JPWO2014017566A1 - Heart rate measuring method and apparatus - Google Patents

Abstract

The heart rate measurement method is a heart rate measurement method for measuring a person's heart rate in a non-contact manner by acquiring a three-dimensional shape change of the chest. The heart rate measuring method includes a first step of acquiring a three-dimensional shape of a human chest abdominal wall surface as three-dimensional coordinate point group data, and a time change of the three-dimensional coordinates at each point constituting the three-dimensional coordinate point group data. The second step of calculating as time series data of 1 and the bandpass filter processing for the first time series data extract the frequency band components of the heartbeat, thereby the second time change of the three-dimensional coordinates due to the heartbeat. And a third step of calculating as the time series data.

Description

  The present invention relates to a heart rate measuring method and apparatus for measuring a heart rate in a non-contact manner.

  Conventionally, heart rate measurement, which is one of vital signs, is generally performed using an ECG (electrocardiograph). Recently, a method using a non-contact measuring means without using a contact measuring means such as ECG has been proposed. For example, Garbe et al. Have proposed a technique using a thermal image (Non-patent Document 1), and Nagae et al. Have proposed a technique using a microwave (Non-Patent Document 2). These non-contact measurement methods are considered to have the effect of reducing the burden and restraint of the subject.

  In addition, Poh et al. Have proposed a method for measuring a heart rate from a color face image using a Web imaging means, and realized heart rate measurement with an inexpensive device (Non-patent Document 3).

Garbey, M .; Nanfei S .; Merla, A .; Pavlidis, I .; “Contact-Free Measurement of Cardiac Pulse Based on the Analysis of Thermal Imagery,” IEEE Transactions on BME, Vol. 54, no. 8, pp. 1418-1426, 2007 Nagae, D .; Mase, A .; "Measurement of vital signal by microwave reflexometry and application to stress evaluation," APMC 2009. Asia Pacific, pp. 477-480, 2009 Poh, M .; McDuff, D .; J. et al. Picard, R .; W. "Advancements in Noncontact, Multiparameter Physiological Measurements Using a Webcam," IEEE Transactions on BME, Vol. 58, no. 1, pp. 7-11, 1011

  The methods of Non-Patent Documents 1 and 2 require expensive measuring equipment. Further, since the method of Non-Patent Document 3 uses inexpensive color imaging means, the robustness against illumination fluctuation is low, and it is difficult to realize quantitative heartbeat waveform measurement.

  The problem to be solved by the present invention is to easily realize non-contact measurement of a heartbeat waveform with an inexpensive apparatus configuration.

The heart rate measuring method according to the present invention includes:
A heartbeat measurement method for measuring a heartbeat in a non-contact manner by acquiring a three-dimensional shape change of a human chest,
A first step of acquiring a three-dimensional shape of the surface of the chest wall of the person as three-dimensional coordinate point group data;
A second step of calculating the time change of the three-dimensional coordinates at each point constituting the three-dimensional coordinate point group data acquired in the first step as first time-series data;
A third step of calculating a time change of three-dimensional coordinates due to a heartbeat as second time-series data by extracting a component of a heartbeat frequency band by band-pass filter processing on the first time-series data;
It is characterized by having.

  In the first step, it is desirable to acquire the three-dimensional coordinate point group data from an image obtained by imaging the surface of a person's thoracoabdominal wall on which pattern light is projected.

  In the second step, it is desirable to calculate a change in distance between the lens of the imaging means for imaging the chest and abdominal wall surface and the chest and abdominal wall as the first time-series data.

  Further, the second time series data is converted into color information according to the size thereof, and the color information is displayed as a video on the three-dimensional coordinates of each point, thereby visualizing the dynamics of the chest and abdominal wall surface due to the heartbeat. You may have the step to do.

  Calculating the second time-series data for each of a plurality of sections corresponding to the chest and abdomen, and calculating a phase difference between heart rate frequency components of other sections with respect to a heart rate frequency component of a predetermined reference section; You may have.

The heart rate measuring device according to the present invention is:
A heartbeat measuring device that measures a heartbeat in a non-contact manner by acquiring a three-dimensional shape change of a person's chest,
Point cloud data acquisition means for acquiring the three-dimensional shape of the human chest abdominal wall surface as three-dimensional coordinate point cloud data;
Point cloud coordinate time change calculation means for calculating the time change of the three-dimensional coordinates at each point constituting the three-dimensional coordinate point cloud data acquired by the point cloud data acquisition means as first time-series data;
By extracting the component of the heart rate frequency band by band-pass filter processing for the first time series data, the time change of the three-dimensional coordinates at each point caused by the heart rate is calculated as the second time series data. Heart rate origin point cloud coordinate time change calculating means,
It is characterized by providing.

Also, pattern light projection means for projecting pattern light onto the chest abdominal surface,
An imaging means for imaging the chest abdominal surface of the person on which the pattern light is projected,
It is desirable that the point group data acquisition unit acquires the three-dimensional coordinate point group data from an image captured by the imaging unit.

  In addition, it is preferable that the point group coordinate time change calculation unit calculates a change in distance between the lens of the imaging unit and the thoracoabdominal wall as the first time-series data.

  In addition, the second time series data is converted into color information according to the amount of time change, and the color information is displayed as an image on the three-dimensional coordinates of each point. You may have a dynamics visualization means to visualize dynamics.

  Further, the second time series data is calculated for each of the plurality of sections corresponding to the chest and abdomen, and the phase difference for calculating the phase difference of the heart rate frequency component of the other section with respect to the heart rate frequency component of the predetermined reference section You may have a calculation means.

  In the heartbeat measuring method and the heartbeat measuring device according to the present invention, non-contact measurement of a heartbeat waveform can be easily realized with an inexpensive device configuration.

FIG. 1 is a flowchart showing the heartbeat measuring method of the present invention. FIG. 2A is an explanatory diagram showing the three-dimensional acquisition of the thoracoabdominal wall of the present invention, and FIG. 2B is a configuration diagram of the arithmetic unit. Example 1 FIG. 3A is an explanatory diagram showing a projected static pattern, and FIG. 3B is an explanatory diagram showing dimensions of the static pattern. Example 1 FIG. 4 is a flowchart showing the three-dimensional shape restoration algorithm. Example 1 FIG. 5 is a drawing-substituting photograph showing a wavy pattern image projected on the chest and abdominal wall. Example 1 FIG. 6 is a drawing-substituting photograph showing a three-dimensional view of the thoracoabdominal wall restored to the three-dimensional shape. Example 1 FIG. 7A is an original waveform of the time change of three-dimensional coordinates at a point on the thoracoabdominal wall, and FIG. 7B is a graph showing a waveform of the original waveform after bandpass filter processing. (Example 2) FIG. 8 is a graph showing the result of simultaneous measurement with ECG. (Example 2) FIG. 9 is a graph showing the coincidence with ECG. (Example 2) FIG. 10 is a drawing-substituting photograph showing the temporal distribution of the three-dimensional shape of the thoracoabdominal wall. (Example 2) FIG. 11 is a drawing-substituting photograph explaining sample points set on the chest and abdominal wall. (Example 3) FIG. 12 is a graph showing the time change of the three-dimensional coordinates at the sample points D1 to D6. (Example 3) FIG. 13 is a drawing-substituting photograph showing the phase difference of the heart rate frequency component at the sample point set on the thoracoabdominal wall. (Example 3) FIG. 14 is a drawing-substituting photograph showing a wavy line pattern image projected onto the chest and abdominal wall. Example 4 FIG. 15 is a graph showing the results of simultaneous measurement with ECG. Example 4

  The heartbeat measuring method according to the present embodiment measures a person's heartbeat in a non-contact manner by acquiring a three-dimensional shape change of the chest and abdominal wall surface, and has three steps shown in FIG.

  First, in the first step, the active stereo method is used to acquire the three-dimensional shape of the human thoracoabdominal wall surface as three-dimensional coordinate point group data. As will be described later, the three-dimensional coordinate point group data is acquired from an image obtained by imaging the surface of the chest and abdominal wall of the person onto which the pattern light is projected.

  Next, in the second step, the time change is calculated for the three-dimensional coordinates at each point constituting the three-dimensional coordinate point group data acquired in the first step, and the first time-series data is acquired. As will be described later, the first time-series data is calculated as a change in the distance between the lens of the imaging means for imaging the chest and abdominal wall surface and the chest and abdominal wall.

  In the subsequent third step, band pass filter processing is performed on the first time-series data, and the components of the heart rate frequency band are extracted. Thereby, the time change of the three-dimensional coordinate by the heartbeat in each point is calculated as 2nd time series data.

  Furthermore, in the heartbeat measuring method according to the present embodiment, as a step following the third step, the time change of the three-dimensional coordinates due to the heartbeat at each point calculated in the third step is changed according to the magnitude. Converted to information. Then, the color information is displayed as an image on the three-dimensional coordinates of each point, thereby visualizing the dynamics of the chest and abdominal wall surface by heartbeat.

  Further, as will be described later, second time-series data is calculated for a plurality of sections corresponding to the chest and abdomen, and the phase difference between the heart rate frequency components of the other sections with respect to the heart rate frequency component of the predetermined reference section is calculated. The form provided with the step may be sufficient.

  The above heart rate measuring method can be realized using the following heart rate measuring device. The heart rate measuring device measures a heart rate in a non-contact manner by acquiring a three-dimensional shape change of the chest and abdominal wall surface, and includes a point cloud data acquisition unit, a point group coordinate time change calculation unit, and a heart rate described below. The origin point coordinate time change calculation means is configured.

  In the point cloud data acquisition means, the active stereo method is used to acquire the three-dimensional shape of the human chest abdominal wall surface as the three-dimensional coordinate point cloud data.

  In the point group coordinate time change calculation means, the time change is calculated for the three-dimensional coordinates at each point constituting the three-dimensional coordinate point group data acquired by the three-dimensional coordinate point group data, and the time series data (first time Acquire series data).

  In the heartbeat-derived point cloud coordinate time change calculation means, bandpass filter processing is performed on the first time-series data, and the components of the heartbeat frequency band are extracted. Thereby, signal components caused by body movements other than the heartbeat, that is, respiration and convulsive tremor are removed, and the time change of the three-dimensional coordinates due to the heartbeat at each point is calculated as the second time series data.

  Furthermore, in addition to the above configuration, the heartbeat measuring device converts the time change of the three-dimensional coordinates due to the heartbeat at each point calculated by the heartbeat-derived point cloud coordinate time change calculation means into color information according to the magnitude. In addition, dynamic visualizing means for displaying the color information as a video at the three-dimensional coordinates of each point may be provided. Thereby, the dynamics of the chest and abdominal wall surface by the heartbeat can be visualized.

  Further, the heartbeat measuring device calculates second time-series data for a plurality of sections corresponding to the chest and abdomen, and calculates a phase difference between the heartbeat frequency components of other sections with respect to the heartbeat frequency component of a predetermined reference section. A form provided with a phase difference calculation means may be sufficient.

  The point cloud data acquisition means, the point cloud coordinate time change calculation means, the heartbeat-derived point cloud coordinate time change calculation means, the dynamic visualization means, and the phase difference calculation means are respectively a point cloud data acquisition section and a point cloud coordinate time change calculation section. The heartbeat-derived point cloud coordinate time change calculation unit, the dynamics visualization unit, and the phase difference calculation unit may be configured to be incorporated into the arithmetic device and processed. In addition, the heartbeat measuring device includes an output unit that outputs data calculated by each of the above means to a monitor or the like.

  Hereinafter, based on an Example, the heart rate measuring method using the heart rate measuring device is explained in full detail. In the present embodiment, acquisition of the three-dimensional shape of a person's chest and abdominal wall is performed using the active stereo method. As shown in FIG. 2A, the heartbeat measuring device includes a pattern light projection unit 21, an imaging unit 22, and an arithmetic unit 25. As shown in FIG. 2B, the arithmetic device 25 includes a point cloud data acquisition unit, a point cloud coordinate time change calculation unit, and a heartbeat-derived point cloud coordinate time change calculation unit. Pattern light 23 is projected from the pattern light projection means 21 onto the chest and abdominal wall of the subject (person) 24 and photographed by the imaging means 22.

  The pattern light projection means 21 is preferably a liquid crystal projector, but a pattern projector using a combination of a transmissive diffraction grating and a laser light source may be used. Alternatively, the pattern light may be projected by transmitting halogen light, LED light, or laser light to a negative film on which a pattern of pattern light is drawn. From the light source of the pattern light projection means 21, light having a wavelength in the visible light to near infrared light band is appropriately selected and output. As the image pickup means 22, a CCD camera or a CMOS camera compatible with high-speed shooting is used.

  In this embodiment, the static pattern shown in FIG. 3 is projected. This pattern is a single color and includes sinusoidal vertical and horizontal curves arranged in a grid. Since the pattern is static, there is no need for synchronization, and no synchronization is required for shooting. Therefore, measurement at a very high frame rate is possible.

  A three-dimensional shape restoration algorithm in this embodiment is shown in FIG. This algorithm is based on inventions made by the inventors of the present invention (Japanese Unexamined Patent Application Publication No. 2009-300277, Japanese Patent Application No. 2011-157249 (Japanese Unexamined Patent Application Publication No. 2013-24608), Japanese Patent Application No. 2011-158175 (Japanese Unexamined Patent Application Publication No. 2013-24655). Issue gazette)).

  In the restoration algorithm, first, line detection is performed from a photographed image. By the optimization by Belief-Propagation (BP), a monochromatic grid line is stably separated and detected vertically and horizontally. The intersection is calculated from the detected vertical and horizontal lines, and a graph is created with the intersection as a node. Next, the position of the epipolar line corresponding to each node is calculated on the pattern light projection means 21 pattern, and if there is an intersection of grids on the line, this is determined as a corresponding candidate. At this time, since a plurality of correspondence candidates are usually found, the optimum combination of the correspondence candidates at each point is obtained using BP. Since the restoration result is sparse as it is, the depth at each pixel is finally obtained by using interpolation and matching of the pattern and the observed image in units of pixels to obtain a dense three-dimensional shape.

  The static pattern projected by the pattern light projection means 21 is not a pattern whose correspondence is uniquely determined by image processing, but uses a grid pattern composed of vertical and horizontal wavy lines in order to give information on the priority of correspondence. Since the wavy line pattern is a simple pattern, it can be easily detected as a curve in the image, and the position can be obtained with subpixel accuracy by calculating the peak of the luminance value.

  Although a periodic sine wave pattern is used as the wavy line, the wavy grid has information useful for detecting corresponding points. In the heartbeat measuring method of the present invention, the intersection of vertical and horizontal wavy lines is used as a feature point. The arrangement of the intersection points is determined by the interval between the wavy lines and the wavelength. Although wavy lines with a constant interval and wavelength are used, as described below, since the interval between the longitudinal wave lines is not an integer multiple of the wavelength of the transverse wave line, the pattern around the intersection has a locally unique pattern, and It can be used as a feature amount.

  The local pattern around the intersection is not unique among the entire projection pattern. In FIG. 3B, when Sx and Sy are the intervals between adjacent wavy lines, and Wx and Wy are wavelengths, Nx = 1 cm (Sx; Wx) = Sx; Ny = 1 cm (Sy; Wy) = Sy, The same pattern occurs for every Nx and Ny wavy lines along the horizontal and vertical axes, respectively. Here, 1 cm (a; b) is the least common multiple of a and b, and the subscripts x and y represent values along the horizontal and vertical axes, respectively. However, the local pattern is a pattern that can be identified in each cycle. FIG. 3B is an example of a pattern including Sx = 10; Sy = 11; Wx = Wy = 14; Ax = Ay = 1 (unit is pixel). In this example, one period is 7 vertical lines and 14 horizontal lines. Therefore, 98 (= 7 × 14) types of intersections exist in a rectangle formed by one cycle.

  In stereo matching, the corresponding point candidates are limited to points on the epipolar line. When the intersection of a certain pattern light projection means image and the epipolar line are located within an appropriate distance, the intersection of the pattern light projection means image is selected as one of the corresponding point candidates. The number of candidates depends on the intersection position of the imaging means image. Since the corresponding point candidates are sparsely distributed in the pattern light projection means image, the number of corresponding candidates can be significantly reduced compared to normal stereo vision in which candidate points are searched for in pixel units.

  By the active stereo method, a three-dimensional shape is restored from a wave line pattern image projected on the subject's chest wall, and a heartbeat waveform is calculated from a time change of the three-dimensional shape of the chest wall. Since it is intended for heart rate measurement, the imaging means 22 used for image acquisition is compatible with high-speed imaging. For the measurement of the heartbeat waveform, it is preferable that the imaging means 22 supports imaging at a frame rate of 60 FPS or higher. Here, it is assumed that the subject 24 is sitting on a chair with a backrest during image acquisition.

  In the restored point cloud data, the distribution density differs depending on the location, and it is impossible to strictly take correspondence between points between frames. Therefore, in the XYZ three-dimensional coordinate system in which the lens of the imaging means 22 is the origin and the depth direction is the Z axis, a two-dimensional lattice point is set on the XY plane, and the point group data is spatially resampled in a lattice shape to newly add a third order Find original coordinates. Then, correlation between frames is performed at the same lattice point, and a heart rate waveform is calculated by calculating a change in depth between consecutive frames. That is, the heartbeat waveform corresponds to a change (speed) in distance between the chest and abdominal wall and the imaging means lens. Note that any interpolation method may be used for spatial resampling of the three-dimensional point cloud data. For example, a linear interpolation triangulation method with low calculation cost is used.

  Since the calculated heartbeat waveform includes body movements of noise and respiratory components, the heartbeat waveform is calculated by extracting only the heartbeat component by performing bandpass filter processing with a transmission band of 0.4 to 5 Hz. Is called.

  In order to verify the validity of the heartbeat measuring method and heartbeat measuring apparatus of the present invention, actual measurement was performed using a prototype system. In the prototype system, SILICON VIDEO (R) monochrome 643M manufactured by EPIX was used as the high-speed imaging means 22. In 643M, a VGA image can be taken at a maximum frame rate of 211 FPS. Here, image acquisition at 100 FPS is performed. The focal length of the imaging means lens was 8 mm. Further, Epson liquid crystal pattern light projection means EB-1750 was used as the pattern light projection means 21 for projecting the wavy line pattern. The distance between the pattern light projection unit lens and the imaging unit lens was set to 600 mm.

  Here, the subject (person) 24 is a healthy male (age 41 years old, height 1.71 m, weight 62 kg). Prior to the measurement, a measurement agreement is exchanged with the subject. The subject's upper body is a white t-shirt. The measurement time was 30 seconds, and after the start of measurement, breathing was stopped for about 10 seconds, and normal breathing was performed for the remaining time.

  The measurement results are shown below. FIG. 5 is a wavy line pattern image of the chest wall acquired by the high speed imaging means. Three-dimensional point cloud data is restored from the wavy pattern image using the restoration algorithm shown in the first embodiment, and the three-dimensional point cloud data is resampled in a grid pattern on the XY plane. FIG. 6 is a three-dimensional view of the thoracoabdominal wall that is three-dimensionally restored from the resample point cloud data.

  FIG. 7A and FIG. 7B show the original waveform and the waveform of the original waveform after the band-pass filter processing with respect to the time change of the three-dimensional coordinates of the points indicated by x in FIG. In FIG. 7A, the original waveform contains large high-frequency noise. On the other hand, as a result of extracting only the frequency component of 0.4 to 5 Hz including the main signal component of the heartbeat by the bandpass filter processing using the fast Fourier transform, a periodic waveform is obtained as shown in FIG. It was.

  Furthermore, the validity of the calculated heartbeat waveform was examined by simultaneous measurement with ECG. The ECG used is a compact wireless ECG logger manufactured by LOGICAL PRODUCT. An ECG electrode was placed under the left chest of the subject. FIG. 8 shows the measurement result of the heartbeat waveform using ECG. From FIG. 8, it was considered that the coordinate change of the point cloud appearing on the surface of the abdominal wall was sub-millimeter order in each frame from the amplitude of the waveform after filtering. Further, FIG. 8 shows that the appearance interval of the peak of the heartbeat waveform by the heartbeat measuring method of the present invention and the appearance interval of the peak (R wave) of the ECG waveform are substantially the same.

  FIG. 9 is a Bland-Altman diagram summarizing the appearance intervals of peaks in each method. FIG. 9 reveals that there is sufficient coincidence between the peak appearance intervals of both methods, and it was proved that the heart rate can be calculated by the heart rate measuring method of the present invention.

  FIG. 10 shows the spatial distribution of the temporal change (change amount between frames) of the three-dimensional shape of the thoracoabdominal wall at the grid points of spatial resampling changed to color information (here, grayscale), and the restored chest wall portion It is expressed overlaid on a three-dimensional shape. FIG. 10 shows that the heartbeat appears on the left side of the chest wall where the heart exists. By visualizing a slight change in the body surface due to the heartbeat in this way, it is possible to visually recognize the heartbeat dynamics appearing on the surface of the abdominal chest wall.

  In the three-dimensional surface of the thoracoabdominal region restored in Example 2, lattice-like sample points A1, A2, A3, A4, A5, A6, B1, B2, B3, B4, B5, as shown in FIG. B6, C1, C2, C3, C4, C5, C6, D1, D2, D3, D4, D5, D6, E1, E2, E3, E4, E5, E6, F1, F2, F3, F4, F5, F6 The phase of the heart rate frequency component was set and examined.

  FIG. 12 is a time change of the three-dimensional coordinates at the sample points D1, D2, D3, D4, D5, and D6. From FIG. 12, it can be seen that there is a phase delay in the upper sample points such as D3, D4, D5, and D6 compared to the lower sample points such as D1 and D2.

  FIG. 13 is a phase map shown as a contour diagram in which the phase difference at each sample point is obtained so that the phase at the sample point A1 becomes a reference (zero). It is clear that there is a phase lag from below to above, indicating a change in the chest wall that accompanies the mechanical function of the heart (constriction and expansion mode). It is known that the heart contraction and dilation mode shows various dynamics depending on the heart disease. Therefore, by using the heart rate measurement method of the present invention to present a phase map of chest wall fluctuation, It can be used for understanding.

  In the above embodiment, the three-dimensional shape reconstruction of the chest and abdomen obtained by projecting the sine wave pattern is performed. However, in the present invention, the projected pattern is not limited to the sine wave pattern. For example, it may be a dot matrix pattern in which point groups are periodically arranged. FIG. 14 is an image of the chest and abdomen projected with the dot matrix pattern.

  FIG. 15 shows temporal changes in three-dimensional coordinates calculated from the three-dimensional shape of the chest and abdomen restored by dot matrix pattern projection. In this embodiment, simultaneous measurement with an electrocardiogram is performed. From FIG. 15, it can be seen that the time change of the three-dimensional coordinates obtained by the dot matrix pattern projection changes in the same period as the electrocardiogram waveform, similarly to the wave line pattern projection of the second embodiment.

  It should be noted that the present invention can be variously modified and modified without departing from the scope of the present invention. Further, the above-described embodiment is for explaining the present invention, and does not limit the scope of the present invention.

  This application is based on Japanese Patent Application No. 2012-163796 filed on July 24, 2012. In this specification, the specification of Japanese Patent Application No. 2012-163796 and the entire claims are incorporated by reference.

  With the heartbeat measuring method and heartbeat measuring device of the present invention, non-contact measurement of a heartbeat waveform can be easily realized with an inexpensive device configuration. For example, a heartbeat interval can be easily measured. Application as a device can be expected. In addition, since heart rate dynamics can be visualized by the heart rate measurement method and heart rate measurement device, screening for heart disease that has been performed based on information such as heart rate pattern and dynamic area by palpation and auscultation until now, May be implemented using vision. Thus, the present invention can be practically used as a new biological measurement technique in the medical field and can be expected to contribute to the development of medical technology.

21 Pattern light projection means 22 Imaging means 23 Pattern light 24 Subject 25 Computing device

[0005]
FIG. 9 is a graph showing coincidence with ECG. (Example 2)
[FIG. 10] FIG. 10 is a drawing-substituting photograph showing the temporal distribution of the three-dimensional shape of the thoracoabdominal wall. (Example 2)
FIG. 11 is a drawing-substituting photograph explaining sample points set on the chest and abdominal wall. Example 3
[FIG. 12] FIG. 12 is a graph showing the time change of three-dimensional coordinates at sample points D1 to D6. Example 3
FIG. 13 is a drawing-substituting photograph showing the phase difference of the heart rate frequency component at the sample point set on the thoracoabdominal wall. Example 3
MODE FOR CARRYING OUT THE INVENTION [0019]
The heartbeat measuring method according to the present embodiment measures a person's heartbeat in a non-contact manner by acquiring a three-dimensional shape change of the chest and abdominal wall surface, and has three steps shown in FIG.
[0020]
First, in the first step, the active stereo method is used to acquire the three-dimensional shape of the human thoracoabdominal wall surface as three-dimensional coordinate point group data. As will be described later, the three-dimensional coordinate point group data is acquired from an image obtained by imaging the surface of the chest and abdominal wall of the person onto which the pattern light is projected.
[0021]
Next, in the second step, the time change is calculated for the three-dimensional coordinates at each point constituting the three-dimensional coordinate point group data acquired in the first step, and the first time-series data is acquired. As will be described later, the first time-series data is calculated as a change in the distance between the lens of the imaging means for imaging the chest and abdominal wall surface and the chest and abdominal wall.
[0022]
In the subsequent third step, band pass filter processing is performed on the first time-series data, and the components of the heart rate frequency band are extracted. This

[0012]
The heartbeat dynamics appearing on the surface of the abdominal chest wall can be visually confirmed.
Example 3
[0051]
In the three-dimensional surface of the thoracoabdominal region restored in Example 2, lattice-like sample points A1, A2, A3, A4, A5, A6, B1, B2, B3, B4, B5, as shown in FIG. B6, C1, C2, C3, C4, C5, C6, D1, D2, D3, D4, D5, D6, E1, E2, E3, E4, E5, E6, F1, F2, F3, F4, F5, F6 The phase of the heart rate frequency component was set and examined.
[0052]
FIG. 12 is a time change of the three-dimensional coordinates at the sample points D1, D2, D3, D4, D5, and D6. From FIG. 12, it can be seen that there is a phase delay in the upper sample points such as D3, D4, D5, and D6 compared to the lower sample points such as D1 and D2.
[0053]
FIG. 13 is a phase map shown as a contour diagram in which the phase difference at each sample point is obtained so that the phase at the sample point A1 becomes a reference (zero). It is clear that there is a phase lag from below to above, indicating a change in the chest wall that accompanies the mechanical function of the heart (constriction and expansion mode). It is known that the heart contraction and dilation mode shows various dynamics depending on the heart disease. Therefore, by using the heart rate measurement method of the present invention to present a phase map of chest wall fluctuation, It can be used for understanding.
[0054]
[0055]

[0013]
[0056]
It should be noted that the present invention can be variously modified and modified without departing from the scope of the present invention. Further, the above-described embodiment is for explaining the present invention, and does not limit the scope of the present invention.
[0057]
This application is based on Japanese Patent Application No. 2012-163796 filed on July 24, 2012. In this specification, the specification of Japanese Patent Application No. 2012-163796 and the entire claims are incorporated by reference.
Industrial applicability [0058]
With the heartbeat measuring method and heartbeat measuring device of the present invention, non-contact measurement of a heartbeat waveform can be easily realized with an inexpensive device configuration. For example, a heartbeat interval can be easily measured. Application as a device can be expected. In addition, since heart rate dynamics can be visualized by the heart rate measurement method and heart rate measurement device, screening for heart disease that has been performed based on information such as heart rate pattern and dynamic area by palpation and auscultation until now, May be implemented using vision. Thus, the present invention can be practically used as a new biological measurement technique in the medical field and can be expected to contribute to the development of medical technology.
Explanation of symbols [0059]
21 Pattern light projection means 22 Imaging means 23 Pattern light 24 Subject 25 Computing device

Claims (10)

A heartbeat measurement method for measuring a heartbeat in a non-contact manner by acquiring a three-dimensional shape change of a human chest,
A first step of acquiring a three-dimensional shape of the surface of the chest wall of the person as three-dimensional coordinate point group data;
A second step of calculating the time change of the three-dimensional coordinates at each point constituting the three-dimensional coordinate point group data acquired in the first step as first time-series data;
A third step of calculating a time change of three-dimensional coordinates due to a heartbeat as second time-series data by extracting a component of a heartbeat frequency band by band-pass filter processing on the first time-series data;
A heart rate measurement method characterized by comprising:   The heartbeat measuring method according to claim 1, wherein in the first step, the three-dimensional coordinate point group data is acquired from an image obtained by imaging a thoracoabdominal wall surface on which pattern light is projected.   The heart rate measuring method according to claim 2, wherein, in the second step, a change in distance between a lens of an imaging unit that images the chest and abdominal wall surface and the chest and abdominal wall is calculated as the first time-series data.   The second time-series data is converted into color information according to the magnitude of the time change amount, and the color information is displayed as an image on the three-dimensional coordinates of each point. The heart rate measuring method according to any one of claims 1 to 3, further comprising a step of visualizing.   Calculating the second time-series data for each of a plurality of sections corresponding to the chest and abdomen, and calculating a phase difference between heart rate frequency components of other sections with respect to a heart rate frequency component of a predetermined reference section; The heartbeat measuring method according to any one of claims 1 to 3, wherein the heartbeat measuring method is provided. A heartbeat measuring device that measures a heartbeat in a non-contact manner by acquiring a three-dimensional shape change of a person's chest,
Point cloud data acquisition means for acquiring the three-dimensional shape of the human chest abdominal wall surface as three-dimensional coordinate point cloud data;
Point cloud coordinate time change calculation means for calculating the time change of the three-dimensional coordinates at each point constituting the three-dimensional coordinate point cloud data acquired by the point cloud data acquisition means as first time-series data;
By extracting the components of the frequency band of the heartbeat by band-pass filter processing with respect to the first time series data, the heartbeat for calculating the time change of the three-dimensional coordinates at each point caused by the heartbeat as the second time series data Origin point cloud coordinate time change calculating means,
A heartbeat measuring device comprising: Pattern light projection means for projecting pattern light onto the chest and abdomen surface;
An imaging means for imaging the chest abdominal surface of the person on which the pattern light is projected,
The heartbeat measuring apparatus according to claim 6, wherein the point group data acquisition unit acquires the three-dimensional coordinate point group data from an image captured by the imaging unit.   The heartbeat measuring apparatus according to claim 7, wherein the point group coordinate time change calculating means calculates a change in distance between the lens of the imaging means and the chest and abdominal wall as the first time-series data.   The second time-series data is converted into color information according to the magnitude of the time change amount, and the color information is displayed as an image on the three-dimensional coordinates of each point. The heartbeat measuring device according to claim 6, further comprising dynamic visualization means for visualizing the heartbeat.   Phase difference calculation means for calculating the second time-series data for each of a plurality of sections corresponding to the chest and abdomen, and calculating a phase difference between heart rate frequency components of other sections with respect to a heart rate frequency component of a predetermined reference section The heartbeat measuring device according to any one of claims 6 to 8, wherein the heartbeat measuring device is provided.