Expiration visualization system, expiration visualization method, expiration evaluation system and expiration evaluation method

Abstract

The invention provides an expiration visualization system and method, an expiration evaluation system and method, and a health evaluation system and method, which have a simple structure and can reliably visualize expiration even when a front image of the face of a subject is acquired by a camera. This expiration visualization system is characterized by being provided with: an infrared camera that detects light in an infrared region radiated from the face of a subject and acquires image data; a processing unit that generates processed image data by subtracting, from first image data acquired at a first time (any selected time), second image data acquired at a second time (before any selected time) prior to the first time (any selected time); and an output unit that outputs the processed image data.

Description

Technical Field

[0001] The present invention relates to an exhalation visualization system and method, an exhalation evaluation system and method, and a health evaluation system and method. Background Art

[0002] Biometric signals such as body temperature, breathing, and heart rate fluctuations are important parameters for understanding a person's health status. The emergence of wearable devices such as smart watches makes it easy to measure these biometric signals.

[0003] However, if these biological signals can be measured contactlessly, continuous measurement can be performed without applying stress to the subject. For example, biological sensing using camera images is expected to be applied to sensorless monitoring at the bedside or remote diagnosis and treatment.

[0004] For example, non-patent document 1 discloses contactless measurement of heartbeat and respiration based on facial images taken by a CMOS camera and a thermal imaging camera. However, the disclosure of non-patent document 1 uses multiple cameras, complicating the system and increasing manufacturing costs.

[0005] Non-patent document 2 attempts to combine infrared cameras and CO 2 The breathing rate is visualized by combining bandpass filters of absorption bands. However, in the disclosure of Non-Patent Document 2, in order to distinguish exhalation from the background, the subject needs to face sideways with respect to the camera, which significantly limits the use scenarios.

[0006] Prior art literature

[0007] Non-patent literature

[0008] Non-patent literature 1: G. Sun et al, "Remote sensing of multiple vital signs using a CMOS camera-equipped infrared thermography system and its clinical application in rapidly screening patients with suspected infectious diseases", International Journal of Infectious Diseases 55 (2017) 113-117

[0009] Non-patent document 2: J. Fei et al, “Imaging Breathing Rate in the CO2 Absorption Band”, Proceedings of the 2005 IEEE 700-705 Summary of the invention

[0010] Technical problem to be solved by the invention

[0011] One of the objects of the present invention is to provide a breath visualization system and method that has a simple structure and can reliably visualize breath even when a frontal image of the face of a subject is captured by a camera. Another object of the present invention is to provide a breath evaluation system and method that is convenient by utilizing the breath visualization. Another object of the present invention is to provide a health evaluation system and method that includes the breath evaluation system and method.

[0012] Technical solutions for solving technical problems

[0013] In order to solve the above-mentioned problem, a first aspect of the present invention provides an exhalation visualization system, characterized by comprising:

[0014] an infrared camera that detects light in an infrared region radiated from the face of the subject and acquires image data;

[0015] a processing unit that subtracts second image data acquired at a second time before the first time from first image data acquired at a first time to generate processed image data; and

[0016] The output unit outputs the processed image data. The first time is a time t1 that can be arbitrarily selected. The second time is a time t2 that can be arbitrarily selected before the first time (for example, t1>t2 according to the 24-hour system).

[0017] In the exhalation visualization system of the present invention, preferably, the infrared camera detects light in the mid-infrared region.

[0018] Furthermore, in the exhalation visualization system of the present invention, it is preferable that the processing unit uses an average of image data acquired over a predetermined period before the first time as the second image data.

[0019] In addition, a second aspect of the present invention provides a breath visualization method, characterized by comprising:

[0020] A step of detecting light in the infrared region emitted from the face of the subject and acquiring image data;

[0021] Subtracting second image data acquired at a second time before the first time from first image data acquired at a first time to generate processed image data; and

[0022] The step of outputting the processed image data.

[0023] In addition, a third aspect of the present invention provides a breath evaluation system, characterized by comprising:

[0024] an infrared camera that detects light in an infrared region radiated from the face of the subject and acquires image data;

[0025] a processing unit that subtracts second image data acquired at a second time before the first time from first image data acquired at a first time to generate processed image data;

[0026] a setting unit that sets a region of interest below the nostrils of the subject in the processed image data; and

[0027] The evaluation unit calculates a flow rate of exhaled air of the subject based on a temporal change in the signal intensity in the region of interest.

[0028] In the exhalation evaluation system of the present invention, preferably, the evaluation unit calculates an opening area of ​​the nostril based on a diameter of the nostril, and calculates a flow rate of the exhaled air using the flow velocity and the opening area.

[0029] In the exhalation evaluation system of the present invention, it is preferable that the evaluation unit evaluates the energy metabolism of the subject based on the flow rate of the exhaled air with reference to data indicating a relationship between energy metabolism and the exhaled volume.

[0030] In addition, a fourth aspect of the present invention provides a breath evaluation method, characterized by comprising:

[0031] A step of detecting light in the infrared region emitted from the face of the subject and acquiring image data;

[0032] A step of subtracting second image data acquired at a second time earlier than the first time from first image data acquired at a first time to generate processed image data;

[0033] A step of setting a region of interest below the nostrils of the subject in the processed image data; and

[0034] The step of calculating the flow rate of the exhaled air of the subject based on the temporal change of the signal intensity in the region of interest.

[0035] Effects of the Invention

[0036] According to the present invention, the structure is simple, and even when a frontal image of the face of a subject is acquired by a camera, exhalation can be reliably visualized. Therefore, the present invention can be used in various scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic diagram of the system 1 and method according to the present embodiment.

[0038] Figure 2 : is a diagram showing an example of the time change of the exhalation intensity of the subject P.

[0039] Figure 3 is a flow chart showing the visual steps of exhalation.

[0040] Figure 4 It is a block diagram showing the functional structure of the system 1.

[0041] Figure 5 1 is a block diagram showing the hardware configuration of a computer 100 included in the system 1 .

[0042] Figure 6 is a flow chart showing the evaluation procedure of exhalation.

[0043] Figure 7 This is a flowchart showing the detailed steps of breath evaluation.

[0044] Figure 8 This is a diagram showing an example of a (slice-reconstructed) image formed by setting a region of interest (ROI) and arranging the ROI along the time axis.

[0045] Fig. 9 FIG. 1 is a diagram showing an example of a method of calculating the tilt from an image reconstructed by slices.

[0046] Fig.10 It is a diagram showing another method of setting ROI, reorganizing slices, and extracting the inclination.

[0047] Fig.11 This is a diagram showing an example of the calculation procedure of the exhaled volume.

[0048] Fig.12 FIG. 1 is a diagram showing an example of a method of extracting heartbeat data from a temperature image of the face of a subject person P. FIG. DETAILED DESCRIPTION

[0049] Below, while referring to the attached Figure 1The representative embodiments of the breath visualization system and method and the breath evaluation system and method of the present invention are described in detail. However, the present invention is not limited to these drawings. In addition, the drawings are used to conceptually illustrate the present invention, and therefore, in order to facilitate understanding, the size, ratio or quantity may be exaggerated or simplified as needed.

[0050] First, the exhalation visualization system and method of the present invention will be described, and then the exhalation evaluation system and method will be described. In the following description, the exhalation visualization system and the exhalation evaluation system are collectively referred to as a system 1.

[0051] 1. Visualization of exhalation

[0052] Reference Figure 1 - Figure 5 , a method and system for visualizing the exhalation of a subject P will be described. The subject P is typically a patient, but is not limited thereto.

[0053] 1-1. Exhalation visualization method

[0054] like Figure 3 As shown, visualization of exhaled breath in this embodiment includes steps such as (1) photographing with an infrared camera, (2) processing of image data, and (3) output of processed image data. Each step will be described in detail below.

[0055] (1) Photography using an infrared camera

[0056] like Figure 1 As shown, the image data of the face of the subject P is acquired via the infrared camera 20 ( Figure 3 In the present embodiment, the characteristic point is that even when the infrared camera 20 acquires a frontal image of the face of the subject P, the exhalation can be reliably visualized, but the subject P may be tilted relative to the infrared camera 20 or may face sideways.

[0057] The infrared camera 20 includes a detector for detecting light in the infrared region (wavelength 0.7 μm to 1,000 μm), and the detector may be any of a quantum infrared detector and a thermal infrared detector.

[0058] The infrared camera 20 preferably detects light in the mid-infrared region, particularly light with a wavelength of 3 μm to 5 μm. 2 The strong absorption band exists at 4.2μm-4.3μm, so the CO in the breath of the subject P can be 2 The flow is reflected in the image data.

[0059] The frame rate of the infrared camera 20 is not particularly limited as long as the exhalation of the subject person P can be sufficiently recognized, and for example, 30 fps and 60 fps can be used.

[0060] In this embodiment, it is assumed that the subject P is stationary (for example, sitting on a chair or sleeping), but the subject P may also be active (for example, may also move or exercise by walking or running). In order to track the face of the active subject P, a known tracking technology can be used.

[0061] (2) Image data processing

[0062] The processed image data ( Figure 3 Step S2). The first moment is a moment t1 that can be arbitrarily selected, for example, it can be set to the present. At this time, the first image data can be said to be current image data. The second moment is a moment t2 that can be arbitrarily selected before the first moment, that is, a moment in the past. Therefore, for example, according to the 24-hour system, t1>t2 holds true, and the second image data can be said to be past image data. In addition, some corrections or corrections can be added to the first image data and the second image data before the processed image data is generated.

[0063] In this process, it is preferred to use the average of the image data captured during the past prescribed period as the second image data. Here, for example, 10 seconds to 50 seconds can be used as the prescribed period, and 20 seconds to 45 seconds is more preferred. However, in the case where the breathing of the subject P is irregular, it is preferred to use a prescribed time of more than 20 seconds. For example, when the frame rate is 60fps, 2400 pieces of image data are generated in 40 seconds.

[0064] The average of past image data is calculated by, for example, adding all pixel values ​​in a predetermined period to each corresponding pixel and dividing the result by the total number of images, but the present invention is not limited thereto.

[0065] The generation of processed image data may be performed each time the infrared camera generates image data, or may be performed at a preset time interval. In the latter case, each time a predetermined number of image data is generated in the infrared camera, one piece of processed image data is generated.

[0066] The generation of processed image data can also be performed using, for example, free software "ImageJ" (https: / / imagej.net / ij / index.html). Of course, other software having the same type of functions can also be used.

[0067] (3) Output of processed image data

[0068] Output processed image data ( Figure 3 Step S3). As the output method, for example, display on a display such as a monitor, transmission to an external display or projector, printing, etc. can be cited. The processed image data can also be stored in a storage device.

[0069] exist Figure 2 The image shown in the upper left corner of is an example of image data displayed on a computer monitor. By displaying the images in a time series order, the flow of gas from the nostrils of the subject P can be visually grasped. For example, on the computer monitor, the exhalation movement of the subject P appears as the swing of a plurality of thin line-shaped stripes, which can be grasped as the swing of white mist.

[0070] This is Figure 2 This is also shown in the lower right graph of . That is, the signal intensity in the area A below the mouth of the subject P changes with a certain degree of regularity (ie, periodicity).

[0071] Of course, in Figure 2 The image illustrated in the upper left corner of the figure can be displayed on the display alone or in conjunction with a graph.

[0072] By using the breath visualization method of this embodiment, it is possible to visualize the breath of the subject P facing the camera. That is, even in the presence of a background such as the face of the subject P, the breath of the subject P can be visually grasped.

[0073] 1-2. Exhalation Visualization System

[0074] A system 1 for realizing the above-mentioned exhalation visualization method will be described.

[0075] like Figure 4 As shown, the system 1 includes an infrared camera 20, a processing unit 11, and an output unit 13. The system 1 for implementing the breath visualization method does not necessarily include the setting unit 15 and the evaluation unit 17. The processing unit 11 and the output unit 13 can be implemented as one function of the computer 10.

[0076] The infrared camera 20 is described in the above 1-1(1). In the present embodiment, as the infrared camera 20, FLIR A6700sc manufactured by Teledyne FLIR LLC can be used.

[0077] The main features of the FLIR A6700sc are as follows.

[0078] Resolution: 640x512 pixels

[0079] Temporal resolution: 60fps

[0080] Wavelength sensitivity: 3-5μm

[0081] Dynamic range: 14bit

[0082] Temperature resolution: 0.018°C

[0083] The processing unit 11 executes the above-mentioned step 1-1(2). In order to realize the processing unit 11, the computer 100 includes a CPU 101, a RAM 103, and a ROM 105 (see Figure 5 The CPU 101 is a circuit that functions as the processing unit 11 by reading programs and various data stored in the ROM 105 into the RAM 103 and executing them.

[0084] The output unit 13 executes the above-mentioned step 1-1(3). In order to realize the output unit 13, the computer 100 includes a communication interface 107 and an output device 111. The output device 111 includes a computer display, a projector, a head-mounted display, and a printer. In addition, the communication interface 107 is used for communication with the computer display, projector, head-mounted display, and printer as external devices.

[0085] In addition, the communication interface 107 is also used to obtain image data from the infrared camera 20. In addition, as the input device 109, a mouse, a keyboard, and a touch panel are included. The input device 109 may also include the infrared camera 20.

[0086] 2. Evaluation of exhalation

[0087] Reference Figure 4 - Fig.11 , a method and system for evaluating the exhalation of a subject P are described.

[0088] 2-1. Breathing evaluation method

[0089] like Figure 6 As shown, the breath evaluation method of this embodiment includes, in addition to the above-mentioned (1) photographing with an infrared camera and (2) processing of image data (steps S11 and S12), evaluation of breath (step S13). The breath evaluation method may include the above-mentioned (3) output of processed image data, or may not include it.

[0090] The evaluation of exhalation includes (4) setting of a region of interest, (5) calculation of the speed and flow rate of exhalation, and (6) evaluation. The following describes steps (4), (5), and (6) in detail.

[0091] (4) Setting of focus area

[0092] In the above 1-1(2), a region of interest (ROI) is set on the processed image data ( Figure 7 ROI refers to the area selected as the operation object in the image data, also known as "object area" or "region of interest".

[0093] The ROI is set below the nostril of the subject P. For example, Figure 8 As shown on the left side of , the ROI can be set in a linear or strip-shaped area extending downward from the nostrils of the subject P (that is, on the exhalation flow path), but is not limited thereto.

[0094] The ROI can be set manually by the user or automatically using pattern recognition technology based on the position of the nose and mouth or the flow of exhaled air.

[0095] (5) Calculation of exhalation speed and flow

[0096] Next, the flow velocity of the exhaled air of the subject P is calculated based on the temporal change of the signal intensity on the ROI. The flow velocity is calculated, for example, as follows.

[0097] That is, by extracting ROIs from a set of processed image data and arranging them in chronological order (re-slice), an image for analysis is generated ( Figure 7 Step S22). Figure 8 In the example of the analysis image shown on the right side of FIG, the distance x along the ROI is shown in the horizontal axis direction, and the time t is shown in the vertical axis direction. Therefore, the stripe pattern shown in this figure shows how the signal intensity on the ROI changes with time. The flow (flow rate) of the exhaled air can be grasped from this analysis image ( Figure 7 Step S23).

[0098] Therefore, the angle θ and the length (distance) L of the stripe-shaped band are obtained from the analysis image. The distance L can be appropriately corrected.

[0099] An example of calculation of the angle θ and the distance L is given. Fig. 9 The image on the left is Figure 8 The area B in the image is enlarged. Focusing on the stripe-like band in the image, a line Q is drawn. Then, the length L of the line and the angle θ formed with the horizontal axis are calculated.

[0100] like Fig. 9As shown on the right side of , there is a known relationship between the distance L, the angle θ, the displacement dx along the x-axis, and the displacement dt along the time axis. Using this relationship, dx=Lcosθ and dt=Lsinθ are calculated from the calculated angle θ and distance L, and the inclination of the belt (line Q), that is, the speed of exhalation v=dx / dt, is calculated from the calculated dx and dt.

[0101] More specifically, the speed v can be calculated, for example, as follows.

[0102] That is, the rules of x and L are length, and t is time.

[0103] However, in the image processing scenario, x and L are given by the distance between pixels (number of pixels), so it is necessary to use the relationship between the number of pixels and the distance (mm / pixel) to correct the number of pixels to the distance. In addition, t is time, which is determined by the frame rate. For example, at 60fps, it is 0.0167sec / pixel, which is used for correction.

[0104] A line Q is drawn using image processing software and analyzed, thereby determining the distance L and angle θ of the line Q.

[0105] Therefore, dx and dt are calculated as follows.

[0106] dx (mm) = Lcosθ (pixel) × distance correction factor (mm / pixel)

[0107] dt (sec) = Lsinθ (pixel) × time correction coefficient (sec / pixel)

[0108] As a result, the speed v is obtained by the following formula.

[0109] v(mm / sec)=dx(mm) / dt(sec)

[0110] Alternatively, L can be treated as a length instead of a pixel. In this case, dx needs to be used directly to convert dt from the number of pixels to time.

[0111] Therefore, by knowing the exhalation speed v, it is possible to determine whether the subject P is breathing. In addition, the breathing rhythm of the subject P can be known from the exhalation speed v (for example, refer to Figure 2 ).

[0112] exist Fig.10 and Fig.11 However, the present invention is not limited to the following processing examples.

[0113] exist Fig.10 In the image data shown on the left side of FIG. , from the nostril of the subject P to the exhalation (CO 2) is set on the flow path. Next, the portion corresponding to the ROI is extracted from a set of image data, and the extracted elements are arranged in a time series order, thereby generating an image for analysis (refer to Fig.10 The image for analysis may also be subjected to noise removal using a smoothing filter such as a Gaussian filter. In addition, the contrast of the image may be enhanced by using a contrast transformation such as enhanced contrast.

[0114] Then, edge detection methods such as Canny edge detection are used to extract the inclination of the stripe pattern in the image. Further, line detection methods such as ridge detection (Hough transform) are used to set the ROI based on the inclination of the stripe pattern. The approximate formula for the inclination is obtained based on the XY coordinates of the ROI (refer to Fig.10 on the right side of the image).

[0115] Or, in Fig.11 In the image data shown above, from the nostril of the subject P to the exhalation (CO 2 ) is set on the flow path. In the figure, the upstream ROI is represented as ROI_1, and the downstream ROI is represented as ROI_2. Next, the distance Δx between these ROIs is calculated. Then, the CO in each ROI is read. 2 Fluctuating charts (e.g. Fig.11 The time difference Δt of each peak of ROI_1 is shown in the graph below. In the graph, the peak of ROI_1 arrives at 3.440 seconds, while the peak of ROI_2 arrives at 3.856 seconds, so the time difference Δt=0.416 is obtained. From this, the exhaled flow rate v=Δx / Δt can be calculated.

[0116] Use the calculated exhalation velocity v to calculate the exhalation flow rate F( Figure 7 Step S24).

[0117] That is, the opening area S of the nostril of the subject P is obtained from the processed image data, and the exhalation flow rate F is calculated using the opening area S and the exhalation speed v. For example, the nostril of the subject P is approximated as a circle, and the opening area S is calculated based on the diameter D of the nostril: 2 π / 4, and the exhaled air flow rate F is calculated by F=S×V.

[0118] (6) Evaluation

[0119] Next, the exhalation velocity v and the exhalation flow rate F obtained are used to evaluate the exhalation of the subject P ( Figure 7As an example of evaluation of exhalation, in addition to the presence or absence and rhythm of breathing mentioned above, evaluation of energy metabolism can also be mentioned.

[0120] In the evaluation of energy metabolism, the energy metabolism of the subject P is estimated from the exhaled air flow rate with reference to data indicating the relationship between energy metabolism and exhaled air volume. As data indicating the relationship between energy metabolism and exhaled air volume, publicly available data can be used. For example, the data of "Human Exhaled CO2" based on Japanese Industrial Standard (JIS A 1406:1974) can be used. 2 "CO2 (for adult males)" and "CO2 (for labor intensity)" based on the Air Conditioning and Hygiene Engineering Society Standard (HASS102-1996) 2 This data can also be stored in the computer 10.

[0121] 2-2. Breathing evaluation system

[0122] A system 1 for realizing the above-mentioned breath evaluation method will be described.

[0123] like Figure 4 As shown, the system 1 includes a setting unit 15 and an evaluation unit 17 in addition to the infrared camera 20 and the processing unit 11 described above. The system 1 for implementing the breath evaluation method does not necessarily need to include the output unit 13.

[0124] Since the infrared camera 20 and the processing unit 11 have been described above, their description here will be omitted.

[0125] The setting unit 15 can execute the steps described in 2-1(4) above. The evaluation unit 17 can execute the steps described in 2-1(5) above. The setting unit 15 and the evaluation unit 17 can be executed as a function of the above-mentioned computer 10. That is, the CPU 101 reads the program and various data stored in the ROM 105 into the RAM 103 and executes them, thereby functioning as the setting unit 15 and the evaluation unit 17 (see Figure 5 ). In addition, the processing unit 11, the output unit 13, the setting unit 15 and the evaluation unit 17 may be executed by one computer or by a plurality of computers.

[0126] 3. Measurement of heart rate and body temperature

[0127] In the present embodiment, not only the breath of the subject P is measured, but also the heartbeat and body temperature of the subject P are measured. The following is an overview of the heartbeat and body temperature measurement.

[0128] 3-1. Heart rate measurement

[0129] The face of the subject P is photographed by the infrared camera 20, and a temperature image showing the temperature distribution of the face of the subject P is obtained (see Fig.12 That is, the infrared camera 20 is used as a thermal imager. Image processing is performed on the temperature image to generate a binary image, and a portion above a predetermined temperature (for example, 35.6° C.) is extracted from the generated image (refer to Fig.12 ). Fig.12 The graph shown below shows an example of the time change of the temperature in the extracted part.

[0130] By applying a filter such as a Butterworth bandpass filter (0.8-1.5 Hz) to the data representing the temporal change of the temperature in the extracted portion, the data is band-limited, and then frequency analysis (Fourier transform) is performed on the band-limited data to obtain heart rate data.

[0131] Therefore, the setting unit 15 sets a predetermined part (eg, the corner of the eye) as the ROI in the temperature image data acquired from the infrared camera 20. The evaluation unit 15 can obtain heartbeat data by extracting temperature data from the part and performing frequency analysis every time the temperature image data is acquired.

[0132] 3-2. Body temperature measurement

[0133] By photographing the face of the subject P with the infrared camera 20, the body temperature data of the subject P can also be obtained. For example, the temperature of the corner of the eye of the subject P can be measured with the infrared camera 20 and set as a representative value of the body temperature. The reason for this is that according to a certain study, the skin at the corner of the eye of a person is not easily affected by external factors such as the outside air, while the skin of other parts of the head is greatly affected by external factors. For example, if it is cold outside, the temperature of the forehead will drop significantly compared to the temperature inside, but such a large change will not occur at the corner of the eye.

[0134] Therefore, for example, the setting unit 15 sets a predetermined part (for example, the corner of the eye) in the temperature image data acquired from the infrared camera 20 as the ROI. The evaluation unit 15 only needs to extract temperature data from the part every time the temperature image data is acquired.

[0135] 4. Effects of this embodiment

[0136] In this embodiment, CO contained in human breath is used. 2Infrared camera 20 with sensitivity to absorption. In addition, image data for display is generated by subtracting past shooting data from real-time or latest shooting data. As a result, the exhalation of subject P facing infrared camera 20 can be visualized. That is, there is no need to turn the camera to the side or prepare multiple types of cameras in order to distinguish the exhalation of subject P from the background in terms of temperature as in the conventional method.

[0137] Therefore, for example, if the patient's exhalation is visualized through a monitor, the doctor can monitor the patient's breathing without touching the patient. Therefore, in addition to telemedicine, this technology can also be expected to be used in screening patients infected with infectious diseases such as COVID-19.

[0138] In addition, the breathing of not only people who are still relative to the camera, but also people who are sleeping, moving, or training can be evaluated. Therefore, the present technology can be widely used, for example, to evaluate breathing disorders during sleep (such as sleep apnea syndrome), evaluate the effect of training, determine changes in physical conditions, monitor, and monitor systems, emergency notification systems, etc.

[0139] In addition, in this embodiment, it is possible to analyze visualized exhalation and evaluate the exhaled volume. Thus, it is possible to evaluate the energy metabolism rate of the subject P in a non-contact manner without using an existing exhaled gas analyzer (e.g., a human calorimeter, a Douglas bag, or an indirect calorimetric device). Therefore, this technology can be used for daily health observation and health management of people.

[0140] Furthermore, by analyzing the specific part of the subject P photographed by the infrared camera 20, the body temperature (deep body temperature) and heartbeat of the subject P can be evaluated. That is, visualization of breathing, energy metabolism rate, deep body temperature and heartbeat can be evaluated at the same time using one infrared camera 20. Therefore, the components of the system can be reduced and the manufacturing cost can be suppressed.

[0141] In addition to human exhalation, this technology can also be used for CO 2 Fire fighting equipment, CO 2 Refrigerant air conditioning equipment, hot water supply equipment, combustion appliances (stoves, etc.), CO 2 Detection of CO in storage facilities, etc. 2 of leakage.

[0142] As mentioned above, although the typical embodiment of this invention was demonstrated, this invention is not limited to these, Various design changes are possible, and these are also included in this invention.

[0143] For example, to obtain the heart rate, a filter can be applied to the data representing the time variation of temperature, and peak detection can be performed on the processed data to obtain the peak-to-peak time. Alternatively, the heart rate (bpm) can also be obtained at 60 / (peak-to-peak time).

[0144] Because infrared CO 2 Since there are a plurality of absorption bands, visualization of exhaled breath similar to the present embodiment can be achieved by using, for example, a far-infrared camera (having a sensitivity of 8 to 14 μm).

[0145] For body temperature measurement, machine learning can also be used for facial landmark detection. For example, machine learning libraries such as dlib (http: / / dlib.net / ) provide tools that can be used to detect the corners of the eyes (such as 40 and 43). Therefore, it is possible to track the corners of the eyes while measuring the body temperature of the same part.

[0146] An evaluation of the respiratory rate can be included as one of the evaluations of exhalation. Figure 2 In the lower right graph of , a filter with a passband of 0.1-0.4 Hz is applied and the respiratory frequency is obtained by Fourier transforming the filtered data. Figure 2 The respiratory rate can be obtained by performing peak detection and peak-to-peak processing on the graph shown.

[0147] Alternatively, it is possible to detect vascular abnormalities such as carotid stenosis, arteriosclerosis, and hypertension using a temperature image of the face of the subject P captured by the infrared camera 20. That is, since the temperature image includes information on the facial artery branching from the carotid artery, it is expected that vascular abnormalities can be detected based on changes in temperature distribution near the eye socket, for example.

[0148] Specifically, a threshold is set for the temperature image of the face of the subject P. A high temperature area exceeding the threshold (e.g., below the eye socket, above the eye socket, near the nasolabial groove, etc.) is selected, and the total area, the left-right difference or left-right ratio of the area, and the left-right difference of the average temperature of the area are measured. Here, the total area can be used as an indicator of hypertension or arteriosclerosis, and the left-right difference or left-right ratio of the area and the average temperature can be used as an indicator of carotid artery stenosis. The area can also be replaced by the number of pixels belonging to the corresponding area.

[0149] For example, in the temperature image of the subject P, the total area of ​​the high temperature region around both eyes can be calculated, the areas of the left and right regions can be compared, or the average temperatures of the left and right regions can be compared. This processing can be performed by the above-mentioned free software ImageJ, for example.

[0150] In addition, there are several papers that suggest the relationship between facial temperature and carotid artery stenosis. It is pointed out that due to carotid artery stenosis, the blood flow of the artery changes, and thus the temperature distribution on the path of the artery running through the face may change. Therefore, when no left-right difference is substantially found, the subject P may be judged to be healthy, or when the left-right balance is found to be broken, it may be judged that the subject P may have unilateral stenosis.

[0151] This abnormality detection method can be incorporated into the system 1 , whereby the system 1 can evaluate various health conditions of the subject P.

[0152] Explanation of symbols

[0153] 1 System

[0154] 10 Computer

[0155] 11 Processing Department

[0156] 13 Output section

[0157] 15. Setting section

[0158] 17 Evaluation Department

[0159] 19 Storage

[0160] 20 Infrared Camera

Claims

1. An exhalation visualization system, characterized in that: have: an infrared camera that detects light in an infrared region radiated from the face of the subject and acquires image data; a processing unit that subtracts second image data acquired at a second time before the first time from first image data acquired at a first time to generate processed image data; as well as An output unit outputs the processed image data.

2. The exhalation visualization system according to claim 1, characterized in that: The infrared camera detects light in the mid-infrared region.

3. The exhalation visualization system according to claim 1 or 2, characterized in that: The processing unit uses an average of image data acquired over a predetermined period before the first time as the second image data.

4. A method for visualizing exhalation, characterized in that: include: A step of detecting light in the infrared region emitted from the face of the subject and acquiring image data; Subtracting second image data acquired at a second time before the first time from first image data acquired at a first time to generate processed image data; and The step of outputting the processed image data.

5. A breath evaluation system, characterized in that: have: an infrared camera that detects light in an infrared region radiated from the face of the subject and acquires image data; a processing unit that subtracts second image data acquired at a second time before the first time from first image data acquired at a first time to generate processed image data; a setting unit that sets a region of interest below the nostrils of the subject in the processed image data; as well as The evaluation unit calculates a flow rate of exhaled air of the subject based on a temporal change in the signal intensity in the region of interest.

6. The breath evaluation system according to claim 5, characterized in that: The evaluation unit calculates an opening area of ​​the nostril based on a diameter of the nostril, and calculates a flow rate of the exhaled air using the flow velocity and the opening area.

7. The breath evaluation system according to claim 6, characterized in that: The evaluation unit evaluates the energy metabolism of the subject based on the exhaled air flow rate with reference to data indicating a relationship between energy metabolism and exhaled air volume.

8. A breath evaluation method, characterized in that: include: A step of detecting light in the infrared region emitted from the face of the subject and acquiring image data; A step of subtracting second image data acquired at a second time earlier than the first time from first image data acquired at a first time to generate processed image data; A step of setting a region of interest below the nostrils of the subject in the processed image data; and The step of calculating the flow rate of the exhaled air of the subject based on the temporal change of the signal intensity in the region of interest.

Images