JP2014188237A - Pulse wave detection device, pulse wave detection method, and pulse wave detection program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve detection accuracy of a pulse wave propagation velocity.SOLUTION: A pulse wave detection device 10 acquires an image of the biological body of a subject which is captured by an imaging device, divides the image into a plurality of regions and performs predetermined statistical processing to pixel values possessed by pixels included in each of the regions for each of wavelength components so as to calculate a representative value for each of the regions, measures a timing at which the representative value in each of the regions where synthesis among the respective wavelength components is performed varies between the frames of the images, and calculates amounts of delays of pulse waves, from a temporal difference of the timing at which the representative value varies in each of the regions.

Description

  The present invention relates to a pulse wave detection device, a pulse wave detection method, and a pulse wave detection program.

  Changes in the volume of blood delivered from the heart, so-called pulse waves, are detected. Not only is the pulse wave itself an important indicator of vital signs, but useful biological information can also be obtained from the pulse wave velocity.

  The pulse wave propagation speed can be obtained from pulse wave delay times between parts and distances between parts by measuring pulse waves at a plurality of parts of the living body. For example, the pulse wave propagation velocity is an index useful for diagnosing the progress of arteriosclerosis, and for example, blood vessel age can be measured from the propagation velocity. Further, the pulse wave propagation speed is also related to the systolic blood pressure, and the pulse wave propagation speed can be used for blood pressure measurement. Furthermore, fluctuations in the pulse cycle are also related to the autonomic nerves, and the function of the sympathetic and parasympathetic nerves is diagnosed from the low-frequency and high-frequency components of fluctuations, and the ratio of low-frequency and high-frequency components. Also useful.

  In general, pulse waves measure changes in skin pressure due to blood vessel expansion while the pressure sensor is in contact with a living body, or irradiate blood vessels with visible or infrared light, and the amount of transmitted or reflected light. It is collected by detecting changes. However, when using a pressure-sensitive sensor, there are disadvantages such as the trouble of contacting a measuring instrument with a living body, and when detecting a change in the amount of light, additional hardware such as a light source is provided.

  Therefore, in order to detect a pulse wave without contacting a measuring instrument to the living body under ambient light such as sunlight or room light, a change in luminance of an image of a part of the subject's living body, for example, a face photographed. Techniques have been proposed for detecting a pulse wave using the. As an example of such a technique, there is a method of improving a signal-to-noise ratio by applying independent component analysis (ICA) to a signal component of an image obtained by photographing a subject's face by a camera.

JP 2008-301915 A JP-A-10-283482

  However, in the above technique, since the time resolution depends on the frame frequency of the imaging apparatus, the detection accuracy of the pulse wave velocity is limited.

  For example, if it is between the neck and the face, the pulse wave is propagated in about 100 ms. On the other hand, in general, an image pickup apparatus with a frame frequency of about 30 Hz is often employed. Thus, even if it is attempted to calculate a propagation time of about 100 ms with an imaging apparatus having only a time resolution of about 33 ms, the accuracy is naturally limited. Further, since the fluctuation of the pulse cycle is about 10 ms shorter than the propagation speed, it is difficult to accurately obtain the fluctuation of the pulse period more than the pulse wave propagation speed. Even if it is so, it is not easy to adopt a high-speed imaging device with a short frame frequency because of the high hurdles in terms of cost.

  An object of one aspect is to provide a pulse wave detection device, a pulse wave detection method, and a pulse wave detection program that can improve the detection accuracy of a pulse wave propagation velocity.

  The pulse wave detection device according to one aspect includes an acquisition unit that acquires an image in which a living body of a subject is captured by an imaging device. Further, the pulse wave detection device includes a dividing unit that divides the image into a plurality of regions. Further, the pulse wave detection device includes a statistical processing unit that calculates a representative value of each region for each wavelength component by executing predetermined statistical processing for each wavelength component on the pixel values of the pixels included in each region. Further, the pulse wave detection device includes a measurement unit that measures a timing at which a representative value for each region synthesized between the wavelength components changes between frames of the image. Further, the pulse wave detection device includes a calculation unit that calculates a delay amount of the pulse wave from a time difference in timing at which the representative value changes in each region.

  According to one embodiment, the detection accuracy of the pulse wave velocity can be improved.

FIG. 1 is a block diagram illustrating a functional configuration of the pulse wave detection device according to the first embodiment. FIG. 2 is a diagram illustrating an example of a face image. FIG. 3 is a diagram illustrating an example of a face image. FIG. 4 is a diagram illustrating an example of the relationship between the luminance change and the sampling time. FIG. 5 is a diagram illustrating an example of the relationship between the luminance change and the sampling time. FIG. 6 is a diagram illustrating an example of the relationship between the luminance change and the sampling time. FIG. 7 is a diagram illustrating an example of the relationship between the luminance change and the sampling time. FIG. 8 is a diagram illustrating an example of a time difference in luminance change timing illustrated in FIG. FIG. 9 is a diagram illustrating an example of the relationship between the luminance change and the sampling time. FIG. 10 is a diagram illustrating an example of a time difference in luminance change timing illustrated in FIG. 9. FIG. 11 is a flowchart illustrating a procedure of delay amount calculation processing according to the first embodiment. FIG. 12 is a diagram illustrating an example of an image showing a subject. FIG. 13 is a flowchart illustrating a procedure of delay amount calculation processing according to the application example 2. FIG. 14 is a schematic diagram illustrating an example of a computer that executes a pulse wave detection program according to the first and second embodiments.

  Embodiments of a pulse wave detection device, a pulse wave detection method, and a pulse wave detection program disclosed in the present application will be described below with reference to the drawings. Note that this embodiment does not limit the disclosed technology. Each embodiment can be appropriately combined within a range in which processing contents are not contradictory.

[Configuration of pulse wave detector]
First, the functional configuration of the pulse wave detection device according to the present embodiment will be described. FIG. 1 is a block diagram illustrating a functional configuration of the pulse wave detection device according to the first embodiment. The pulse wave detection device 10 shown in FIG. 1 uses a subject's pulse wave, that is, a subject's pulse wave, i.e., an image obtained by photographing the subject without bringing the measuring instrument into contact with the living body under ordinary ambient light such as sunlight or room light. A pulse wave detection process for detecting a change in the volume of blood accompanying the pulsation of the heart is executed.

  As one aspect, the pulse wave detection device 10 can be implemented by installing a pulse wave detection program provided as package software or online software on a desired computer. For example, the pulse wave detection program is installed not only on communication terminals such as smartphones, mobile phones, and PHS (Personal Handyphone System), but also on mobile terminal devices including tablet terminals and slate terminals that do not have communication functions. Accordingly, the mobile terminal device can function as the pulse wave detection device 10. Here, the portable terminal device is illustrated as an implementation example of the pulse wave detection device 10, but the pulse wave detection program may be installed in a fixed terminal such as a personal computer.

  As shown in FIG. 1, the pulse wave detection device 10 includes a camera 11, an image memory 11a, an acquisition unit 12, a first dividing unit 13g, a second dividing unit 13r, and statistical processing units 14g1 to 14gn. And statistical processing units 14r1 to 14rn. Furthermore, the pulse wave detection device 10 includes BPFs (Band-Pass Filters) 15g1 to 15gn and BPFs 15r1 to 15rn, difference calculation units 16-1 to 16-n, binarization units 17-1 to 17-n, A measurement unit 18 and a calculation unit 19 are included.

  Such a pulse wave detection device 10 may include various functional units included in known mobile terminal devices in addition to the functional units illustrated in FIG. 1. For example, when the pulse wave detection device 10 is implemented as a tablet terminal or a slate terminal, the pulse wave detection apparatus 10 may further include an input device, a display device, or an inputable and displayable touch panel. In addition, when the pulse wave detection device 10 is mounted as a mobile terminal, it further includes functional units such as an antenna, a carrier communication unit that performs communication via a carrier network, and a GPS (Global Positioning System) receiver. It doesn't matter.

  Among the functional units shown in FIG. 1, the camera 11 is an imaging device on which an imaging element such as a charge coupled device (CCD) or a complementary metal oxide semiconductor (CMOS) is mounted. For example, the camera 11 can be equipped with three or more light receiving elements such as R (red), G (green), and B (blue). As an example of mounting the camera 11, a digital camera or a web camera may be connected via an external terminal, or when the camera is mounted from the time of shipment, the camera can be used. In addition, although the case where the pulse wave detection device 10 includes the camera 11 is illustrated here, the pulse wave detection device 10 does not necessarily include the camera 11 when an image can be acquired via a network or a storage device. Also good.

  Here, the pulse wave detection device 10 is configured so that when the application program for executing the above-described pulse wave detection process is preinstalled or installed, the camera 11 captures an image of a subject whose pulse wave is easily detected. An image shooting operation can be guided. In the following description, the application program may be referred to as a “pulse wave detection application”.

  When the pulse wave detection application is activated via an input device (not shown), the camera 11 is activated. In response to this, the camera 11 starts photographing the subject accommodated in the photographing range of the camera 11. At this time, when shooting an image showing the face of the subject, the pulse wave detection app displays the image taken by the camera 11 on a display device (not shown) and displays the target position where the nose of the subject is projected as an aim. It can also be made. As a result, among the facial parts such as the subject's eyes, ears, nose and mouth, an image in which the subject's nose is within the center of the imaging range can be taken. Then, the pulse wave detection application stores an image in which the face of the subject is captured by the camera 11 in the image memory 11a. In the following, an image showing a face may be referred to as a “face image”.

  The image memory 11a is a storage device that stores images. As one aspect, each time a subject's face is photographed by the camera 11, the subject's face image is stored in the image memory 11a. At this time, the image memory 11a may store a moving image encoded by a predetermined compression encoding method, or may store a set of still images showing the face of the subject. Hereinafter, assuming that a moving image in which the face of the subject is captured by the camera 11 is acquired, an image of each frame forming the moving image may be described as a “frame image”. Here, the case where the face image taken by the camera 11 is stored is illustrated, but the face image received via the network may be stored.

  The acquisition unit 12 is a processing unit that acquires an image. As one aspect, the acquisition unit 12 acquires the face image of the subject stored in the image memory 11a. As another aspect, the acquisition unit 12 acquires an image from an auxiliary storage device such as a hard disk or an optical disk or a removable medium such as a memory card or a USB (Universal Serial Bus) memory that stores an image of the face of the subject. You can also As a further aspect, the acquisition unit 12 can also acquire a face image received from an external device via a network. In addition, although the acquisition part 12 illustrated the case where a process is performed using image data, such as two-dimensional bitmap data and vector data obtained from the output by image pick-up elements, such as CCD and CMOS, it is output from one detector. It is also possible to acquire the processed signal as it is and execute the subsequent processing.

  Here, the pulse wave detection device 10 according to the present embodiment uses three wavelength components included in the image, that is, signals of two wavelength components of the R component and the G component among the R component, the G component, and the B component. The case where a pulse wave is detected is illustrated. For example, each time the face image stored in the image memory 11a is read, the acquisition unit 12 outputs a face image G having a G component luminance value for each pixel to the first division unit 13g and R for each pixel. The face image R having the luminance value of the component is output to the second dividing unit 13r.

  The first dividing unit 13g and the second dividing unit 13r are processing units that divide a face image. That is, the first dividing unit 13g and the second dividing unit 13r except that the first dividing unit 13g is in charge of dividing the face image G and the second dividing unit 13r is in charge of dividing the face image R. Similar processing is executed. Here, processing executed by the first dividing unit 13g will be described on behalf of the first dividing unit 13g and the second dividing unit 13r.

For example, each time the face image G is input from the acquisition unit 12, the first dividing unit 13 g executes a process described below. 2 and 3 are diagrams illustrating examples of face images. As shown in FIG. 2, the first dividing unit 13g performs image processing such as template matching on the face image 200, thereby performing facial parts such as the left eye 210a of the subject from the face image 200 having a G component luminance value. The right eye 210b, chin 211, and the like are extracted. Then, the first dividing unit 13g divides the partial image in which the face image is divided by the two types of face parts into N regions along the direction in which the pulse wave propagates. For example, as shown in FIG. 3, the first dividing unit 13g extends in the horizontal direction from the position where the straight eye Le extends horizontally from the position where the left eye 210a or the right eye 210b exists and the position where the jaw 211 exists. A partial image sandwiched between the straight lines Lj is extracted. On top of that, the first division portion 13g is divided partial image flanked by straight lines Le and linear Lj propagation direction of the pulse wave, i.e., the N regions of P ₁ to P _n along the vertical direction of the image To do.

Thus, the reason why the face image is divided along the propagation direction of the pulse wave is to make the timing at which the pulse wave arrives in each of the regions P _{1 to} P _n substantially the same. That is, since blood is sent out from the heart along the artery, the pulse wave reaches in the region where the longitudinal direction extends in the direction perpendicular to the propagation direction of the pulse wave, that is, in the horizontal direction of the image. The pulse wave detection device 10 uses the point that the timings to be considered are substantially the same.

Although the case where the first dividing unit 13g divides the face image is illustrated here, the same processing is executed in the second dividing unit 13r. The first division section 13g and the second divided portion 13r is divided manner by one of the processing unit, for example, when the position or size of the N regions of P ₁ to P _n is determined, the other The processing unit can be notified of the positions and sizes of the N areas P _{1 to} P _n . Thus, the face image may be divided without causing the other processing unit to perform image processing such as template matching. Although the case where the division method is determined using the face image G is illustrated here, the division method may be determined using the original image, that is, the original face image.

In the processing unit subsequent to the first dividing unit 13g and the second dividing unit 13r, processing is performed for each of the two wavelength components of the G component and the R component, and for each of the divided regions P _{1 to} P _n. Is executed. Therefore, the first division unit 13g outputs the G signal of the region _{P 1} divided from the face image G to statistical processing section 14g1, and outputs a G signal of the region _{P 2} to the statistical processing section 14G2, · · ·, and it outputs a G signal of the region _{P n} to the statistic processing unit 14Gn. The second division part 13r outputs the R signal of a region _{P 1} divided from the face image R to the statistical processing section 14r1, and outputs the R signal of the region _{P 2} to the statistical processing section 14r2, · · · and it outputs the R signal of the region _{P n} to the statistic processing unit 14Rn.

  The statistical processing units 14g1 to 14gn and 14r1 to 14rn are processing units that perform predetermined statistical processing on the pixel values included in the region for each wavelength component or for each region. The statistical processing units 14g1 to 14gn and 14r1 to 14rn have statistical processing units 14g1 to 14gn in charge of G signal statistical processing and statistical processing units 14r1 to 14rn in charge of R signal statistical processing. The contents of the process are the same. Therefore, the statistical processing executed by the statistical processing units 14g1 to 14gn on behalf of the statistical processing units 14g1 to 14gn and 14r1 to 14rn will be described.

As an embodiment, the statistical processing unit 14g1 averages the luminance value of the G component having the pixels included in the region P _1. Here, the case of calculating the average value is exemplified, but the median value and the mode value may be calculated. In addition to the weighted average, an arbitrary average process such as a weighted average or a moving average is executed. You can also Thus, the average value of the luminance value of the G component having the pixels included in the region P ₁ is calculated as a representative value representing the G component of the region P _1. Further, the statistical processing unit 14g2 calculates a representative value of the G component in the region _{P 2,} the statistical processing unit 14gn calculates a representative value of the G component in the region _{P n.} Also in the statistical processing unit 14R1～14rn, the representative value of the R component of the region _{P 1,} a representative value of the R component of the region _{P 2,} · · ·, a representative value of the R component of the region _{P n} is calculated.

  Each of the BPFs 15g1 to 15gn and 15r1 to 15rn is a bandpass filter that passes only signal components in a predetermined frequency band and removes signal components in other frequency bands. These BPFs 15g1 to 15gn and 15r1 to 15rn may be implemented by hardware or may be implemented by software.

  These BPFs 15g1 to 15gn and 15r1 to 15rn pass signal components in a pulse wave frequency band of 0.7 Hz or more and less than 4 Hz that can be taken by a pulse wave and in a frequency band of 42 bpm or more and less than 240 bpm when converted per minute.

Thus, BPF15g1 G signal of the pulse wave frequency band region _{P 1} is extracted by, G signal of the pulse wave frequency band region _{P 2} is extracted by BPF15g2, further pulse wave frequency band region _{P n} by BPF15gn G signals are extracted. On the other hand, an R signal in the pulse wave frequency band of the region P ₁ is extracted by the BPF 15r1, an R signal in the pulse wave frequency band of the region P ₂ is extracted by the BPF 15r2, and further, an R signal in the pulse wave frequency band of the region P _n is extracted by the BPF 15rn. A signal is extracted.

The difference calculation unit 16-1 to 16-n is for each region, and the G signal of the pulse wave frequency band of the output regions _P 1 to P _n by BPF15g1～15gn, output by BPF15r1~15rn regions _P 1 to P It is a processing part which calculates the difference with R signal of _n pulse wave frequency band.

One The embodiment, the difference calculation unit 16-1 subtracts the R signal of the pulse wave frequency band regions _{P 1} output by BPF15r1 from G signals outputted pulse wave frequency band regions _{P 1} by BPF15g1. Similarly, the difference calculation unit 16-2 subtracts the R signal of the pulse wave frequency band of the output region _{P 2} by BPF15r2 from G signal of the pulse wave frequency band of the output region _{P 2} by BPF15g2. Similarly, the difference calculation unit 16-n, for the G signal of the pulse wave frequency band of the output region _{P n} by BPF15gn, subtracts the R signal of the pulse wave frequency band of the output region _{P n} by BPF15rn. Thus, the result of region P ₁ to P _n noise component contained in the G signal of the pulse wave frequency band is canceled by the R signal of the pulse wave frequency band regions P ₁ to P _n, region P ₁ to P _A luminance signal with a high S / N ratio can be obtained for each _n . Thus, the time-series data of the luminance signal obtained for each of the regions P _{1 to} P _n can be used as a pulse waveform.

Here, the case where the R signal in the pulse wave frequency band is subtracted from the G signal in the pulse wave frequency band in each of the regions P _{1 to} P _n has been described. From the viewpoint of improving the SN ratio of the luminance signal, Such a method can also be adopted. For example, the pulse wave detection device 10 has a frequency band corresponding to noise between the G signal and the R signal, that is, a frequency band of 3 bpm or more and less than 20 bpm in which noise such as blinking of the body light or flickering of environmental light is likely to appear. The ratio of absolute intensity values at is calculated. After that, the pulse wave detection device 10 performs an operation of subtracting the R signal in the pulse wave frequency band multiplied by the ratio of the absolute intensity value in the frequency band corresponding to the noise from the G signal in the pulse wave frequency band. . As a result, when the difference between the G signal and the R signal is calculated to cancel the noise, the noise component can be reduced while suppressing the strength of the signal component in the frequency band that can be taken by the pulse wave from decreasing. .

The binarization units 17-1 to 17-n are processing units that binarize the luminance signals of the regions P _{1 to} P _n output by the difference calculation units 16-1 to 16-n. These binarization units 17-1 to 17-n have the same procedure for binarizing the luminance signal, although the input luminance signal is different in each of the areas P _{1 to} P _n . Therefore, the processing executed by the binarization unit 17-1 on behalf of the binarization units 17-1 to 17-n will be described.

As an embodiment, the binarization unit 17-1 performs binarization on whether there is a change in the luminance signal of a region P ₁ between frames. For example, the binarization unit 17-1, a luminance signal of a region P ₁ obtained in the current frame, the difference value between the luminance signal of a region P ₁ obtained in the previous frame is predetermined It is determined whether it is larger than the threshold value. As an example, zero can be used as the threshold value in order to detect a slight luminance change between frames. As another example, when noise such as blinking of illumination is mixed in the face image photographed by the camera 11, a value larger than zero can be adopted as the threshold value. At this time, the binarization unit 17-1 outputs “0” when the difference in the value of the luminance signal between the current frame and the previous frame is equal to or less than the threshold value, while the current frame “1” is output when the difference in value of the luminance signal between the previous frame and the previous frame is greater than the threshold value. In the binarization units 17-2 to 17-n, similarly to the binarization unit 17-1, the luminance signals of the respective regions P _{2 to} P _n are binarized.

The measuring unit 18 is a processing unit that measures the timing of the luminance change in each of the regions P _{1 to} P _n . As an aspect, the measurement unit 18 determines whether or not there is an area where the values output from the binarization units 17-1 to 17-n have changed from “0” to “1” between frames. . At this time, if there is a region where the value has changed from “0” to “1”, it can be said that the rising edge of the luminance change has been detected in the current frame, and the pulse wave has propagated to the region. Can be estimated. In this case, the measurement unit 18 associates the area where the rise of the luminance change is detected with the sampling time of the face image where the rise of the luminance change of the area is detected, and saves it in an internal memory (not shown). Here, the “sampling time” refers to the time when an image is taken by the camera 11 and is determined by the frame frequency of the camera 11. For example, when the frame frequency of the camera 11 is 30 Hz, the sampling time arrives at 33 ms intervals starting from the sampling time when the camera 11 starts capturing an image.

Here, the time resolution exhibited by the pulse wave detection device 10 according to the present embodiment will be described. 4 and 5 are diagrams illustrating an example of the relationship between the luminance change and the sampling time. 4 and 5, it is assumed that the division number N is 5, and the luminance changes in the regions P ₁ , P ₂ , P ₃ , P ₄ , and P ₅ are illustrated in the order from the subject's heart. Among these, FIG. 4 also shows the magnitude of the luminance change between the sampling points. On the other hand, FIG. 5 shows a luminance change binarized to “0” or “1” at each sampling point. The solid line shown in FIGS. 4 and 5 is a luminance change generated along with the pulse wave when the pulse wave generated at a certain timing T0 is propagated, and the broken line is a timing T0 + α slightly delayed from the timing. (Where α is a time shorter than the sampling time interval of 33 ms) is assumed to be a luminance change generated along with the pulse wave when the pulse wave is propagated.

As indicated by the solid line in Fig. 4, when the timing T0, only the region _{P 1} has started the luminance change before the sampling time T2, during the remaining regions _P 2 to P ₅ in the sampling time T2~T3 The brightness change has started. In this case, as shown by the solid line shown in FIG. 5, in the region P _1, until the sampling time T1 while "0" is output, "1" is output from the sampling time T2. In the other regions P _{2 to} P ₅ , “0” is output until the sampling time T2, while “1” is output from the sampling time T3.

Further, as shown by the broken line shown in FIG. 4, when the timing T0 + alpha, luminance change between the regions _P 1 to P ₄ in the sampling time T2~T3 has start, after the remaining regions _{P 5} at the sampling time T3 The brightness change has started. In this case, as indicated by a broken line in FIG. 5, in the areas P _{1 to} P ₄ , “0” is output until the sampling time T2, while “1” is output from the sampling time T3. In the remaining regions P _5, while up to the sampling time T3 "0" is output, "1" is output from the sampling time T4. In this way, a change in α time that is smaller than the sampling time interval can be expressed as a change for each sampling time.

6 and 7 are diagrams illustrating an example of the relationship between the luminance change and the sampling time. 6 and 7, similarly to FIGS. 4 and 5, the division number N is 5, and the luminance changes in the regions P ₁ , P ₂ , P ₃ , P ₄ , and P ₅ are in order from the subject's heart. It is assumed that it is illustrated. Among these, FIG. 6 also shows the magnitude of the luminance change between the sampling points. On the other hand, FIG. 7 shows a luminance change binarized to “0” or “1” at each sampling point. The solid line, the broken line, the alternate long and short dash line, and the thin line shown in FIGS. 6 and 7 are luminance changes that occur with the pulse wave when the pulse wave generated at the timings T0, T0 + α, T0 + 2α, and T0 + 3α propagates, respectively. I will do it.

As indicated by the solid line in Fig. 6, when the timing T0, the luminance variation between the regions P ₁ to P ₄ in the sampling time T1~T2 has start, the luminance changes after the remaining regions P ₅ at the sampling time T2 Has started. In this case, as indicated by the solid line in FIG. 7, in the areas P _{1 to} P ₄ , “0” is output until the sampling time T1, while “1” is output from the sampling time T2. In the remaining regions P _5, while up to the sampling time T2 "0" is output, "1" is output from the sampling time T3.

Further, as indicated by the broken line in FIG. 6, when the timing is T0 + α, the luminance change starts between the sampling times T1 and T2 in the regions P _{1 to} P ₃ , and the sampling time T 2 in the regions P _{4 to} P _5. Later, the luminance change started. In this case, as indicated by the broken line shown in FIG. 7, in the areas P _{1 to} P ₃ , “0” is output until the sampling time T1, while “1” is output from the sampling time T2. In the remaining regions P _{4 to} P ₅ , “0” is output until the sampling time T2, while “1” is output from the sampling time T3.

Further, as a chain line shown in FIG. 6, when the timing T0 + 2.alpha, regions _P 1 to P has started luminance change between the _two at sampling times T1 to T2, a region _P 3 in to P ₅ sampling time T2 After this, the luminance change starts. In this case, as indicated by the alternate long and short dash line in FIG. 7, in the areas P _{1 to} P ₂ , “0” is output until the sampling time T1, while “1” is output from the sampling time T2. In the remaining regions P _{3 to} P ₅ , “0” is output until the sampling time T2, while “1” is output from the sampling time T3.

Further, as the thin line shown in FIG. 6, when the timing T0 + 3.alpha., Luminance change between the regions _{P 1} sampling time T1~T2 has start, after the remaining region _P 2 to P ₅ in the sampling time T2 The brightness change has started. In this case, as the thin line shown in FIG. 7, in the region P _1, until the sampling time T1 while "0" is output, "1" is output from the sampling time T2. In the remaining regions P _{2 to} P ₅ , “0” is output until the sampling time T2, while “1” is output from the sampling time T3.

FIG. 8 shows a summary of luminance changes for each of the regions P _{1 to} P ₅ over these timings T0 to T0 + 3α. FIG. 8 is a diagram illustrating an example of a time difference in luminance change timing illustrated in FIG. Figure 8 is shown over the farthest place in region P ₅ located detects the rise of the luminance change in the 1-frame state timing T0~T0 + 3α binary before sampling time T2 of the sampling time T3 from the subject's heart Has been.

As shown in FIG. 8, the timing T0, the area P at the time point of the sampling time T2 immediately before the rising of the brightness change is detected at _5, all the rise of luminance change is other than the region P ₅ regions P ₁ to P ₄ Has been detected. Further, at the timing T0 + alpha, no luminance change in a region P ₄ where the rising of the previous timing T0 in luminance variation has been detected at the sampling time T2, in the region P ₁ to P ₃ is detected the rise of luminance change Yes. In this example, the pulse wave propagates through one sampling, that is, between the regions P _{1 to} P ₅ during the frame period, that is, between the four regions. Propagating between them is 1/4 of the frame period. Therefore, since the manner of change by one region differs between the timing T0 and the timing T0 + α, it means that there is a time difference of ¼ of the frame period between the timing T0 and the timing T0 + α. That is, in this example, α is ¼ period.

Further, at the timing T0 + 2α, there is no luminance change in the areas P _{3 to} P ₄ where the rising of the luminance change was detected at the timing T0 at the time of the sampling time T2, and the rising of the luminance change in the areas P _{1 to} P _2. It has been detected. This means that there is a time difference between the timing T0 and the timing T0 + 2α in a period in which the rise of the luminance change is 2/4 of the frame period. Further, at the timing T0 + 3.alpha., No change in luminance in the region P ₂ to P ₄ of the rise of the previous timing T0 in luminance variation has been detected at the sampling time T2, detects the rise of the luminance change in only the area P ₁ ing. This means that there is a time difference between the timing T0 and the timing T0 + 3α in a period in which the rise of the luminance change is 3/4 of the frame period.

In this way, by comparing the time difference of the luminance change timing between the plurality of regions, the timing at which the pulse wave propagates can be measured with a time resolution equal to or less than the frame period. For example, in the example of FIG. 5, since the rise of the luminance change is shifted by one sampling between the regions P ₁ and P ₅ , one sampling can be divided into four, resulting in a fourfold improvement in time resolution. Can be made. For example, when the frame frequency of the camera 11 is 30 Hz, the time resolution can be improved to about 8 ms (≈33 ms / 4), which is 1/4 of the frame period 33 ms.

Calculating unit 19 is a processing unit that calculates a delay amount of the pulse wave of each region P ₁ to P _n from the time difference between the timing of the luminance change area P ₁ to P _n. As one aspect, the calculation unit 19 starts calculating the delay amount of the pulse wave when the rising edge of the luminance change is detected in the Nth region _Pn . For example, the calculation unit 19 counts the number of areas in which “1” is set in the current frame among the areas in which “0” is set in the previous frame, that is, the number of areas in which the rising of the luminance change is detected. Count m. Here, assuming that the delay amount of the pulse wave in each region is equal, the delay amount of one region is “d”, the frame period which is the period of one frame is “Ts”, and the rise of the luminance change between frames is detected. The delay amount d can be calculated by the following calculation formula (1), where “m” is the number of the regions that have been set. The calculation unit 19 calculates a pulse wave delay amount d by substituting known “Ts” and “m” into the following calculation formula (1).

  d = Ts / m (1)

FIG. 9 is a diagram illustrating an example of the relationship between the luminance change and the sampling time. FIG. 10 is a diagram illustrating an example of a time difference in luminance change timing illustrated in FIG. 9. 9, it is assumed that the division number N is 11, and the luminance changes in the regions P ₁ , P ₂ , P ₃ ,..., P ₁₁ are illustrated in order from the subject's heart. FIG. 9 shows the magnitude of the luminance change between the sampling points. FIG. 10 shows a binary state at the sampling times T1 and T2.

As shown in FIG. 9, only the region P ₁ has started the luminance change before the sampling time T1, in the region P ₂ to P _9, and the start of the luminance change between the sampling time T1~T2 , the luminance change is initiated during the region _P 10 _{to P 11} at the sampling time T2 to T3. In this case, as shown in FIG. 10, the sampling time T1, the output is "0" of the region _P 2 _{to P 11} outputs while a "1" in the area _{P 1,} and the rise of only the luminance change area _{P 1} Is detected. Further, the sampling time T2, the output is "0" in the area _P 10 _{to P 11} output regions _P 1 to P ₉ While becomes "1", to the region _P 2 to P ₉ Following region _{P 1} The rising edge of the luminance change is detected.

Under such circumstances, the calculation unit 19 counts the number m of areas in which the rising of the luminance change is detected between the sampling times T1 and T2. That is, the areas where “0” is output in the frame of the sampling time T1 are areas P _{2 to} P ₁₁ excluding the area P ₁ . Of these areas P _{2 to} P ₁₁ , areas where “1” is output in the frame at the sampling time T ₂ are areas P _{2 to} P ₉ . For this reason, eight of the areas P _{2 to} P ₉ in which the rising of the luminance change is detected at the sampling time T ₂ are counted as “m”.

  When the number m becomes known in this way, the pulse wave delay amount d in one region can be calculated by substituting known “Ts” and “m” into the above-described calculation formula (1). That is, when the frame period Ts (= 33 ms) and the number m (= 8) are substituted into the above calculation formula (1), the delay amount d can be calculated as 4.2 ms by the calculation of “33/8”.

  Thus, if the delay amount d is obtained, the length of the short side of the region, that is, the length of the region in the vertical direction is also known, so the length of the short side of the region is determined from the scale on the image to the scale of the real space. For example, the amount of movement M when the required time is the delay amount d can be calculated by converting into meters. Therefore, the pulse wave propagation velocity can also be calculated by dividing the movement amount M by the delay amount d.

  If the delay amount d is obtained, the time at which the pulse wave propagates for each region exceeds the time resolution of the camera 11 using the sampling time stored in association with each region by the measurement unit 18. It can also be determined accurately. For example, the region where the delay of the pulse wave is the longest from the heart, that is, the region where the number assigned to P is the maximum among the regions where the rise of the luminance change is detected at the same sampling time is specified as the reference region. . Then, it is assumed that the pulse wave has propagated to the reference region at the time of sampling time ± 0, and the global time corresponding to the sampling time is set as the pulse wave propagation time. Then, the pulse wave propagation time is calculated by “pulse wave propagation time of the reference region−distance from the reference region × delay amount d” for the region closer to the heart than the reference region. The pulse wave propagation time can be calculated by “pulse wave propagation time of region + distance from reference region × delay amount d” ”. Note that the distance from the reference region refers to the number of regions in which the pulse wave propagation time is calculated being separated from the reference region.

  Furthermore, if the pulse wave propagation time is obtained, the pulse period and fluctuation of the pulse period can also be calculated. For example, the pulse wave propagation time can be calculated sequentially in the same region, and the pulse cycle can be calculated by subtracting the pulse wave propagation time calculated immediately before from the pulse wave propagation time calculated this time. The fluctuation of the pulse period can also be calculated by calculating the difference between the periods.

  The pulse wave delay amount, propagation speed, propagation time, pulse wave cycle fluctuation and pulse waveform obtained in this way are output to an arbitrary output destination including a display device (not shown) of the pulse wave detection device 10. Can do. For example, a measurement program that measures blood vessel age, blood pressure, etc. using pulse wave propagation speed, diagnosis of autonomic nerve function from fluctuations in pulse rate and pulse cycle, diagnosis of heart disease etc. from pulse waveform When the program is installed in the pulse wave detection device 10, a measurement program or a diagnostic program can be output. In addition, a server device that provides a measurement program or a diagnostic program as a Web service can be used as an output destination. Further, a terminal device used by a person concerned of the user who uses the pulse wave detection device 10, for example, a caregiver or a doctor, can be used as the output destination. This also enables monitoring services outside the hospital, for example, at home or at home. Needless to say, the measurement results and diagnostic results of the measurement program and the diagnostic program can also be displayed on the terminal devices of the parties concerned including the pulse wave detection device 10.

  The acquisition unit 12, the first dividing unit 13g, the second dividing unit 13r, the statistical processing units 14g1 to 14gn and 14r1 to 14rn, the BPFs 15g1 to 15gn and 15r1 to 15rn, and the difference calculating unit 16 -1 to 16-n, binarization units 17-1 to 17-n, measurement unit 18, and calculation unit 19 detect pulse waves in a CPU (Central Processing Unit), MPU (Micro Processing Unit), or the like. This can be realized by executing the program. Each functional unit described above can also be realized by a hard wired logic such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

  Further, a semiconductor memory element or a storage device can be adopted for the image memory 11a. For example, examples of the semiconductor memory device include a video random access memory (VRAM), a random access memory (RAM), a read only memory (ROM), and a flash memory. Further, examples of the storage device include storage devices such as a hard disk and an optical disk.

[Process flow]
Subsequently, the flow of processing of the pulse wave detection device according to the present embodiment will be described. FIG. 11 is a flowchart illustrating a procedure of delay amount calculation processing according to the first embodiment. This delay amount calculation process is a process that is started every time an image is stored in the image memory 11a by the camera 11 and repeatedly executed until no image is stored in the image memory 11a. Note that when an interruption operation is received via an input device (not shown) or the like, the delay amount calculation process can be stopped.

  As shown in FIG. 11, the acquisition unit 11 acquires an image stored in the image memory 11a (step S101). Subsequently, the first dividing unit 13g or the second dividing unit 13r sets a measurement range of a luminance change determined by face parts extracted from the face image, for example, eyes and chin (step S102).

  In addition, the first dividing unit 13g and the second dividing unit 13r set the measurement range set in step S102 for each face image having a G component pixel value and each face image having an R component pixel value. The image of the corresponding part is divided into N areas along the propagation direction of the pulse wave (step S103). As a result, the image corresponding to the measurement range in the face image having the G component pixel value is divided into N areas, and the portion corresponding to the measurement range in the face image having the R component pixel value. Are divided into N regions.

Then, the statistical processing units 14g1 to 14gn and 14r1 to 14rn calculate the representative value of the area by averaging the luminance values of the pixels included in the area for each wavelength component (step S104). Specifically, the statistical processing unit 14g1~14gn calculates the representative value of the G component in the region _{P 1,} a representative value of the G component in the region _{P 2,} · · ·, a representative value of the G component in the region _{P n.} Further, the statistical processing unit 14r1~14rn calculates the representative value of the R component of the region _{P 1,} a representative value of the R component of the region _{P 2,} · · ·, a representative value of the R component of the region _{P n.}

Subsequently, BPF 15g1 to 15gn and 15r1 to 15rn have a pulse wave frequency band of 0.7 Hz to less than 4 Hz that can be taken by a wavelength component and a region, and a frequency band of 42 bpm to less than 240 bpm when converted per minute. A signal component is extracted (step S105). Specifically, BPF15g1 extracts the G signal of the pulse wave frequency band regions _{P 1,} BPF15g2 extracts the G signal of the pulse wave frequency band regions _{P 2,} further, BPF15gn the region _{P n} The G signal in the pulse wave frequency band is extracted. Further, BPF15r1 extracts the R signal of the pulse wave frequency band regions _{P 1,} BPF15r2 extracts the R signal of the pulse wave frequency band regions _{P 2,} BPF15rn the region _{P n} of the pulse wave frequency band R signal is extracted.

Then, the difference calculation unit 16-1 to 16-n is for each region, and the G signal of the pulse wave frequency band of the output regions _P 1 to P _n by BPF15g1～15gn, region _{P 1} output by BPF15r1~15rn The noise is canceled by calculating the difference from the R signal in the pulse wave frequency band of ~ _Pn (step S106). Thus, the result of region P ₁ to P _n noise component contained in the G signal of the pulse wave frequency band is canceled by the R signal of the pulse wave frequency band regions P ₁ to P _n, region P ₁ to P _A luminance signal with a high S / N ratio can be obtained for each _n .

Thereafter, the binarization units 17-1 to 17-n binarize the luminance signals of the regions P _{1 to} P _n output by the difference calculation units 16-1 to 16-n (Step S107).

  Subsequently, the measurement unit 18 determines whether there is an area where the values output from the binarization units 17-1 to 17-n have changed from “0” to “1” between frames, that is, It is determined whether or not a rise in luminance change has been detected in the region (step S108).

Here, when the rise of the luminance change is detected in any _one of the regions P _{1 to} P _n (step S108 Yes), it can be estimated that the pulse wave has propagated to the region. In this case, the measurement unit 18 associates the area where the rise of the brightness change is detected with the sampling time of the face image where the rise of the brightness change of the area is detected (step S109). Save to

Thereafter, when the region where the rising edge of the luminance change is detected is the Nth region, that is, when the region _Pn is the farthest from the heart and the delay of the pulse wave is the maximum (step S110 Yes), the following Execute the process. That is, the calculation unit 19 counts the number m of areas in which the rising of the luminance change is detected in the current frame (step S111).

After that, the calculation unit 19 calculates the pulse wave delay amount d in each of the regions P _{1 to} P _n by dividing the frame period Ts by the number m of the regions counted in step S111 (step S112). Subsequently, the calculation unit 19 calculates the pulse wave propagation speed and the pulse wave propagation time of each region using the pulse wave delay amount d calculated in step S112 (step S113).

As described above, after the pulse wave propagation speed and the pulse wave propagation time of each region are calculated, or when the rise of the brightness change is not detected in the region P _n (step S108 No or step S110 No), the process proceeds to step S101. Returning, the processing from step S101 to step S113 is repeated.

[Effect of Example 1]
As described above, the pulse wave detection device 10 according to the present embodiment divides an image obtained by photographing a subject's living body into a plurality of regions, and the pulse wave delay amount from the time difference in timing at which the luminance changes in each region. Is calculated. For this reason, in the pulse wave detection device 10 according to the present embodiment, the time resolution for measuring the timing at which the pulse wave propagates can be improved below the frame period of the moving image. Therefore, according to the pulse wave detection device 10 according to the present embodiment, it is possible to improve detection accuracy such as pulse wave delay amount, pulse wave propagation speed, pulse wave arrival time, pulse wave cycle fluctuation, and the like.

  Although the embodiments related to the disclosed apparatus have been described above, the present invention may be implemented in various different forms other than the above-described embodiments. Therefore, another embodiment included in the present invention will be described below.

[Application Example 1]
In the first embodiment, the calculation method for calculating the delay amount of the pulse wave has been described under the assumption that the delay amount of the pulse wave is constant between the regions P _{1 to} P _n. The amount of delay of the pulse wave can also be calculated by the method.

  That is, the pulse wave propagation speed varies depending on the part of the living body. For this reason, even if the length of the propagation direction of the pulse wave in each region is constant, the delay amount of the region is not always the same. Therefore, in order to further improve the detection accuracy of the delay amount of each region, the pulse wave detection device 10 uses a frequency at which each region becomes the rising change point of the luminance change, The amount of delay can also be calculated.

  For example, the pulse wave detection device 10 binarizes the luminance signal of each region over a plurality of pulse wave periods. After that, the pulse wave detection device 10 calculates the number Ni of the change point region located closest to the heart that is the start point of the pulse wave in the region where the rise of the luminance change is detected in the same sampling time. Count. Specifically, when the number of changes in luminance between the area i and the area i + 1 is “Ni”, the frame period is “Ts”, and the number of delay areas for one period is “m”, The amount of delay can be calculated by the following calculation formula (2). Note that “Σ” in the following calculation formula (2) is the sum of i = 1 to m.

  di = Ni × Ts × ΣNi (2)

  In this way, by calculating the pulse wave delay amount di of each region using the above calculation formula (2), even if the length of the propagation direction of the pulse wave of each region is constant, The delay amount of the pulse wave can be calculated with higher accuracy. That is, it can be estimated that the greater the number of times Ni that has become the change point region, the greater the delay amount of the region. In this case, the delay amount having a value larger than that of the other areas is calculated by the above calculation formula (2). On the other hand, it can be estimated that the smaller the number of times Ni that has become the change point region is, the smaller the value of the delay amount in the region is. In this case, the amount of delay having a value smaller than that of other regions is calculated by the above calculation formula (2).

[Application 2]
Further, the pulse wave detection device 10 can also calculate the delay amount of each region according to the measured value of the systolic blood pressure. In general, it is known that the pulse wave propagation velocity V1 can be approximated by the following linear expression (3) with respect to the systolic blood pressure P1. However, the slope a1 and the intercept a0 of the following linear expression (3) have different values depending on individual differences. For this reason, it is preferable to calibrate by prior measurement.

  V1 = a1 × P1 + a0 (3)

  It is possible to measure systolic blood pressure by calculating the difference in pulse wave propagation time at a plurality of sites in the subject's living body using the above-mentioned primary equation and calculating the pulse wave propagation speed from the distance between the sites. it can. However, since the pulse wave propagation speed changes with blood pressure, the amount of delay in each region also changes with blood pressure.

  Therefore, as an example, when the pulse wave detection device 10 obtains the blood pressure from the propagation velocity from the chest pulse wave to the facial pulse wave, first, the pulse wave propagation velocity is assumed assuming a normal systolic blood pressure P1. V1 is determined. After that, the pulse wave detection device 10 calculates the delay amount of each region for each region from the pulse wave propagation velocity V1, measures the pulse wave propagation time for each region of the chest and face from the delay amount for each region, The pulse wave velocity V2 is obtained from the difference in the pulse wave propagation time to the face, and the systolic blood pressure P2 is obtained from the pulse wave velocity V2. If the blood pressure P1 is different from the initially assumed blood pressure P1, the pulse wave detecting device 10 corrects the initially assumed blood pressure P1 and repeats the above calculation until the error becomes sufficiently small.

  FIG. 12 is a diagram illustrating an example of an image showing a subject. FIG. 12 shows a mode in which the area is divided into N parts for each of the chest and the face. As shown in FIG. 12, the chest is divided into N regions of a chest region 1 to a chest region Nb in order from the heart that is the starting point of the pulse wave. On the other hand, the head is divided into N areas of face area 1 to face area Nf in order from the heart that is the starting point of the pulse wave.

  FIG. 13 is a flowchart illustrating a procedure of delay amount calculation processing according to the application example 2. This process is started when a delay amount calculation instruction is given via an input device (not shown) or the like. As shown in FIG. 13, the pulse wave detection device 10 acquires the first blood pressure P1 as a reference, that is, the systolic blood pressure measured at the time of calibration from an internal main storage device, an auxiliary storage device, or an external device ( Step S301). Then, the pulse wave detection device 10 calculates the pulse wave propagation velocity V1 from the first blood pressure P1 acquired in step S301 (step S302).

  Subsequently, the pulse wave detection device 10 uses the pulse wave propagation velocity V1 calculated in step S302, and among the plurality of regions, the first region close to the heart from which the pulse wave starts, for example, each region 1 of the chest A delay amount Dbi of .about.Nb is calculated (step S303). The delay amount Dbi can be calculated by calculating by dividing Lbi by the pulse wave propagation velocity V1, that is, “Lbi / V1” when the distance from the lower end of the chest region 1 to the upper end of the chest region i is “Lbi”. “I” is a natural number of 1 to Nb.

  Then, the pulse wave detection device 10 measures the rise of the luminance change of each region 1 to Nb of the first part from the image acquired by the acquisition unit 12 (step S304). In addition, the pulse wave detection device 10 uses the delay amount Dbi of each region 1 to Nb calculated in step S303 and the sampling time of each region 1 to Nb in which the rise of the luminance change is measured in step S304. Then, the pulse wave propagation time Tb of the region 1 of the first part is calculated (step S305).

  Further, the pulse wave detection device 10 uses the pulse wave propagation velocity V1 calculated in step S302, and a second part farther than the first part from the heart that is the starting point of the pulse wave among a plurality of parts, for example, The delay amount Dfi of each of the head regions 1 to Nf is calculated (step S306). The delay amount Dfi can be calculated by a calculation of dividing Lfi by the pulse wave propagation velocity V1, that is, “Lfi / V1” when the distance from the lower end of the face area 1 to the upper end of the face area i is “Lfi”. “I” is a natural number of 1 to Nf.

  Then, the pulse wave detection device 10 measures the rise of the luminance change in each of the regions 1 to Nf of the second part from the image acquired by the acquisition unit 12 (Step S307). In addition, the pulse wave detection device 10 uses the delay amount Dfi of each of the regions 1 to Nf calculated in step S306 and the sampling time of each of the regions 1 to Nf in which the rise of the luminance change is measured in step S307. Then, the pulse wave propagation time Tf of the region Nf of the second part is calculated (step S308).

  Thereafter, the pulse wave detection device 10 calculates the pulse wave propagation time Tb of the first region 1 calculated in step S305 and the pulse wave propagation time Tf of the second region Nf calculated in step S308. Is used to calculate the delay amount Tbf from the first part to the second part (step S309). The delay amount Tbf from the chest to the face can be calculated by subtracting the pulse wave propagation time Tb of the chest region 1 from the pulse wave propagation time Tf of the head region Nf, that is, “Tf−Tb”.

  Then, the pulse wave detection device 10 calculates the pulse from the delay amount Tbf from the first part to the second part calculated in step S309 and the distance Lbf from the first part to the second part. A wave propagation velocity V2 is calculated (step S310). The pulse wave velocity V2 is calculated by dividing the distance Lbf from the first part to the second part shown in FIG. 12 by the delay amount Tbf from the first part to the second part, that is, “Lbf / Tbf ". Then, the pulse wave detection device 10 calculates the second blood pressure P2 using the pulse wave propagation velocity V2 calculated in step S310 (step S311). This can be calculated by substituting the pulse wave propagation velocity V2 into the primary equation (3) instead of the pulse wave propagation velocity V1.

  Then, the pulse wave detection device 10 determines whether or not the error between the first blood pressure P1 acquired in step S301 and the second blood pressure P2 calculated in step S311 is within a predetermined appropriate range. (Step S312).

  Here, the pulse wave velocity V2 is calculated using the above-described pulse wave velocity V1, but is not a value obtained only from the reference first blood pressure P1, but in each region of the chest and head. It is also determined using the sampling time at which the rise of the luminance change is measured. For this reason, the pulse wave propagation velocity V2 is highly likely to be converged to the accurate pulse wave propagation velocity by the measured value. Therefore, as the error between the second blood pressure P2 calculated using the pulse wave velocity V2 and the first blood pressure increases, the initially assumed pulse wave velocity V1 becomes more dissimilar from the actual pulse wave velocity. Can be regarded as doing. In this case, there is a high possibility that the delay amount of the pulse wave in each region is different from the actual value.

  Therefore, when the error between the first blood pressure P1 and the second blood pressure P2 is outside the appropriate range (step S312 No), the pulse wave detection device 10 executes the following process. That is, the pulse wave detection device 10 replaces the pulse wave propagation velocity V2 calculated in step S310 with the pulse wave propagation velocity V1, and replaces the second blood pressure P2 calculated in step S311 with the first blood pressure P1 (step S313). ), And repeatedly executes the processing from step S303 to step S312 described above.

  Finally, when the error between the first blood pressure P1 and the second blood pressure P2 is within an appropriate range (step S312 Yes), there is a possibility that the delay amount of the pulse wave in each region is different from the actual value. Can be considered low. In this case, the pulse wave delay amount, the pulse wave velocity V1 or V2, and the blood pressure P1 or P2 are output to a predetermined output destination, and the process is terminated.

  In the flowchart shown in FIG. 13, the case where the processing of Step S306 to Step S308 is executed after the processing of Step S303 to Step S305 is illustrated, but these processing should be executed in a reversed order. You can also run in parallel.

  In this manner, the pulse wave detection device 10 can accurately calculate the delay amount of each region using the systolic blood pressure.

[Division number]
The time resolution can be improved as the value of the division number N of the region is increased. However, when the value is excessively increased, the SN ratio may be lowered. For this reason, for example, the delay amount described in the first and second embodiments is calculated by setting the number of divisions to such a number that at least one region where a deviation occurs in the rise of the luminance change between frames is generated. realizable.

[input signal]
In the first embodiment and the second embodiment, the case where two types of R signal and G signal are used as input signals is illustrated. However, any type of signal and any number of signals having different light wavelength components may be used. A number of signals can be input signals. For example, two signals of any combination among signals having different optical wavelength components such as R, G, B, IR, and NIR can be used, or three or more signals can be used.

[Other implementation examples]
In the first embodiment, the case where the pulse wave detection device 10 executes the above-described pulse wave detection processing in a stand-alone manner is illustrated, but it can also be implemented as a client server system. For example, the pulse wave detection device 10 may be implemented as a Web server that provides a pulse wave detection service, or may be implemented as a cloud that provides a pulse wave detection service by outsourcing. As described above, when the pulse wave detection device 10 operates as a server device, a mobile terminal device such as a smartphone or a mobile phone or an information processing device such as a personal computer can be accommodated as a client terminal. When an image showing the face of the subject is acquired from these client terminals via the network, the pulse wave detection process is executed, and the pulse wave detection result and the diagnosis result made using the detection result are responded to the client terminal By doing so, a pulse wave detection service can be provided.

[Distribution and integration]
In addition, each component of each illustrated apparatus does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured. For example, each functional unit included in the pulse wave detection device 10 may be connected as an external device of the pulse wave detection device 10 via a network.
Moreover, you may make it implement | achieve the function of said pulse-wave detection apparatus 10 by having each function part each in another apparatus, and being network-connected and cooperating.

[Pulse wave detection program]
The various processes described in the above embodiments can be realized by executing a prepared program on a computer such as a personal computer or a workstation. In the following, an example of a computer that executes a pulse wave detection program having the same function as that of the above-described embodiment will be described with reference to FIG.

  FIG. 14 is a schematic diagram illustrating an example of a computer that executes a pulse wave detection program according to the first and second embodiments. As illustrated in FIG. 14, the computer 100 includes an operation unit 110a, a speaker 110b, a camera 110c, a display 120, and a communication unit 130. Further, the computer 100 includes a CPU 150, a ROM 160, an HDD 170, and a RAM 180. These units 110 to 180 are connected via a bus 140.

  As shown in FIG. 14, the HDD 170 includes the acquisition unit 12, the first dividing unit 13g, the second dividing unit 13r, the statistical processing units 14g1 to 14gn and 14r1 to 14rn shown in the first embodiment. BPF 15g1-15gn and 15r1-15rn, difference calculation units 16-1 to 16-n, binarization units 17-1 to 17-n, measurement unit 18, and calculation unit 19 The pulse wave detection program 170a to be stored is stored in advance. The pulse wave detection program 170a may be integrated or separated as appropriate, as with each component of each functional unit shown in FIG. In other words, all data stored in the HDD 170 need not always be stored in the HDD 170, and only data necessary for processing may be stored in the HDD 170.

  Then, the CPU 150 reads the pulse wave detection program 170 a from the HDD 170 and develops it in the RAM 180. Accordingly, as shown in FIG. 14, the pulse wave detection program 170a functions as a pulse wave detection process 180a. The pulse wave detection process 180a expands various data read from the HDD 170 in an area allocated to itself on the RAM 180 as appropriate, and executes various processes based on the expanded various data. The pulse wave detection process 180a includes processing executed by each functional unit shown in FIG. 1, for example, processing shown in FIG. 11 and FIG. In addition, each processing unit virtually realized on the CPU 150 does not always require that all processing units operate on the CPU 150, and only a processing unit necessary for the processing needs to be virtually realized.

  Note that the pulse wave detection program 170a is not necessarily stored in the HDD 170 or the ROM 160 from the beginning. For example, each program is stored in a “portable physical medium” such as a flexible disk inserted into the computer 100, so-called FD, CD-ROM, DVD disk, magneto-optical disk, or IC card. Then, the computer 100 may acquire and execute each program from these portable physical media. In addition, each program is stored in another computer or server device connected to the computer 100 via a public line, the Internet, a LAN, a WAN, etc., and the computer 100 acquires and executes each program from these. It may be.

DESCRIPTION OF SYMBOLS 10 Pulse wave detection apparatus 11 Camera 11a Image memory 12 Acquisition part 13g 1st division part 13r 2nd division part 14g1, 14g2, 14gn, 14r1, 14r2, 14rn Statistical processing part 15g1, 15g2, 15gn, 15r1, 15r2, 15rn BPF
16-1, 16-2, 16-n Difference calculation unit 17-1, 17-2, 17-n Binarization unit 18 Measurement unit 19 Calculation unit

Claims (10)

An acquisition unit for acquiring an image of a subject's living body captured by an imaging device;
A dividing unit for dividing the image into a plurality of regions;
A statistical processing unit that calculates a representative value of each region for each wavelength component by executing predetermined statistical processing for each wavelength component on the pixel values of the pixels included in each region;
A measurement unit that measures the timing at which the representative value for each region synthesized between the wavelength components changes between frames of the image;
A pulse wave detection device comprising: a calculation unit that calculates a delay amount of a pulse wave from a time difference between timings at which representative values change in each region. A binarization unit that binarizes the representative value according to whether the representative value changes between frames of the image;
The calculation unit includes:
The pulse wave detection device according to claim 1, wherein the number of regions where binary values are inverted between frames of the image is calculated as a time difference between timings at which representative values change in each region. The calculation unit includes:
The pulse wave detection device according to claim 1, wherein the delay amount between the regions is calculated by distributing the frame period of the image by the number of regions where the binary value is inverted between the frames of the image. The counter further includes a counting unit that counts the number of times that the region of the change point located closest to the heart that is the starting point of the pulse wave in the region where binary inversion is detected in the same frame for each region,
The calculation unit includes:
4. The pulse wave detection according to claim 3, wherein the amount of delay between the regions is calculated by changing the amount of distribution of the frame period of the image to each region according to the number of times the region becomes the change point region. apparatus. The calculation unit includes:
Further calculate the pulse wave propagation speed, pulse wave propagation time, pulse period, pulse period fluctuation, or a combination of pulse wave velocity, pulse wave propagation time, pulse period and pulse period fluctuation from the delay amount between each region. The pulse wave detection device according to any one of claims 1 to 4, wherein the device is a pulse wave detection device. A storage unit that stores the first blood pressure;
The acquisition unit
Obtaining an image including a first part and a second part farther from the heart that is the origin of the pulse wave than the first part;
The measuring unit is
For each of the first part and the second part, measure the timing at which the representative value for each region synthesized between the wavelength components changes between frames of the image,
The calculation unit includes:
The first pulse wave velocity derived from the first blood pressure is used to calculate the amount of delay between the regions where the first part is divided, and each of the parts where the second part is divided Calculate the amount of delay between areas,
While calculating the pulse wave propagation time to the first part based on the delay amount between the areas of the first part and the time difference of the timing at which the representative value changes in each area of the first part The pulse wave propagation time to the second part is calculated based on the delay amount between the areas of the second part and the time difference of the timing at which the representative value changes in each area of the second part. ,
From the difference between the pulse wave propagation time to the first part and the pulse wave propagation time to the second part, the delay amount from the first part to the second part is calculated,
A second pulse wave velocity is calculated from a delay amount from the first part to the second part and a distance between the first part and the second part;
Repeatedly executing the process of deriving the second blood pressure until the difference between the second blood pressure derived from the second pulse wave propagation velocity and the first blood pressure falls within a predetermined range. The pulse wave detection device according to claim 5, wherein the device is a pulse wave detection device. The dividing unit is
The pulse wave detection device according to any one of claims 1 to 6, wherein an image is divided along an artery of the subject. The dividing unit is
Extracting a first face part and a second face part included in the face of the subject from the image, and dividing an image in a range determined by the first face part and the second face part; The pulse wave detection device according to claim 7. Computer
Obtain an image of the subject's living body captured by the imaging device,
Dividing the image into a plurality of regions;
Calculate the representative value for each wavelength component by executing predetermined statistical processing for each wavelength component on the pixel values of the pixels included in each region,
Measure the timing at which the representative value for each region synthesized between the wavelength components changes between frames of the image,
A pulse wave detection method comprising: executing a process of calculating a delay amount of a pulse wave from a time difference between timings at which representative values change in each region. On the computer,
Obtain an image of the subject's living body captured by the imaging device,
Dividing the image into a plurality of regions;
Calculate the representative value for each wavelength component by executing predetermined statistical processing for each wavelength component on the pixel values of the pixels included in each region,
Measure the timing at which the representative value for each region synthesized between the wavelength components changes between frames of the image,
A pulse wave detection program for executing a process of calculating a delay amount of a pulse wave from a time difference between timings at which representative values change in each region.

Images