obtaining a complete waveform of a single pulse period which is subjected to signal processing such as denoising and the like.
Drawings
Fig. 1 is a flowchart of a pulse condition information collecting method based on a smart phone according to an embodiment of the present invention.
Fig. 2(a) and 2(b) are views of a video recording operation in a pulse taking process by using the scheme of the invention, wherein fig. 2(a) is an operation example of obtaining a video at the radial artery of the wrist, and fig. 2(b) is an operation example of a video of the fingertip of the index finger.
FIG. 3 shows the pulse waveform at the radial artery of the wrist without treatment.
FIG. 4 shows the pulse waveform at the tip of the untreated index finger.
Fig. 5 shows a pulse waveform of one cycle obtained by analyzing the algorithm of the present invention.
Fig. 6 shows the pulse waveform measured by a commercially available pulse instrument for the same person.
Detailed Description
One of the core technical requirements of the pulse condition instrument is accurate and convenient pulse condition information acquisition. The existing pulse condition instruments generally need a special pressurizing system (simulating the pressure of fingers during doctor pulse diagnosis), and have the problems of difficult operation, poor portability and the like during operation. In addition, the number of sensors adopted by the conventional pulse condition instruments is limited, and the pulse condition information is usually acquired only by using a single sensor at the wrist, so that the acquired pulse condition information is limited. Another existing defect is poor popularity, and if a common user buys and carries a special pulse instrument, the use and carrying are inconvenient.
In order to overcome the problems in the prior art, the invention provides a pulse condition information acquisition system based on a smart phone, which replaces a signal sensor with a very common smart phone (comprising a camera and a flash lamp) at present, and does not need a special pressurization system, a pressure sensor and other hardware facilities. The method comprises the steps of shooting and recording videos of the blood vessel beating process of two different positions of a radial artery and a finger tip of an index finger by using a camera and a flash lamp of a mobile phone, extracting and analyzing the videos by using a computer or an intelligent mobile phone and combining a matching algorithm provided by the inventor, and calculating and obtaining corresponding pulse condition information.
The invention aims to realize a convenient and efficient pulse condition information acquisition method based on a smart phone, which is convenient and efficient by only using a camera and a flash lamp of the smart phone without other equipment. The method can be applied to home pulse diagnosis and visualization and quantitative analysis of the traditional Chinese medicine pulse diagnosis process.
The pulse condition information acquisition system based on the intelligent mobile phone comprises the following components:
the pulse condition acquisition part comprises:
the smart phone camera is used for clinging to the body part (such as the skin at the radial artery and the finger tip of the index finger) of the pulse condition video to be shot, starting the flash lamp and the camera to shoot so as to obtain the video,
a video control part, when the mobile phone camera is tightly attached to the skin, the flash lamp is turned on and the brightness of the flash lamp is controlled, and simultaneously the video function is turned on,
a pulse condition information video analyzing and extracting part, which processes the signal of the first video shot from the index finger tip and the second video shot from the radial artery, obtains the single-cycle complete waveform through noise filtering, data fusion and the like,
the pulse condition information analyzing and extracting part comprises:
a signal extracting section for decomposing the first video and the second video into a series of frame pictures by frame, and analyzing R-channel information of each frame picture to obtain an unprocessed pulse signal containing a plurality of pulse periods and containing noise, an
A section for performing subsequent signal processing, comprising:
the pulse signals are screened according to the statistical information such as the amplitude intensity and the like of the pulse signals so as to be decomposed into a part of a complete signal set of each period,
a linear function is applied to the signal waveform of each cycle and normalized to eliminate a portion of the influence of zero point drift,
linear interpolation is performed on the processed signals of each period to eliminate discrete signals and portions of the signals of which the periods are not exactly the same in magnitude,
and the complete waveform acquisition part is used for calculating the mean variance of sample values at different sites of the waveform, performing linear weighted combination on signals with different shooting positions, namely different pulse information sources according to the minimum variance principle, and finally obtaining a complete waveform of a single pulse period which is subjected to signal processing such as denoising and the like.
The advantages of the invention include:
1. the invention only uses the camera and the flash lamp of the mobile phone as the equipment without other complicated pressurizing devices and special display systems. The method has the advantages of low cost for users, convenient use and good portability, and can be used for collecting and analyzing the pulse condition information of the traditional Chinese medicine anytime and anywhere.
2. Only the camera of the mobile phone is required to be respectively attached to the finger tip of the index finger and the area near the skin of the radial artery, and the complicated pulse taking manipulation and process of the traditional Chinese medicine do not need to be known. The professional requirement of pulse condition collection is reduced, so that ordinary people without training can finish shooting and recording the video.
3. The method is favorable for objectively summarizing the Chinese medical knowledge and promotes the standardized and objective research and development of the Chinese medical pulse diagnosis;
4. the pulse condition data acquisition is carried out by adopting a common smart phone, and then the computer or the smart phone is used for carrying out video analysis to extract pulse condition information, so that the home pulse diagnosis is very easy to realize;
5. the invention obtains rich pulse condition information through a video processing technology.
6. The invention creatively collects pulse signals at two positions of the radial artery and the finger tip of the index finger and comprehensively analyzes the signals from two different sources, thereby achieving the accuracy which cannot be achieved by the common video shooting method. More accurate analysis of the pulse condition signal is possible. Compared with the pulse diagnosis instrument (only collecting signals of one radial artery or one finger) on the market at present, more information details can be reserved, and the possibility is provided for the subsequent possible automatic analysis and the pulse condition information discrimination.
7. The implementation process of the invention can realize the simulation of the three pulse methods of traditional Chinese medicine pulse diagnosis, namely floating and sinking, and obtains the characteristic information of pulse signals under different forces through the strength of the camera which is tightly attached to the skin at the radial artery.
As shown in fig. 1, according to another aspect of the present invention, there is provided a pulse condition information collecting method based on a smart phone, including:
the pulse condition acquisition step comprises:
when the camera of the smart phone is attached to a body part (such as skin at the radial artery and the tip of the index finger) to be used for shooting the pulse condition video, the flash lamp and the camera are started to shoot so as to obtain the video,
when the camera of the mobile phone is tightly attached to the skin, the flash lamp is turned on, the brightness of the flash lamp is controlled, the video recording function is turned on at the same time,
performing time domain analysis, performing time domain analysis on the videos acquired at the two positions, performing comprehensive analysis and judgment processing on the videos shot at the two positions, and extracting pulse signals from the videos, wherein one pulse condition information acquisition process comprises shooting at two different positions of a radial artery of a wrist and a finger tip of an index finger once respectively to obtain the videos containing the pulse condition signals at the two positions,
the pulse condition information video analysis and extraction comprises the following steps:
performing signal extraction from the video, decomposing the first video and the second video by frames to obtain a series of frame pictures, analyzing R channel information of each frame picture to obtain an unprocessed pulse signal containing a plurality of pulse periods and noise, and
performing subsequent signal processing, including:
the pulse signals are screened according to the statistical information such as the amplitude intensity and the like of the pulse signals so as to be decomposed into a complete signal set of each period,
a linear function is applied to the signal waveform of each cycle and normalized to eliminate the influence of zero point drift,
linear interpolation is carried out on the processed signals of each period to eliminate the influence of discrete signals and different period sizes of each period,
and acquiring a complete waveform, wherein the average variance of sample values of different sites of the waveform is calculated, linear weighted combination is carried out on signals with different shooting positions, namely different pulse information sources according to a minimum variance principle, and finally the complete waveform of a single pulse period is obtained through signal processing such as denoising.
The video captured by the fingertip of the index finger (fig. 2(b)) is processed, and the video captured by the radial artery of the wrist (fig. 2(a)) is processed in the same manner as the fingertip of the index finger.
The part of the video beginning and ending one second (30 frames before and after in this embodiment, which generally contains a lot of noise) is removed, and the rest of the video is decomposed by frames, assuming that the number of the obtained video frames is N and the ordinal number of the frame is used as the abscissa, the average value of the R channel in the RGB channels of each frame of the picture is calculated as the amplitude of the ordinate. The measured amplitude and the corresponding ordinal number are represented by the coordinate axis to obtain the waveform shown in fig. 4 (the same processing is performed on the video shot at the tip of the index finger to obtain the waveform shown in fig. 3).
Let the amplitude be XkWhere k represents the kth frame. Calculating the amplitude X of each frame except the first framekIs increased by
dk+1＝Xk+1-Xk(1)
Making statistics of the mean Ed and variance E [ (d-Ed)2]Selecting dkFrame number k of mean value greater than d plus twice variance, i.e.
{k|argkdk＞Ed+2E[(d-Ed)2]} (2)
And obtaining a time point when the amplitude is obviously and greatly increased, wherein the obtained k set corresponds to the process of increasing the blood pressure in each pulse period corresponding to the k-th frame of the shot video.
And recording the first element of the obtained set of the frame sequence number k into a new set for storing intermediate data, judging the growing node set from the second element to the last element one by one, and recording the element into the new set if the difference value between the element and the last element in the new set is more than 10 frames. The set of the starting point frame number serial numbers of each pulse cycle can be obtained through the operation. By this method, the time point of the start of each pulse wave period can be obtained, and a time-series signal (fig. 3 and 4) including a plurality of periods can be decomposed for each period to obtain a set of signals each having one element as each period.
The number of frames at the beginning and end of each period of the pulse (the end of the previous period is the beginning of the next period) is obtained according to the above method. The obtained pulse signals of each period are processed, and taking the signal of any period as an example, the number of frames corresponding to the signal in the period is set as { t }i，ti+1，...，ti+TiWhere T isiLet l ∈ {1, 2.,. T ] be the number of frames included in a periodic signaliTherefore, therefore
Indicating the signal strength corresponding to a certain frame in the period
From the above equations (1) and (2), the video collected from the radial artery was also analyzed (fig. 4), and the signals collected from the index finger tip (denoted as X) were obtained respectivelyfinger) And signals acquired at the radial artery (denoted as X)artery). It can be found that, compared with the commercially available ordinary pulse diagnosis instrument (the waveform measured by the commercially available pulse diagnosis instrument for the same sample is shown in fig. 6), the pulse diagnosis information acquisition method of the invention does not need special hardware facilities such as a pressurizing device, a pressure sensor and the like, and can also obtain a pulse signal basically consistent with the waveform characteristics of the professional pulse diagnosis instrument. The pulse signals acquired at different positions are all one observation of real human pulse signals, and the pulse signals contain mutually independent noise signals with zero mean value and unequal variance.
The signal strength obtained by shooting a video relative to the radial artery by the finger tip of the index finger is stronger, and the radial artery position follows the pulse taking mode of the traditional Chinese medical pulse diagnosis, so that the detailed characteristics of the signal at the radial artery position are kept more.
Meanwhile, for shooting at the radial artery by a mobile phone camera, the complete process equivalent to the floating, middle and deep three pulse taking methods in the traditional Chinese medicine can be realized according to the difference of force applied to the skin of the radial artery by a lens.
Since during the process of measuring the pulse signal, the measurement noise (such as irregular and minute hand shake) which affects the signal is inevitably generated. The waveform measured by the commercially available pulse condition apparatus of fig. 6 or the waveform measured by the mobile phone at the radial artery of fig. 4 can be seen to have such unavoidable noises (such as zero drift and irregular high-frequency jitter in the waveform). In order to obtain a smooth waveform, noise is removed, and pulse waveform characteristics are kept so as to facilitate subsequent waveform identification and analysis. We need to process the measured waveform.
For a complete signal in any pulse fluctuation period, setting the pulse fluctuation period as Ti。
Order to
Because the signals acquired by the mobile phone video are discrete values in time and the pulse fluctuation time of one period measured in each cycle is not strictly identical (because the pulse periods of human bodies are not strictly identical or the phenomena of irregular heart rate and the like exist), the amplitude between two adjacent signals is estimated by using a linear interpolation method.
Obtaining the processed signal values of the finger tip of the index finger and the radial artery according to the formulas (5) and (6):
for each period of the two signal waveforms obtained by the equations (5) and (6), p can be obtained from the equations (7) and (8) described aboveiWherein i is 1, 1/n, 2/n, …, 1. Let its corresponding random variables be Pfinger，iAnd P isfartery，i
Then:
and finally, obtaining the pulse signal of the collected person according to the mean value and the variance of the periodic signal measured at different positions for subsequent analysis and application. (equations 13,14, FIG. 5).
pi＝Partery，i+α(Pfinger，i-Partery，i) (14)。
Claims (8)
1. Pulse condition information acquisition system based on smart mobile phone includes:
the pulse condition acquisition part comprises:
the intelligent mobile phone camera is used for being attached to the body part of the pulse condition video to be shot, starting the flash lamp and the camera to shoot so as to obtain the video,
a video control part, when the mobile phone camera is tightly attached to the skin, the flash lamp is turned on and the brightness of the flash lamp is controlled, and simultaneously the video function is turned on,
a pulse information video analyzing and extracting part for processing the signals of the first video shot from the first part and the second video shot from the second part, filtering the noise, fusing the data and the like to obtain a single-cycle complete waveform,
the pulse condition information analyzing and extracting part comprises:
a signal extracting section for decomposing the first video and the second video into a series of frame pictures by frame, and analyzing R-channel information of each frame picture to obtain an unprocessed pulse signal containing a plurality of pulse periods and containing noise, an
A section for performing subsequent signal processing, comprising:
the pulse signals are screened according to the statistical information such as the amplitude intensity and the like of the pulse signals so as to be decomposed into a part of a complete signal set of each period,
a linear function is applied to the signal waveform of each cycle and normalized to eliminate a portion of the influence of zero point drift,
linear interpolation is performed on the processed signals of each period to eliminate discrete signals and portions of the signals of which the periods are not exactly the same in magnitude,
and the complete waveform acquisition part is used for calculating the mean variance of sample values at different sites of the waveform, performing linear weighted combination on signals with different shooting positions, namely different pulse information sources according to the minimum variance principle, and finally obtaining a complete waveform of a single pulse period which is subjected to signal processing such as denoising and the like.
2. The smartphone-based pulse condition information acquisition system according to claim 1, wherein:
the part for screening the pulse signals according to the statistical information such as the amplitude intensity and the like of the pulse signals so as to decompose the pulse signals into a complete signal set of each period comprises the following steps:
a portion that calculates an amplitude increment in accordance with:
dk+1＝Xk+1-Xk
calculating an amplitude increment, wherein XkIs the average value of R channels of the k frame picture of the video, which represents the pulse signal intensity,
selecting a portion of the frame ordinal according to:
{k|argkdk＞Ed+2E[(d-Ed)2]}
the number k of the frames is selected,
the section for applying a linear function to the signal waveform of each cycle to eliminate the influence of the zero point shift includes:
a linear function zero drift elimination section based on
Applying a linear function to eliminate null shift, where TiIs the number of frames included in a periodic signal, l ∈ {1, 2iTherefore, therefore
Indicating the corresponding signal strength of a certain frame within the period,
a normalization portion in accordance with
And carrying out normalization processing to ensure that the maximum amplitude values of all periods are consistent, thereby facilitating subsequent signal processing.
3. The smartphone-based pulse condition information acquisition system according to claim 1, wherein:
the part for performing linear interpolation on the processed signals of each period to eliminate the influence of discrete signals and the cycle size of each period not strictly identical comprises a part for performing linear interpolation according to the following relation:
Wherein:
fias a discrete function of the ith signal period,
fithe abscissa of (a) is time, the ordinate is the corresponding signal strength,
pi(x) For the purpose of the corresponding linear interpolation function,
x0，x1belong to fiGiven the set of abscissas of the scatter points,
{pfinger(x)}、{partery(x) The periodic signal amplitude set is obtained by interpolating the acquisition signals at different positions respectively through a linear interpolation function p,
the complete waveform acquisition section includes a section that determines parameters in accordance with the following relationship:
pi＝Partery，i+α(Pfinger，i-Partery，i)，
wherein:
DPfinger，ishowing the variance of the amplitude set of the periodic signal at the time i obtained by the above-mentioned processing and interpolation for the pulse signal collected at the first part,
DPartery，ishowing the variance of the amplitude set of the periodic signal at the time i obtained by the above-mentioned processing and interpolation for the pulse signal collected at the second part,
pithe magnitude of the resulting signal at time i for final processing.
4. The smartphone-based pulse condition information acquisition system according to claim 1 or 3, wherein:
the first part is the tip of the index finger,
the second site is the radial artery site.
5. Pulse condition information acquisition method based on smart phone, its characterized in that includes:
the pulse condition acquisition step comprises:
the camera of the intelligent mobile phone is used for clinging to the body part of the pulse condition video to be shot, the flash lamp and the camera are started for shooting, so as to obtain the video,
a video recording control step, when the mobile phone camera is tightly attached to the skin, the flash lamp is turned on, the brightness of the flash lamp is controlled, and the video recording function is turned on at the same time,
a pulse condition information video analysis and extraction step, signal processing is carried out on a first video shot from a first part and a second video shot from a second part, a single-cycle complete waveform is obtained through noise filtering, data fusion and the like,
the pulse condition information analyzing and extracting steps comprise:
performing signal extraction from the video, including decomposing the first video and the second video by frames to obtain a series of frame pictures, analyzing R channel information of each frame picture to obtain unprocessed pulse signal containing a plurality of pulse periods and containing noise, and
performing subsequent signal processing, including:
the pulse signals are screened according to the statistical information such as the amplitude intensity and the like of the pulse signals so as to be decomposed into a complete signal set of each period,
a linear function is applied to the signal waveform of each cycle and normalized to eliminate the influence of zero point drift,
a step of performing linear interpolation on the processed signals of each period to eliminate the discrete signals and the influence of different periods of different sizes,
and a complete waveform obtaining step, which comprises the steps of calculating the mean variance of sample values of different sites of the waveform, performing linear weighted combination on signals with different shooting positions, namely different pulse information sources according to the minimum variance principle, and finally obtaining a complete waveform of a single pulse period which is subjected to signal processing such as denoising and the like.
6. The pulse condition information acquisition method based on the smart phone as claimed in claim 5, wherein:
the processing of screening the pulse signals according to the statistical information such as the amplitude intensity and the like so as to decompose the pulse signals into a complete signal set of each period comprises the following steps:
calculating a magnitude delta, comprising according to:
dk+1＝Xk+1-Xk
calculating an amplitude increment, wherein XkIs the average value of R channels of the k frame picture of the video, which represents the pulse signal intensity,
selecting a frame ordinal number, including according to:
{k|argkdk＞Ed+2E[(d-Ed)2]} (2)
the number k of the frames is selected,
the process of applying a linear function to the signal waveform of each cycle to eliminate the influence of zero drift includes:
eliminating zero drift of a linear function, comprising according to:
applying a linear function to eliminate null shift, where TiIs the number of frames included in a periodic signal, l ∈ {1, 2iTherefore, therefore
Indicating the corresponding signal strength of a certain frame within the period,
normalization processing, including according to:
And carrying out normalization processing to ensure that the maximum amplitude values of all periods are consistent, thereby facilitating subsequent signal processing.
7. The pulse condition information acquisition method based on the smart phone as claimed in claim 5, wherein:
the processing of performing linear interpolation on each processed periodic signal to eliminate the influence of discrete signals and the non-strict difference in the period size of each period includes processing of performing linear interpolation according to the following relationship:
Wherein:
fias a discrete function of the ith signal period,
fithe abscissa of (a) is time, the ordinate is the corresponding signal strength,
pi(x) For the purpose of the corresponding linear interpolation function,
x0，x1belong to fiGiven the set of abscissas of the scatter points,
{pfinger(x)}、{partery(x) The periodic signal amplitude set is obtained by interpolating the acquisition signals at different positions respectively through a linear interpolation function p,
the full waveform acquisition step includes a process of determining parameters according to the following relationship:
pi＝Partery，i+α(Pfinger，i-Partery，i)，
wherein:
DPfinger，ishowing the variance of the amplitude set of the periodic signal at the time i obtained by the above-mentioned processing and interpolation for the pulse signal collected at the first part,
DPartery，ishowing the variance of the amplitude set of the periodic signal at the time i obtained by the above-mentioned processing and interpolation for the pulse signal collected at the second part,
pithe magnitude of the resulting signal at time i for final processing.
8. The pulse condition information acquisition method based on the smart phone according to claim 5 or 7, wherein:
the first part is the tip of the index finger,
the second site is the radial artery site.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811628312.1A CN111374646B (en) | 2018-12-28 | 2018-12-28 | Non-hardware-dependent pulse condition information acquisition system and method based on smart phone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811628312.1A CN111374646B (en) | 2018-12-28 | 2018-12-28 | Non-hardware-dependent pulse condition information acquisition system and method based on smart phone |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111374646A true CN111374646A (en) | 2020-07-07 |
| CN111374646B CN111374646B (en) | 2021-08-03 |
Family
ID=71212761
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811628312.1A Active CN111374646B (en) | 2018-12-28 | 2018-12-28 | Non-hardware-dependent pulse condition information acquisition system and method based on smart phone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111374646B (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105595971A (en) * | 2015-12-15 | 2016-05-25 | 清华大学 | Pulse condition information collecting system based on video and collecting method thereof |
| CN105769151A (en) * | 2016-02-25 | 2016-07-20 | 北京航空航天大学 | Multipoint pulse wave detection method and device |
| US20170238842A1 (en) * | 2016-02-19 | 2017-08-24 | Covidien Lp | Systems and methods for video-based monitoring of vital signs |
| CN107666853A (en) * | 2015-05-21 | 2018-02-06 | 皇家飞利浦有限公司 | Beat signals are determined according to video sequence |
| CN107920763A (en) * | 2015-07-16 | 2018-04-17 | 普瑞温迪克斯有限责任公司 | Handle biological data |
-
2018
- 2018-12-28 CN CN201811628312.1A patent/CN111374646B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107666853A (en) * | 2015-05-21 | 2018-02-06 | 皇家飞利浦有限公司 | Beat signals are determined according to video sequence |
| CN107920763A (en) * | 2015-07-16 | 2018-04-17 | 普瑞温迪克斯有限责任公司 | Handle biological data |
| CN105595971A (en) * | 2015-12-15 | 2016-05-25 | 清华大学 | Pulse condition information collecting system based on video and collecting method thereof |
| US20170238842A1 (en) * | 2016-02-19 | 2017-08-24 | Covidien Lp | Systems and methods for video-based monitoring of vital signs |
| CN105769151A (en) * | 2016-02-25 | 2016-07-20 | 北京航空航天大学 | Multipoint pulse wave detection method and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111374646B (en) | 2021-08-03 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110269600B (en) | Non-contact video heart rate detection method based on multivariate empirical mode decomposition and combined blind source separation | |
| 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 | |
| CN112949349B (en) | Method and system for displaying pulse condition waveform in real time based on face video | |
| CN109044322A (en) | A kind of contactless heart rate variability measurement method | |
| CN104545864B (en) | Psychological regulation method and apparatus | |
| US20160317041A1 (en) | System and methods for measuring physiological parameters | |
| CN112200162B (en) | Non-contact heart rate measuring method, system and device based on end-to-end network | |
| CN109793506B (en) | Non-contact radial artery waveform extraction method | |
| CN113397555A (en) | Arrhythmia classification algorithm of C-LSTM for physiological parameter monitoring | |
| CN111387959A (en) | Non-contact physiological parameter detection method based on IPPG | |
| CN112294282A (en) | Self-calibration method of emotion detection device based on RPPG | |
| Hernandez-Ortega et al. | A comparative evaluation of heart rate estimation methods using face videos | |
| CN110495888B (en) | Standard color card based on tongue and face images of traditional Chinese medicine and application thereof | |
| CN108937905A (en) | A kind of contactless heart rate detection method based on signal fitting | |
| CN113591769A (en) | Non-contact heart rate detection method based on photoplethysmography | |
| KR20170004804A (en) | A method for estimating respiratory and heart rate using dual cameras on a smart phone | |
| CN113378798A (en) | Pulse video pulse extraction method and system based on video micro-motion amplification | |
| US10779771B2 (en) | Signal processing method and apparatus | |
| CN111374646B (en) | Non-hardware-dependent pulse condition information acquisition system and method based on smart phone | |
| Zhang et al. | Using rear smartphone cameras as sensors for measuring heart rate variability | |
| Karmuse et al. | A robust rppg approach for continuous heart rate measurement based on face | |
| CN113255786B (en) | Video quality evaluation method based on electroencephalogram signals and target salient characteristics | |
| Pansare et al. | Heart Rate Measurement from Face and Wrist Video | |
| CN114601474A (en) | Source domain sample screening method for motor imagery transfer learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |