VIDEO ANALYSIS DEVICE, VIDEO ANALYSIS METHOD, AND PULSE WAVE DETECTION DEVICE

Abstract

To obtain a pulse wave whose signal quality is high.SOLUTION: A time change detection unit (12) outputs time series data of the pixel value of a green pixel as it is. Furthermore, the time change detection unit (12) individually applies a noise reduction filter (21) and a noise reduction filter (22) to at least one of time series data of the pixel value of a red pixel and time series data of the pixel value of a blue pixel and thereby outputs time series data with a reduced noise.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The following disclosure relates to a video analysis device, a video analysis method, and a pulse wave detection device. [Background technology]

[0002] As an example of a device for detecting a pulse wave of a living body, a pulse wave detection device that detects a pulse wave based on an output value (pixel value) of an imaging element (light receiving element) is known. Patent Document 1 discloses a pulse wave measurement device including a pulse wave information acquisition unit that optically acquires pulse wave information from a part of the living body, and a pulse wave feature amount calculation unit that calculates a pulse wave feature amount based on the pulse wave information. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. WO2015 / 098977 Summary of the Invention [Problem to be solved by the invention]

[0004] The above-mentioned conventional techniques have a problem in that they are unable to remove noise contained in the frequency band of the pulse wave, making it impossible to obtain a pulse wave with good signal quality.

[0005] An object of one aspect of the present disclosure is to obtain a pulse wave with good signal quality. [Means for solving the problem]

[0006] In order to solve the above-mentioned problems, a video analysis device according to one embodiment of the present disclosure includes a region of interest detection unit that detects a region of interest of a living body based on a predetermined algorithm from a video of the living body captured by an imaging device having at least red, blue, and green color filters and an imaging unit, a time change detection unit that detects time series data of each pixel value of red pixels, green pixels, and blue pixels among a plurality of pixels included in the region of interest, and a pulse wave detection unit that detects a pulse wave of the living body based on an output from the time change detection unit, and the time change detection unit is configured to output the time series data of the pixel values ​​of the green pixels as is, and to apply a noise reduction filter to at least one of the time series data of the pixel values ​​of the red pixels and the time series data of the pixel values ​​of the blue pixels, thereby outputting the time series data with noise reduced. Effect of the Invention

[0007] According to one aspect of the present disclosure, a pulse wave with good signal quality can be obtained. [Brief description of the drawings]

[0008] [Figure 1] 1 is a block diagram showing a configuration of a main part of a pulse wave detecting device according to a first embodiment. [Diagram 2] 4A and 4B are diagrams illustrating an example of the arrangement of red, green, and blue color filters in a color filter portion. [Diagram 3] 4 is a diagram showing an example of a region of interest detected by a region of interest detection unit; FIG. [Figure 4] 11A and 11B are diagrams illustrating an example of an interpolation process for missing pixel values ​​in each color channel image. [Diagram 5] 13 is a diagram showing an example of time-series data of R pixel values, G pixel values, and B pixel values ​​in a region of interest detected from the region of interest. FIG. [Figure 6] FIG. 13 is a diagram showing frequency components of each pixel value after fast Fourier transform is applied to time series data of R pixel values, G pixel values, and B pixel values ​​in a region of interest detected from the region of interest. [Figure 7] 6 is a diagram showing an example of a pulse wave detected based on the time series data of R pixel values, G pixel values, and B pixel values ​​shown in FIG. 5. FIG. [Figure 8] FIG. 6 is a diagram showing an example of a pulse wave detected based on time series data of R pixel values, G pixel values, and B pixel values ​​shown in FIG. 5 after applying a noise reduction filter to the time series data of R pixel values ​​and the time series data of B pixel values, respectively. [Figure 9] FIG. 2 is a flowchart showing the flow of a series of processes executed by the video analysis device. [Figure 10] 1A and 1B are diagrams illustrating a Bayer array of color filters and another method for generating a full-color image. [Figure 11] FIG. 13 is a block diagram showing a configuration of a main part of a modified example of a pulse wave detecting device. [Figure 12] FIG. 13 is a block diagram showing a configuration of a main part of a modified example of a pulse wave detecting device. [Figure 13] 10 is a block diagram showing the configuration of a main part of a pulse wave detecting device according to a second embodiment. FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0009] [Embodiment 1] A pulse wave detection device 1 according to the first embodiment will be described below. For convenience of explanation, the same reference numerals will be given to components having the same functions as those described in the first embodiment in the following embodiments, and the explanations thereof will not be repeated. For simplicity, the explanations of matters similar to those in the known art will also be omitted as appropriate.

[0010] Please note that each configuration and each numerical value described in this specification is merely an example unless otherwise specified. Therefore, unless otherwise specified, the positional relationship of each member is not limited to the example of each figure. Also, please note that each drawing is for explaining the shape, structure, and positional relationship of each member in a schematic manner, and is not necessarily drawn as it actually is. In this specification, the description "A to B" regarding two numbers A and B means "greater than or equal to A and less than or equal to B" unless otherwise specified.

[0011] (Overview of Pulse Wave Detection Device 1) FIG. 1 is a block diagram showing the configuration of the main parts of a pulse wave detection device 1. The pulse wave detection device 1 detects the pulse wave of a living organism 900 (e.g., a human). The pulse wave detection device 1 is an example of a non-contact type pulse wave detection device. Hereinafter, the pulse wave of the living organism H will be simply referred to as the pulse wave. The pulse wave detection device 1 includes a video analysis device 10 and an imaging device 50 (e.g., a camera). The video analysis device 10 and the imaging device 50 are connected so as to be able to communicate with each other.

[0012] The imaging device 50 captures a video (moving image) showing the living body H. Specifically, the imaging device 50 captures images (still images) that constitute the video for each frame period. In the first embodiment, the imaging device 50 is an RGB (Red, Green, Blue) camera. Therefore, the video in the first embodiment is an RGB color video. The imaging device 50 includes a color filter unit 51 and an imaging unit 52.

[0013] The color filter unit 51 includes at least a red color filter 511R, a green color filter 511G, and a blue color filter 511B. The color filter unit 51 may include other color filters, for example, a yellow color filter or a near-infrared color filter. The imaging unit 52 includes a plurality of imaging elements 520 arranged in an array. The imaging elements 520 are known image sensors. The imaging elements 520 receive light (e.g., external light) that arrives from the living body 900 and passes through the color filter unit 51.

[0014] The imaging element 520 is also referred to as a pixel of the imaging device 50. For this reason, in this specification, the imaging element 520 is also referred to as an imaging pixel. W×H imaging pixels are arranged in the imaging section 52. W represents the number of imaging pixels in the horizontal direction (imaging pixels per row), and H represents the number of imaging pixels in the vertical direction (imaging pixels per column). The imaging pixels output, as pixel values ​​(gradation values), signal values ​​(e.g., current values) of electric signals indicating the intensity (e.g., luminance) of light received by the imaging pixels themselves. More specifically, the imaging pixels output pixel values ​​as analog values.

[0015] In the imaging device 50, an AD converter (not shown) converts the pixel values ​​as analog values ​​into digital values. Then, the imaging device 50 outputs a digital image (an image having digital pixel values). As an example, the pixel values ​​in the first embodiment are sampled as 8-bit digital values. Therefore, the pixel values ​​in the first embodiment are in the range of 0 to 255.

[0016] 2 is a diagram showing an example of the arrangement of red color filters 511R, green color filters 511G, and blue color filters 511B in the color filter section 51. In the imaging device 50, the red color filters 511R, green color filters 511G, and blue color filters 511B are arranged in one-to-one correspondence with the imaging elements 520. One red color filter 511R is arranged so as to cover one imaging element 520. One green color filter 511G is arranged so as to cover another imaging element 520. One blue color filter 511B is arranged so as to cover yet another imaging element 520.

[0017] 2, in the first embodiment, a red color filter 511R, a green color filter 511G, and a blue color filter 511B are arranged in a Bayer array. In the color filter section 51 in the example of FIG. 2, color filter units 510 corresponding to four (two in the horizontal direction × two in the vertical direction) imaging pixels are defined.

[0018] One color filter unit 510 includes (i) one red color filter 511R, (ii) two green color filters 511G (green color filters 511G1 and 511G2), and (iii) one green color filter 511B. In the color filter section 51, the multiple color filter units 510 are repeatedly arranged in the horizontal and vertical directions. In this manner, in the color filter section 51, the red color filter 511R, the green color filter 511G, and the blue color filter 511B are arranged in a mosaic pattern (more specifically, in a checkered pattern) with a number ratio of 1:2:1.

[0019] In this specification, an imaging pixel corresponding to the red color filter 511R is referred to as a red imaging pixel. One red imaging pixel receives light (red light) that has passed through the red color filter 511R, and outputs a red pixel value (R pixel value) that indicates the intensity of the red light. Also, an imaging pixel corresponding to the green color filter 511G is referred to as a green imaging pixel. One green imaging pixel receives light (green light) that has passed through the green color filter 511G, and outputs a green pixel value (G pixel value) that indicates the intensity of the green light. Also, an imaging pixel corresponding to the blue color filter 511B is referred to as a blue imaging pixel. One blue imaging pixel receives light (blue light) that has passed through the blue color filter 511B, and outputs a blue pixel value (B pixel value) that indicates the intensity of the blue light.

[0020] The imaging unit 52 generates (i) an R channel image IMGR (an image showing the distribution of R pixel values ​​output by a plurality of red imaging pixels), (ii) a G channel image IMGRG (an image showing the distribution of G pixel values ​​output by a plurality of green imaging pixels), and (iii) a B channel image IMGB (an image showing the distribution of B pixel values ​​output by a plurality of blue imaging pixels) for each video frame period. IMGR, IMGG, and IMGB all have a resolution W×H. Then, the imaging unit 52 generates an image IMG as a mosaic-like RGB color image by combining IMGR, IMGG, and IMGB. In other words, the imaging unit 52 generates IMG as a RAW image.

[0021] As described above, the imaging unit 52 generates IMG as a RAW image for each frame constituting the video. The imaging unit 52 then generates a video (RAW video) by arranging each generated frame in the order of frame number. The imaging unit 52 supplies the generated video to the video analysis device 10.

[0022] Video analysis device 10 includes region of interest detection unit 11, time change detection unit 12, and pulse wave detection unit 13. Time change detection unit 12 includes two noise reduction filters 21 and 22. Both noise reduction filters 21 and 22 are low-pass filters (noise reduction filters) that reduce noise. Time change detection unit 12 may include other noise reduction filters, such as an average filter, a Gaussian filter, a median filter, a maximum value filter, or a minimum value filter.

[0023] Video analysis device 10 detects a pulse wave by analyzing the video acquired from imaging device 50. More specifically, video analysis device 10 detects a pulse wave by analyzing each frame of the video (each of a plurality of IMGs).

[0024] Region of interest detection unit 11 detects a region of interest (ROI) from IMG based on a predetermined algorithm (e.g., using a known algorithm). A region of interest is a partial region of IMG in which a predetermined part of a living body 900 is shown. FIG. 3 is a diagram showing an example of a region of interest detected by region of interest detection unit 11. In the following description, the region of interest in the example of FIG. 3 is referred to as region of interest 910.

[0025] The IMG in the example of Fig. 3 is an image showing the entire face of the living body 900. In the example of Fig. 3, one cheek of the face is set in advance as a predetermined part. First, region of interest detection unit 11 detects a face region (a region showing the face of the living body 900) from the IMG. Then, region of interest detection unit 11 detects a region within the face region showing one cheek as region of interest 910. Region of interest 910 in the example of Fig. 3 is a rectangular region.

[0026] However, as will be apparent to those skilled in the art, the predetermined portion does not need to be limited to the cheek, but may be any portion from which information regarding blood vessels of the living body 900 can be detected. Other examples of the predetermined portion include the forehead, the root of the nose, the neck, the fingertips, and the palm. The region of interest may be any image region that expresses a change in skin color of the living body over time. The number of regions of interest is not necessarily limited to one. Thus, the IMG is not limited to the example of FIG. 3.

[0027] Time change detection unit 12 detects time series data of each pixel value of red pixels, green pixels, and blue pixels among a plurality of pixels included in region of interest 910. Time series data is a set of data in which each data (pixel value, etc.) that changes with time and the time data at that time are arranged in a chronological order. Pulse wave detection unit 13 detects the pulse wave of living body 900 based on each detected time series data.

[0028] As an example, the time change detection unit 12 calculates (i) an average value Rave of R pixel values ​​included in the region of interest 910, (ii) an average value Gave of G pixel values ​​included in the region of interest 910, and (iii) an average value Bave of B pixel values ​​included in the region of interest 910, for each frame. That is, the time change detection unit 12 derives time series data for each of Rave, Gave, and Bave. Then, the pulse wave detection unit 13 may detect (derive) a pulse wave by analyzing the time series data using a known algorithm. As an example, the pulse wave detection unit 13 may detect a pulse wave using a method disclosed in International Publication WO2020 / 090348.

[0029] However, if pixel values ​​are saturated within the region of interest 910, the accuracy of pulse wave detection by pulse wave detection unit 13 may decrease. Such pixel value saturation may occur, for example, when the brightness of external light is high. Therefore, in the first embodiment, as described below, pulse wave detection unit 13 detects a pulse wave after reducing the influence of pixels whose pixel values ​​are saturated (hereinafter referred to as saturated pixels) among the pixels included in region of interest 910. Note that the process of reducing the influence of saturated pixels may be performed by time change detection unit 12.

[0030] In this specification, "saturation of pixel value" means that the pixel value is a predetermined maximum value that can be set in an image (more precisely, digital image data). As described above, in the first embodiment, the pixel value is sampled as an 8-bit digital value. Therefore, in the first embodiment, "saturation of pixel value" means that the pixel value is 255. In contrast, in the first embodiment, "non-saturation of pixel value" means that the pixel value is 0 to 254. From the above, a saturated pixel in the first embodiment is a pixel having a pixel value of 255. Therefore, in the first embodiment, the pixel value of 255 is also referred to as a saturated pixel value. Note that the pixel value in the above description refers to any one of the R pixel value, the G pixel value, and the B pixel value. In addition, although the pixel value saturation is set to a pixel value of 255, this is not limiting, and a value close to 255 may be set as a threshold value (e.g., 240) and a pixel value equal to or greater than the threshold value may be set as a saturated pixel value.

[0031] FIG. 4 is a diagram for explaining a method of aligning missing information in each pixel when color filters are arranged in a Bayer array. An example of aligning missing information in a pixel receiving light passing through a blue color filter 511B will be explained below. A blue pixel receiving light passing through a blue color filter 511B has blue information, but is missing green and red information. Therefore, a G pixel value in the blue pixel is obtained from each green pixel receiving light passing through a green color filter 511G, which is disposed on each of the top, bottom, left, and right sides of the blue pixel receiving light passing through the blue color filter 511B. Therefore, an R pixel value in the blue pixel is obtained from each red pixel receiving light passing through a red color filter 511R, which is disposed on each of the top right, top left, bottom right, and bottom left of the blue pixel receiving light passing through the blue color filter 511B. This allows three color information (pixel values) to be aligned in one pixel.

[0032] FIG. 5 is a diagram showing an example of time series data of R pixel values, G pixel values, and B pixel values ​​in the region of interest 910 detected from the region of interest 910. 1010 in FIG. 5 shows the time series data of R pixel values, 1020 in FIG. 5 shows the time series data of G pixel values, and 1030 in FIG. 5 shows the time series data of B pixel values. In each graph in FIG. 5, the vertical axis shows the pixel value (brightness value) of each pixel, and the horizontal axis shows the measurement frame (time). As shown in FIG. 5, the time series data of the G pixel values ​​has small noise, so that a pulse wave component is observed. On the other hand, the time series data of the R pixel values ​​and the time series data of the B pixel values ​​have large noise, so that a pulse wave component is not observed.

[0033] As shown in Fig. 5, the time series data of the G pixel values ​​includes luminance changes of the pixel values ​​due to the movement of the living body. Therefore, it is difficult to detect a good pulse wave based only on the time series data of the G pixel values.

[0034] Fig. 6 is a diagram showing frequency components of each pixel value after fast Fourier transform is applied to each time series data of R pixel values, G pixel values, and B pixel values ​​in the region of interest 910 detected from the region of interest 910. 1110 in Fig. 6 shows the frequency components of the R pixel value after fast Fourier transform is applied, 1020 in Fig. 6 shows the frequency components of the G pixel value after fast Fourier transform is applied, and 1030 in Fig. 6 shows the frequency components of the B pixel value after fast Fourier transform is applied. In each graph in Fig. 6, the vertical axis shows the power spectrum, and the horizontal axis shows the frequency.

[0035] As shown in a range 60 in Fig. 6, in the frequency components of the G pixel values ​​after the fast Fourier transform is applied, a power spectrum peak is observed near the frequency of the pulse wave. On the other hand, in the frequency components of the R pixel values ​​to which the fast Fourier transform is applied, a power spectrum peak is not observed near the frequency of the pulse wave. Similarly, in the frequency components of the B pixel values ​​to which the fast Fourier transform is applied, a power spectrum peak is not observed near the frequency of the pulse wave.

[0036] Fig. 7 is a diagram showing an example of a pulse wave detected based on the time series data of the R pixel value, the G pixel value, and the B pixel value shown in Fig. 5. As shown in this diagram, the detected pulse wave contains noise. Therefore, a pulse wave with poor signal quality is detected.

[0037] FIG. 8 is a diagram showing an example of a pulse wave detected based on the time series data of the R pixel value, the G pixel value, and the B pixel value after applying the noise reduction filters 21 and 22 to the time series data of the R pixel value and the time series data of the B pixel value, respectively, among the time series data of the R pixel value, the G pixel value, and the B pixel value shown in FIG. 5. As shown in this figure, noise is removed from the detected pulse wave. Therefore, a pulse wave with good signal quality is detected. The time series data of the R pixel value and the time series data of the B pixel value contain large noise as shown in FIG. 5. However, these time series data also contain a trend similar to the trend contained in the time series data of the G pixel value. Therefore, if the noise reduction filter 21 is applied to the time series data of the R pixel value and the noise reduction filter 22 is further applied to the time series data of the B pixel value, the trend contained in these time series data can be effectively removed.

[0038] The time series data of the pixel values ​​of the red pixels and the time series data of the red pixels of the blue pixels contain almost no pulse wave components. Therefore, the noise reduction filters 21 and 22 may be filters that reduce noise in a band including the frequency band of the pulse wave. This makes it possible to reduce noise in a wide frequency band contained in the time series data of the R pixel values ​​and the time series data of the B pixel values. In this way, by applying the noise reduction filters 21 and 22 to the time series data of the R pixel values ​​and the time series data of the B pixel values ​​that contain almost no pulse wave but are mixed with noise, respectively, it is possible to reduce the influence of the luminance change of the pixel values ​​caused by the body movement of the living body 900, and to detect a pulse wave with good signal quality.

[0039] FIG. 9 is a flow chart showing a series of processes executed by the video analysis device 10. In step S1, a region of interest 910 of the living body 900 is detected based on a predetermined algorithm from an image IMG in which the living body 900 is captured by the imaging device 50. In step S2, the time change detection unit 12 detects time series data of the pixel values ​​of red pixels, green pixels, and blue pixels among a plurality of pixels included in the region of interest 910, and outputs the data to the pulse wave detection unit 13. At this time, the time change detection unit 12 outputs the time series data of the pixel values ​​of the green pixels as is to the pulse wave detection unit 13. Furthermore, the time series data in which noise has been reduced is output to the pulse wave detection unit 13 by applying a noise reduction filter 21 to the time series data of the pixel values ​​of the red pixels and a noise reduction filter 22 to the time series data of the pixel values ​​of the blue pixels. In step S3, the pulse wave detection unit 13 detects the pulse wave of the living body 900 based on the output of the time change detection unit 12, i.e., the time series data of the R pixel value, the G pixel value, and the B pixel value. In step S4, video analysis device 10 determines whether or not the measurement of the pulse wave has been completed. If the determination result is true (YES in step S4), video analysis device 10 ends the execution of the series of processes shown in Fig. 9. If the determination result is false (NO in step S4), region of interest detection unit 11 returns to step S1 and detects region of interest 910 again.

[0040] FIG. 10 is a diagram for explaining a Bayer-arranged color filter 511 and another method for generating a full-color image. In this example, the video analysis device 10 obtains R, G, and B information (pixel value) of one pixel constituting a full-color image from a block of four pixels including one red pixel, two green pixels, and one blue pixel in the shaded portion shown in FIG. 10. For example, for a block of four pixels included in an image captured by the imaging device 50 and a pixel of the full-color image corresponding to the block, information of one red pixel present in the block is taken as R information of the pixel of the full-color image. Furthermore, the average value of information of two G pixels present in the block is taken as G information of the pixel of the full-color image. Furthermore, information of one B pixel present in the block is taken as B information of the pixel of the full-color image. As a result, a full-color image with the number of pixels in the height direction being H÷2 and the number of pixels in the width direction being W÷2 is obtained from an image captured by the imaging device 50 in which the number of pixels in the height direction is H and the number of pixels in the width direction is W.

[0041] In this embodiment, as a condition for applying the noise reduction filters 21 and 22, a threshold may be set for each of the signal-to-noise ratios (SN ratios) of the R pixel values ​​and the B pixel values ​​specific to the imaging device 50. In this case, the time change detection unit 12 determines whether to apply the noise reduction filter 21 to the time series data of the pixel values ​​of the red pixels based on whether the SN ratio of the R pixel values ​​included in the region of interest 910 exceeds a threshold set for the red pixels. Furthermore, the time change detection unit 12 determines whether to apply the noise reduction filter 22 to the time series data of the B pixel values ​​based on whether the SN ratio of the B pixel values ​​included in the region of interest 910 exceeds a threshold set for the blue pixels. More specifically, if the SN ratio of the R pixel values ​​included in the region of interest 910 exceeds the threshold for the red pixels, the time change detection unit 12 determines to apply the noise reduction filter 21 to the time series data of the R pixel values, and if not, determines not to apply the noise reduction filter 21. Furthermore, if the SNR of the B pixel values ​​included in the region of interest 910 exceeds a threshold value for blue pixels, the time change detection unit 12 decides to apply the noise reduction filter 22 to the time series data of the B pixel values, and if the SNR does not exceed the threshold value, it decides not to apply the noise reduction filter 22.

[0042] Alternatively, as a condition for applying the noise reduction filters 21 and 22, a threshold may be set to the sum of transmittance values ​​included in a specific wavelength range for each of the spectral sensitivities of the red color filter 511R and the blue color filter 511B specific to the imaging device 50.

[0043] (Modification) In the first embodiment, an example has been described in which the time change detection unit 12 applies the noise reduction filters 21 and 22 individually to both the time series data of R pixel values ​​and the time series data of B pixel values. However, the present invention is not limited to this, and the time change detection unit 12 may be configured to apply only the corresponding noise reduction filter 21 or 22 individually to at least one of the time series data of R pixel values ​​and the time series data of B pixel values. Such an example configuration will be described below.

[0044] 11 is a block diagram showing the configuration of the main parts of a modified example of pulse wave detection device 1. As shown in this figure, video analysis device 10 constituting pulse wave detection device 1 may be configured to include only noise reduction filter 21 corresponding to red pixels, out of noise reduction filters 21 and 22. In video analysis device 10 of this example, time change detection section 12 applies noise reduction filter 21 to time series data of pixel values ​​of red pixels, but does not apply noise reduction filter 22 to time series data of pixel values ​​of blue pixels.

[0045] Therefore, similarly to the first embodiment, the time change detection unit 12 outputs the time series data of the G pixel values ​​to the pulse wave detection unit 13 as is, and also applies the noise reduction filter 21 to the time series data of the R pixel values ​​to output the time series data of the R pixel values ​​with noise reduced to the pulse wave detection unit 13. Furthermore, unlike the first embodiment, the time change detection unit 12 outputs the time series data of the B pixel values ​​to the pulse wave detection unit 13 as is. As a result, the pulse wave detection unit 13 detects a pulse wave based on the time series data of the B pixel values ​​and the time series data of the G pixel values ​​with no noise reduction, and the time series data of the R pixel value with noise reduced. As a result, a pulse wave with good signal quality can be detected.

[0046] 12 is a block diagram showing the configuration of the main parts of a modified example of pulse wave detection device 1. As shown in this figure, video analysis device 10 constituting pulse wave detection device 1 may be configured to include only noise reduction filter 22 corresponding to blue pixels out of noise reduction filters 21 and 22. In video analysis device 10 of this example, time change detection section 12 applies noise reduction filter 22 to time series data of pixel values ​​of blue pixels, but does not apply noise reduction filter 21 to time series data of pixel values ​​of red pixels.

[0047] Therefore, similarly to the first embodiment, the time change detection unit 12 outputs the time series data of the G pixel values ​​to the pulse wave detection unit 13 as it is, and also applies the noise reduction filter 22 to the time series data of the B pixel values ​​to output the time series data of the B pixel values ​​with noise reduced to the pulse wave detection unit 13. Furthermore, unlike the first embodiment, the time change detection unit 12 outputs the time series data of the R pixel values ​​to the pulse wave detection unit 13 as it is. As a result, the pulse wave detection unit 13 detects a pulse wave based on the time series data of the R pixel values ​​and the time series data of the G pixel values ​​with no noise reduction, and the time series data of the B pixel values ​​with noise reduced. As a result, similarly to the first embodiment, a pulse wave with good signal quality can be detected.

[0048] [Embodiment 2] 13 is a block diagram showing the configuration of the main parts of a pulse wave detection device 1A according to embodiment 2. As shown in this figure, pulse wave detection device 1A includes a video analysis device 10A and an imaging device 50 (e.g., a camera). The video analysis device 10A and the imaging device 50 are connected so as to be able to communicate with each other. The configuration of the imaging device 50 is the same as that of embodiment 1, and therefore a detailed description thereof will be omitted.

[0049] Pulse wave detection unit 13A includes a pulse wave noise filter 31. Pulse wave noise filter 31 is a filter that removes noise in a band other than the frequency band of the pulse wave from the pulse wave detected by pulse wave detection unit 13 A. Pulse wave detection unit 13A removes noise from the pulse wave by applying pulse wave noise filter 31 to the detected pulse wave.

[0050] The time series data of the pixel values ​​of the red pixels and the time series data of the red pixels of the blue pixels contain almost no pulse wave components. Therefore, the noise reduction filters 21 and 22 may be filters that reduce noise in a band that includes the frequency band of the pulse wave. Furthermore, the noise reduction filters 21 and 22 may be filters that reduce noise at a frequency lower than the frequency of the noise removed by the pulse wave noise filter 31. By using these noise reduction filters 21 and 22 and the pulse wave noise filter 31, the signal quality of the pulse wave can be further improved.

[0051] In this embodiment, as in the first embodiment, it is possible to set a threshold value for determining whether or not to apply noise reduction filters 21 and 22. Note that pulse wave detection unit 13A always applies pulse wave noise filter 31 to the detected pulse wave.

[0052] [Software implementation example] The functions of pulse wave detection device 1 (hereinafter referred to as the "device") can be realized by a program for causing a computer to function as the device, and by a program for causing a computer to function as each control block of the device (in particular, each part included in video analysis device 10 or 10A).

[0053] In this case, the device includes a computer having at least one control device (e.g., a processor) and at least one storage device (e.g., a memory) as hardware for executing the program. The control device and storage device execute the program to realize each function described in each of the above embodiments.

[0054] The program may be non-transitory and may be recorded in one or more computer-readable recording media. The recording media may or may not be included in the device. In the latter case, the program may be provided to the device via any wired or wireless transmission medium.

[0055] In addition, some or all of the functions of each of the control blocks can be realized by a logic circuit. For example, an integrated circuit in which a logic circuit that functions as each of the control blocks is formed is also included in the scope of the present disclosure. In addition, the functions of each of the control blocks can be realized by, for example, a quantum computer.

[0056] Furthermore, each process described in each of the above embodiments may be executed by AI (Artificial Intelligence). In this case, the AI ​​may be executed by the control device or another device (for example, an edge computer or a cloud server).

[0057] [Additional Notes] One aspect of the present disclosure is not limited to the above-described embodiments, and various modifications are possible within the scope of the claims. An embodiment obtained by appropriately combining the technical means disclosed in each of the different embodiments is also included in the technical scope of one aspect of the present disclosure. Furthermore, by combining the technical means disclosed in each of the embodiments, new technical features can be formed. [Explanation of symbols]

[0058] 1. 1A pulse wave detector 10, 10A Video analysis device 11 Region of interest detection section 12 Time change detection section 13, 13A Pulse wave detector 21, 22 Noise reduction filter 31 Pulse wave noise filter 50 Imaging device 51 Color filter section 52 Imaging unit 520 Image sensor 510 Color filter unit 511R Red color filter 511G Green color filter 511B Blue color filter 900 Living organisms 910 Area of ​​Interest IMG Image (one frame of footage showing a living body)

Claims

1. a region of interest detection unit that detects a region of interest of a living body based on a predetermined algorithm from an image of the living body captured by an imaging device having at least red, blue, and green color filters and an imaging unit; a time change detection unit that detects time series data of each pixel value of red pixels, green pixels, and blue pixels among a plurality of pixels included in the region of interest; a pulse wave detection unit that detects a pulse wave of the living body based on an output from the time change detection unit, The time change detection unit outputs the time series data of the pixel values ​​of the green pixels as is, and also applies a noise reduction filter to at least one of the time series data of the pixel values ​​of the red pixels and the time series data of the pixel values ​​of the blue pixels, thereby outputting the time series data with noise reduced.

2. the pulse wave detection unit applies a pulse wave noise filter to the detected pulse wave to reduce noise from the pulse wave; 2. The video analysis device according to claim 1, wherein the noise reduction filter is a filter that reduces noise having a frequency lower than a frequency of noise reduced by the pulse wave noise filter.

3. 3. The video analysis device according to claim 1, wherein the noise reduction filter is a filter that reduces noise in a frequency band that includes a frequency band of the pulse wave.

4. A video analysis device according to any one of claims 1 to 3, A pulse wave detection device comprising the above imaging device.

5. a region of interest detection step of detecting a region of interest of a living body based on a predetermined algorithm from an image of the living body captured by an imaging device having at least red, blue, and green color filters and an imaging unit; a time change detection step of detecting time series data of pixel values ​​of red pixels, green pixels, and blue pixels among a plurality of pixels included in the region of interest; a pulse wave detection step of detecting a pulse wave of the living body based on an output from the time change detection step, In the time change detection step, the time series data of the pixel values ​​of the green pixels is output as is, and a noise reduction filter is applied to at least one of the time series data of the pixel values ​​of the red pixels and the time series data of the pixel values ​​of the blue pixels, thereby outputting noise-reduced time series data.

Images