Biological information detection system

Abstract

To provide a biological information detection system capable of achieving improvement in detection accuracy of biological information using image processing.SOLUTION: In a biological information detection system 1, biological information detection processing includes: first processing S106 for dividing an entire region A1 of an image taken by imaging means 40 into a plurality of divided regions B1; second processing S107 for calculating an SN ratio based on image information that varies with time in each divided region B1; third processing S108 for comparing each SN ratio of each divided region B1 with a predetermined value, extracting a divided region B1 in which the SN ratio is the predetermined value or more, and setting it as at least one first skin region C1; fourth processing S111 for setting a hue range for detection on the basis of hue information extracted from the first skin region C1; and fifth processing S112 and S113 for setting a region included in the hue range for detection of the entire region A1 as a second skin region C2, and detecting biological information from the second skin region C2.SELECTED DRAWING: Figure 1

Description

Technical Field

[0001] The present invention relates to a biological information detection system.

Background Art

[0002] Patent Document 1 discloses an example of a conventional biological information detection system. This biological information detection system includes a camera that captures an image of a subject and a control unit that controls the camera.

[0003] Based on the image captured by the camera, the control unit detects a pulse wave, which is the biological information of the subject. Specifically, the control unit has a first color space conversion unit, a second color space conversion unit, an independent component analysis unit, a Fourier transform unit, and a signal selection and averaging unit. By using these units, the control unit extracts the hue of human skin from the image data of the subject's face captured by the camera, and detects a pulse wave in the detection area where the hue is extracted.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, in the above conventional biological information detection system, it is necessary to consider individual differences in the hue of human skin, and the range of the hue to be extracted has to be increased. However, in this biological information detection system, if the range of the hue to be extracted is increased, there is a risk of misrecognizing clothes, background, etc. reflected in the camera as human skin and using them as the detection area of the pulse wave. As a result, it is difficult to improve the detection accuracy of the pulse wave.

[0006] In view of the above conventional situation, the present invention aims to solve the problem of providing a biological information detection system capable of improving the detection accuracy of biological information using image processing.

Means for Solving the Problem

[0007] The biological information detection system of the present invention includes imaging means for imaging an image of a subject, a control unit that controls the imaging means and executes a biological information detection process for detecting biological information including at least the vital state of the subject based on the image and digitized biological information, and is provided with The biological information detection process includes a first process of dividing the entire area of the image into a plurality of divided areas, a second process of calculating an SN ratio based on image information that varies over time in each of the divided areas, a third process of comparing each of the SN ratios of the divided areas with a predetermined value, extracting the divided areas where the SN ratio is equal to or greater than the predetermined value, and setting them as at least one first muscle area, a fourth process of setting a detection hue range based on the hue information extracted from the first muscle area, a fifth process of setting, as a second muscle area, an area included in the detection hue range in the entire area, and detecting the biological information from the second muscle area, and is characterized by having

[0008] In the biological information detection system of the present invention, the control unit executes a biological information detection process. In the first process, the second process, and the third process, each of the SN ratios calculated in a plurality of divided areas obtained by dividing the entire area of the image captured by the imaging means is compared with a predetermined value, and the divided areas where the SN ratio is equal to or greater than the predetermined value are extracted and set as at least one first muscle area.

[0009] Here, the SN ratio becomes a high value only in the human muscle area and a low value in areas different from the muscle area, such as clothes and background. Therefore, the control unit can estimate that the area with a high SN ratio is highly likely to be the human muscle area, and can suppress the misrecognition of the areas with a low SN ratio as the human muscle, such as clothes and background. Thus, the first muscle area can be set with high accuracy.

[0010] Then, in the fourth process, the control unit can accurately set the detection hue range suitable for each individual by setting the detection hue range based on the hue information extracted from the first skin area. As a result, since the necessity of setting a large detection hue range in consideration of individual differences decreases, the control unit can accurately set the second skin area suitable for detecting biological information from the entire area of the image captured by the imaging means in the fifth process.

[0011] Therefore, the biological information detection system of the present invention can realize an improvement in the detection accuracy of biological information using image processing.

[0012] Also, in this biological information detection system, the control unit executes the calculation of the signal-to-noise ratio (SN ratio), which has a large computational load due to frequency analysis, in the first half of the biological information detection process, and detects the biological information without calculating the SN ratio in the second half of the biological information detection process. Thereby, this biological information detection system can reduce the computational load compared to a configuration that always repeats the calculation of the SN ratio and the detection of biological information.

[0013] Furthermore, in this biological information detection system, the control unit sets the area included in the detection hue range as the second skin area and detects biological information from the second skin area. Thereby, this biological information detection system can set the second skin area following the movement of the subject even when the relative positional relationship between the subject and the imaging means changes due to the movement of the subject, so that biological information can be detected while suitably corresponding to the movement of the subject.

[0014] The imaging means preferably captures an image of a subject who has entered a predetermined room. The biological information detection system of the present invention preferably includes an entrance detection means for detecting whether or not the subject has entered the room. Then, each time the entrance detection means detects that the subject has entered the room, the control unit preferably starts the biological information detection process, executes the first to fifth processes, and then repeatedly executes the fifth process.

[0015] Image information related to the skin hue of a person included in an image captured by the imaging means may vary due to the influence of lighting in a predetermined room or external light entering the room, which varies greatly depending on time zones such as morning, noon, and night, even for the same individual. In this regard, with the above configuration, each time the subject enters the room, the control unit can set the detection hue range according to the lighting in the room or the situation of the external light entering the room. As a result, this biometric information detection system can set the detection hue range more accurately, thus further improving the detection accuracy of biometric information using image processing.

[0016] After the control unit executes the first to fourth processes, while the second skin area can be set in the fifth process, the fifth process is repeatedly executed. When the second skin area cannot be set in the fifth process, it is desirable to stop repeating the fifth process and execute from the first process.

[0017] When the second skin area cannot be set in the fifth process, it is considered that the subject has moved outside the imaging range of the imaging means (frame out). In this case, it is highly likely that the subject to be imaged by the imaging means later is not the same as the subject imaged earlier. In this regard, with the above configuration, the detection hue range can be set each time the second skin area cannot be set in the fifth process, so it can suitably respond to the replacement of the subject.

[0018] The imaging means preferably captures an image of a subject who has entered a predetermined room. The room is preferably a bathroom. And the subject is preferably a bather.

[0019] With this configuration, the imaging means captures an image of a bather who has entered the bathroom after taking off their clothes. Therefore, in the third process, the control unit can extract many divided areas with an SN ratio of a predetermined value or more and set them as the first skin area, so the amount of information for setting the detection hue range increases. As a result, this biometric information detection system can set the detection hue range more accurately, thus further improving the detection accuracy of biometric information using image processing.

Advantages of the Invention

[0020] According to the biological information detection system of the present invention, it is possible to improve the detection accuracy of biological information using image processing.

Brief Description of the Drawings

[0021]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Modes for Carrying Out the Invention

[0022] Hereinafter, embodiments embodying the present invention will be described with reference to the drawings.

[0023] (Embodiment) As shown in FIG. 1, the biological information detection system 1 of the embodiment is an example of a specific aspect of the biological information detection system of the present invention and is applied to a house H1. The house H1 is provided with a bathroom R1 and a kitchen R2. The bathroom R1 is an example of the "predetermined room" of the present invention. In the bathroom R1, a bathtub 3, a mixing faucet 5, a shower 5A, and lighting 4 are installed.

[0024] The biological information detection system 1 includes a hot water supply device 90, a bathroom remote control 10, a kitchen remote control 20, and a door switch 6S. The door switch 6S is an example of the "room entry detection means" of the present invention. Further, the biological information detection system 1 includes a camera 40 provided integrally with the bathroom remote control 10. The camera 40 is an example of the "imaging means" of the present invention.

[0025] <Hot water supply device> The hot water supply device 90 is installed in the house H1. Since the hot water supply device 90 has a well-known configuration, the description will be simplified. It has a gas burner, a heat exchanger, a circulation pump, etc. (not shown). The gas burner burns fuel gas such as city gas to generate high-temperature combustion gas. The hot water supply device 90 circulates the water supplied from a water supply or the like in the heat exchanger and performs heat exchange with the combustion gas generated by the gas burner to heat the water and supply hot water.

[0026] The hot water supply device 90 supplies hot water to the mixing faucet 5 and the shower 5A via the pipe P5. Also, the hot water supply device 90 supplies hot water to the bathtub 3 via the bathtub supply pipe P3A. The hot water supply device 90 can execute an automatic filling operation for automatically filling the bathtub 3, a supplementary hot water operation for adding hot water to the bathtub 3, and a supplementary water operation for adding water to the bathtub 3.

[0027] Furthermore, the hot water supply device 90 can execute a reheating operation in which hot water is circulated among the bathtub 3, the bathtub return pipe P3B, the heat exchanger, and the bathtub supply pipe P3A by a circulation pump (not shown) and heated by the gas burner.

[0028] The hot water supply device 90 has a control unit 91. The control unit 91 is an electronic circuit unit including a CPU (not shown), a storage unit 91M composed of storage elements such as a ROM and a RAM, an interface circuit, etc. The control unit 91 executes control processing related to the operations of the gas burner, heat exchanger, circulation pump, etc. of the hot water supply device 90.

[0029] The storage unit 91M stores various programs and setting information for operating the hot water supply device 90 and the like. Also, the storage unit 91M appropriately stores various information acquired by the control unit 91 during the operation of the hot water supply device 90 and the like.

[0030] The hot water supply device 90 has a water level sensor 93. The water level sensor 93 includes a pressure sensor that measures the pressure in the internal piping connected to the bathtub 3 via the bathtub supply pipe P3A and the bathtub return pipe P3B. The water level sensor 93 detects the water level of the bathtub 3 based on the hydrostatic pressure measured by the pressure sensor when the water surface in the bathtub 3 is in a static state. When the bather P1 is in the bathtub 3, the water level is significantly higher than when the bather P1 is not in the bathtub 3. The water level sensor 93 detects whether the bather P1 is in the bathtub 3 based on such a change in the water level.

[0031] <Bathroom remote control> The bathroom remote control 10 is installed on the inner wall of the bathroom R1 that faces the bather P1 in the bathtub 3. That is, the bathroom remote control 10 is installed at a position where the integrally provided camera 40 can photograph the bather P1 in the bathtub 3 and the surrounding area. The bathroom remote control 10 is connected to the control unit 91 of the hot water supply device 90 so as to be capable of wired communication.

[0032] As shown in FIG. 2, the bathroom remote control 10 has an input unit 12B, a display unit 12D, a speaker 12S, a wireless communication unit 13, and a control unit 11.

[0033] The bathroom remote control 10 transmits the operation performed by the bather P1 on the input unit 12B to the hot water supply device 90 to remotely control the hot water supply device 90. Further, the bathroom remote control 10 displays various information such as the operation status and setting information transmitted from the hot water supply device 90 on the display unit 12D or outputs it as sound by the speaker 12S.

[0034] As shown in FIG. 1, the wireless communication unit 13 performs wireless communication with the wireless router 7 installed in the house H1 by Wi-Fi (registered trademark) or the like. The control unit 11 can connect to the external network NW1 via the wireless communication unit 13 and the wireless router 7 and perform network communication with an information processing terminal such as an external server 9 connected to the network NW1. The external server 9 is a support server of the biological information detection system 1.

[0035] As shown in FIG. 2, the control unit 11 is an electronic circuit unit including a CPU (not shown), a storage unit 11M composed of storage elements such as a ROM and a RAM, and an interface circuit or the like.

[0036] The control unit 11 controls the operations of the input unit 12B, the display unit 12D, and the speaker 12S of the bathroom remote controller 10, and also controls the operation of the camera 40. Further, the control unit 11 controls the wired communication between the bathroom remote controller 10 and the control unit 91 of the hot water supply device 90.

[0037] The storage unit 11M stores various programs and setting information for operating the bathroom remote controller 10 and the camera 40. Further, the storage unit 11M appropriately stores various information acquired by the control unit 11 during the operations of the bathroom remote controller 10 and the camera 40. The programs stored in the storage unit 11M include the biometric information detection processing program shown in FIG. 3.

[0038] As shown in FIG. 2, the control unit 11 has an image processing unit 11A and a biometric information detection unit 11B. The image processing unit 11A and the biometric information detection unit 11B function when the control unit 11 executes the biometric information detection processing program shown in FIG. 3. The processing contents of the image processing unit 11A and the biometric information detection unit 11B will be described later.

[0039] <Kitchen remote controller> As shown in FIG. 1, the kitchen remote controller 20 is installed in the kitchen R2 of the house H1. The kitchen remote controller 20 is connected to the control unit 91 of the hot water supply device 90 so as to be capable of wired communication.

[0040] Since the kitchen remote controller 20 has a well-known configuration, the description will be simplified. Similar to the bathroom remote controller 10, the kitchen remote controller 20 remotely operates the hot water supply device 90 in response to the operation of the occupant in the kitchen R2, and displays or outputs various information such as the operation status and setting information transmitted from the hot water supply device 90 by voice.

[0041] The kitchen remote controller 20 has a wireless communication unit (not shown). The wireless communication unit of the kitchen remote controller 20 performs wireless communication with the wireless router 7 installed in the house H1 via Wi-Fi (registered trademark) or the like. The control unit 91 of the water heater 90 is connected to the external network NW1 via the wireless communication unit of the kitchen remote controller 20 and the wireless router 7, and can perform network communication with an information processing terminal such as an external server 9 connected to the network NW1.

[0042] <Door switch> The door switch 6S is a well-known sensor that detects the opening and closing of the door 6 of the bathroom R1. The signal wiring of the door switch 6S is connected to the control unit 11 and the control unit 91.

[0043] The door switch 6S detects whether the bather P1 has entered the bathroom R1 based on the opening and closing of the door 6, and transmits the detection result to the control unit 11 and the control unit 91. The control unit 11 and the control unit 91 can consider that the bather P1 has entered the bathroom R1 when the door switch 6S detects the opening of the door 6.

[0044] <Camera> The camera 40 is a digital camera capable of shooting videos. The images captured by the camera 40 are composed of RGB format image information. The frame rate when the camera 40 shoots a video is, for example, 10 to 60 (frames / second).

[0045] When the control unit 11 executes the biometric information detection processing program shown in FIG. 3, the camera 40 is controlled by the control unit 11 to perform a shooting operation of shooting an image of the bather P1 who has entered the bathroom R1 as shown in FIG. 1. The bather P1 is an example of the "subject" of the present invention.

[0046] The camera 40 performs a shooting operation so that at least the head P1H of the bather P1 is included in the captured image. The image information of the image captured by the camera 40 is transmitted to the image processing unit 11A of the control unit 11.

[0047] <Biometric information detection processing> Based on the image captured by the camera 40, more specifically, based on the physical state of the bather P1 included in the image captured by the camera 40, the control unit 11 executes a biological information detection process for detecting biological information including at least the vital state of the bather P1, which is digitized biological information.

[0048] The control unit 11 stores the detected biological information in the storage unit 11M for use as data for the health management of the bather P1, and also uses it to observe changes in the physical condition of the bather P1 and to determine the presence or absence of safety problems of the bather P1.

[0049] The physical state of the bather P1 included in the image captured by the camera 40 includes at least a part of the image of the body of the bather P1 included in the captured image of the camera 40 and the sequential changes in the hue, brightness, and saturation of a plurality of pixels constituting the image.

[0050] The biological information including at least the vital state of the bather P1, which is digitized biological information, is a pulse wave, heart rate, respiration, etc. A pulse wave is a wave of the propagation of the change in arterial pressure (pulse pressure) accompanying the heartbeat of the heart, and distortion occurs in the waveform while being transmitted to the periphery.

[0051] As shown in FIG. 2, in this embodiment, the image processing unit 11A processes the image captured by the camera 40, and the biological information detection unit 11B detects the pulse wave and respiration as digitized information based on the image processing result.

[0052] For example, the pulse wave is detected by the sequential change in the luminance of the face part. The respiration rate, the degree of the depth of respiration, etc. are detected by digitizing the fluctuations of the nostrils, mouth, chest, etc. and performing arithmetic processing. Then, the biological information detection unit 11B performs arithmetic processing on the information extracted from the detected pulse wave and detects the heart rate, electrocardiogram waveform, heart rate variability analysis index (LF / HF, etc.) as digitized information.

[0053] In order to improve the detection accuracy of the biometric information of the bather P1, when the power supplies of the hot water supply device 90 and the bathroom remote controller 10 are turned on, the control unit 11 executes the biometric information detection processing program shown in FIG. 3.

[0054] First, in step S101, based on the detection result of the door switch 6S, the control unit 11 detects whether the bather P1 has entered the bathroom R1.

[0055] If the result in step S101 is "No", the control unit 11 repeats step S101. When the result in step S101 becomes "Yes", the control unit 11 proceeds to step S102.

[0056] When the control unit 11 proceeds to step S102, it causes the camera 40 to start a shooting operation. The camera 40 performs video shooting at a predetermined frame rate.

[0057] Next, the control unit 11 proceeds to step S103 and acquires a new frame (image) by the camera 40. An example of the new frame (image) acquired in step S103 is shown in FIG. 4. FIG. 4 shows a state where the bather P1 in the bathtub 3 is immersed in the hot water up to the shoulders. The entire area A1 of the image is an area surrounded by a large rectangular frame indicated by a thick line.

[0058] Next, the control unit 11 proceeds to step S104 shown in FIG. 3 and determines whether the detection hue range is not set. The detection hue range is set in step S111. Therefore, the description of the detection hue range will be given in step S111.

[0059] If the result in step S104 is "No", the control unit 11 proceeds to step S112. The processing after step S112 will be described later. On the other hand, if the result in step S104 is "Yes", the control unit 11 proceeds to step S106.

[0060] When the control unit 11 proceeds to step S106, as shown in FIG. 4, the entire region A1 of the image of the new frame acquired in step S103 is divided into a plurality of divided regions B1. Step S106 is an example of the "first process" of the present invention.

[0061] Each divided region B1 is a region surrounded by a plurality of thin lines arranged in a grid pattern. In the example shown in FIG. 4, the entire region A1 of the image is divided into 168 divided regions B1 arranged in 12 columns vertically and 14 columns horizontally.

[0062] Next, the control unit 11 proceeds to step S107 shown in FIG. 3, and calculates the signal-to-noise ratio (SN ratio) based on the image information that changes over time in each divided region B1 divided in step S106. Step S107 is an example of the "second process" of the present invention.

[0063] A specific example of a method for calculating the SN ratio based on the image information that changes over time in each divided region B1 will be described. When the temporal change in the luminance of the plurality of pixels constituting one divided region B1 is subjected to FFT conversion, if a part of the body (especially the skin) of the bather P1 is included in that divided region B1, a signal corresponding to the pulse wave of the bather P1 is detected. If a part of the body of the bather P1 is not included in that divided region B1, a signal corresponding to the pulse wave of the bather P1 is not detected. Then, for that divided region B1, the sum of the pulse components (power spectrum) and the sum of all components are calculated. The SN ratio = (pulse component / total component). The larger the proportion of the body part of the bather P1 included in that divided region B1, the larger the SN ratio. When the pulse can be detected without noise, the SN ratio = 1 (the higher the better). This process is executed for all the divided regions B1.

[0064] Next, the control unit 11 proceeds to step S108, compares each of the SN ratios of the divided regions B1 calculated in step S107 with a predetermined value (for example, 0.8), and extracts the divided regions B1 whose SN ratios are equal to or greater than the predetermined value and sets them as at least one first skin region C1. Step S108 is an example of the "third process" of the present invention.

[0065] The predetermined value is a threshold value set so that there is a high possibility that a part of the body (especially the skin) of the bather P1 is included in the divided area B1. An example of the first skin area C1 set in step S108 is shown in FIG. 5. The first skin area C1 is the divided area B1 surrounded by the thick broken line BL1.

[0066] Next, the control unit 11 proceeds to step S109 shown in FIG. 3 and determines whether the first skin area C1 exists.

[0067] If the result in step S109 is "No", the control unit 11 determines that the biological information of the bather P1 cannot be detected. Then, the control unit 11 returns to step S103 and repeats steps S103 to S109. On the other hand, if the result in step S109 is "Yes", the control unit 11 proceeds to step S111.

[0068] When the control unit 11 proceeds to step S111, it sets a detection color range based on the hue information extracted from the first skin area C1 set in step S108. Then, the control unit 11 proceeds to step S112. Step S111 is an example of the "fourth process" of the present invention.

[0069] Here, the hue information extracted from the first skin area C1 is information that expresses the hues of a plurality of pixels constituting the first skin area C1 in a well-known color system.

[0070] In this embodiment, since the discrimination process becomes complicated when using the RGB color system as the color system, the HSV color space or the YCbCr color space in which colors can be easily defined by variables is used. The HSV color space represents the color space with three independent elements: hue, saturation, and value. The YCbCr color space represents the color space using a luminance signal Y and two color difference signals, so it is easy to reduce the processing load. Note that, as the color system, the Munsell color system, the L*a*b* color system, etc. can also be used.

[0071] A specific example of a method for setting a detection hue range will be given. For example, the average value of the hue information extracted from all the first skin regions C1 is calculated, and based on this average value, "[average value - α to average value + β]" or "[average value × (1 - α) to average value × (1 + β)]" may be set as the detection hue range. Also, based on the minimum value and the maximum value of the hue information extracted from all the first skin regions C1, "[minimum value to maximum value]" or "[minimum value + α to maximum value - β]" may be set as the detection hue range.

[0072] When shifting from step S104 or step S111 to step S112, the control unit 11 sets, as the second skin region C2, the region included in the detection hue range set in step S111 within the entire region A1. Thereby, portions that were not set in the first skin region C1 (skin portions included in the divided region B1 with a small skin ratio) can also be set as the second skin region C2, and it is also possible to follow body movements within the entire region A1.

[0073] An example of the second skin region C2 set in step S112 is shown in FIG. 6. The second skin region C2 is the hatched region. Around the mouth, eyes, nostrils, and eyebrows of the bather P1, the possibility of being set in the second skin region C2 is lower compared to the cheeks, forehead, jaw, neck, etc. of the bather P1.

[0074] Next, the control unit 11 shifts to step S113 shown in FIG. 3 and detects biological information from the second skin region C2 set in step S112. The content of the detection process of the biological information is as described above.

[0075] Steps S112 and S113 are an example of the "fifth process" of the present invention.

[0076] Next, the control unit 11 proceeds to step S114, stores the biological information detected in step S113 in the storage unit 11M, and uses it to observe changes in the physical condition of the bather P1 or to determine whether there are any safety problems with the bather P1. Then, if necessary regarding changes in the physical condition or safety problems of the bather P1, the control unit 11 notifies the bather P1 via the display unit 12D or the speaker 12S. After that, the control unit 11 proceeds to step S116.

[0077] When the control unit 11 proceeds to step S116, it determines whether the bather P1 has exited as detected by the camera 40 and the door switch 6S.

[0078] If the bather P1 does not frame out of the shooting range of the camera 40, or if the door switch 6S does not detect the opening of the door 6 even when the bather P1 frames out of the shooting range of the camera 40, it becomes "No" in step S116. The control unit 11 determines that the bather P1 has not exited the bathroom R1, returns to step S103, and repeats steps S103 to S116.

[0079] On the other hand, when the bather P1 frames out of the shooting range of the camera 40 and the door switch 6S detects the opening of the door 6, it becomes "Yes" in step S116. The control unit 11 determines that the bather P1 has exited the bathroom R1 and proceeds to step S117.

[0080] When the control unit 11 proceeds to step S117, after ending the shooting operation of the camera 40, it ends this program. Then, until the power of the water supply device 90 and the bathroom remote control 10 is turned off, the control unit 11 repeats the biological information detection processing program shown in FIG. 3.

[0081] That is, every time the door switch 6S detects that the bather P1 has entered the bathroom R1, the control unit 11 starts the biological information detection processing, executes from the first process (step S106) to the fifth process (steps S112, S113), and then repeatedly executes the fifth process (steps S112, S113).

[0082] <Effect> In the biological information detection system 1 of the embodiment, as shown in FIG. 3, the control unit 11 executes biological information detection processing. In the first process (step S106), the second process (step S107), and the third process (step S108), each of the SN ratios calculated in a plurality of divided regions B1 obtained by dividing the entire region A1 of the image captured by the camera 40 is compared with a predetermined value. As shown in FIG. 4, the divided regions B1 having an SN ratio equal to or higher than the predetermined value are extracted and set as at least one first skin region C1.

[0083] Here, the SN ratio becomes a high value only in the human skin region, and becomes a low value in a region different from the skin region, for example, clothes, background, etc. Therefore, the control unit 11 can estimate that the region with a high SN ratio is highly likely to be a human skin region, and can suppress the misrecognition of the region with a low SN ratio as a human skin by estimating that it is likely to be clothes, background, etc. Therefore, the first skin region C1 can be set with high accuracy.

[0084] Then, as shown in FIG. 3, in the fourth process (step S111), the control unit 11 can accurately set a detection hue range according to each individual by setting a detection hue range based on the hue information extracted from the first skin region C1. As a result, since the necessity of setting a large detection hue range in consideration of individual differences is reduced, in the fifth process (steps S112 and S113), as shown in FIG. 6, the control unit 11 can accurately set a second skin region C2 suitable for detecting biological information from the entire region A1 of the image captured by the camera 40.

[0085] Therefore, the biological information detection system 1 of the embodiment can realize an improvement in the detection accuracy of biological information using image processing.

[0086] Also, in this biological information detection system 1, the control unit 11 calculates the signal-to-noise ratio (SNR), which involves frequency analysis and has a large computational load, in the first half of the biological information detection process. In the second half of the biological information detection process, the control unit 11 detects the biological information without calculating the SNR. As a result, this biological information detection system 1 can reduce the computational load compared to a configuration that always repeats the calculation of the SNR and the detection of biological information.

[0087] Furthermore, in this biological information detection system 1, the control unit 11 sets the region included in the detection hue range as the second skin region C2, and detects biological information from the second skin region C2. As a result, even when the bather P1 moves and the relative positional relationship with the camera 40 changes, this biological information detection system 1 can set the second skin region C2 following the movement of the bather P1, so that it can detect biological information while suitably corresponding to the movement of the bather P1.

[0088] Also, in this biological information detection system 1, every time the door switch 6S detects that the bather P1 has entered the bathroom R1, the control unit 11 starts the biological information detection process, executes from the first process (step S106) to the fifth process (steps S112, S113), and then repeatedly executes the fifth process (steps S112, S113). The image information regarding the hue of the skin of a person included in the image captured by the camera 40 may vary due to the influence of the lighting 4 in the bathroom R1 and the external light entering the bathroom R1 changing greatly depending on time zones such as morning, noon, and night, even for the same individual. In this regard, with the above configuration, the control unit 11 can set the detection hue range according to the lighting 4 in the bathroom R1 and the situation of the external light entering the bathroom R1 every time the bather P1 enters the bathroom R1. As a result, this biological information detection system 1 can set the detection hue range more accurately, so that it can further improve the detection accuracy of biological information using image processing.

[0089] Furthermore, in this biological information detection system 1, the camera 40 captures an image of the bather P1 who has entered the bathroom R1 after taking off their clothes. With this configuration, in the third process (step S108), the control unit 11 can extract many divided regions B1 with an SN ratio equal to or greater than a predetermined value and set them as the first skin region C1, so that the amount of information for setting the detection hue range increases. As a result, this biological information detection system 1 can set the detection hue range more accurately, and thus can further improve the detection accuracy of biological information using image processing.

[0090] As described above, the present invention has been described with reference to the embodiments. However, the present invention is not limited to the above embodiments, and it goes without saying that it can be appropriately modified and applied without departing from the gist thereof.

[0091] (Modification example) The biological information detection system of the modification example modifies the biological information detection processing program shown in FIG. 3 for the biological information detection system 1 of the embodiment as follows. That is, between step S112 and step S113, a step for determining "whether the second skin region could be set in step S112" is added. If "Yes", the process proceeds to step S113. If "No", after changing the detection hue range to unset, the process returns to step S103. Other configurations of the modification example are the same as those of the embodiment.

[0092] In such a biological information detection system of the modification example, after the control unit 11 executes the first process (step S106) to the fourth process (step S111), while the second skin region C2 can be set in the fifth process (steps S112, S113), the fifth process (steps S112, S113) is repeatedly executed. When the second skin region C2 cannot be set in the fifth process (steps S112, S113), the repetition of the fifth process (steps S112, S113) is stopped and the process is executed from the first process (step S106).

[0093] If it becomes impossible to set the second skin area C2 in the fifth process (steps S112 and S113), it is considered that the bather P1 has moved outside the shooting range of the camera 40 (has gone out of the frame). In this case, it is highly likely that the bather P1 captured by the camera 40 later is not the same as the bather P1 captured earlier. In this regard, with the above configuration, since the detection hue range can be set every time it becomes impossible to set the second skin area C2 in the fifth process (steps S112 and S113), it is possible to suitably respond to the replacement of the bather P1.

[0094] In the embodiment, the predetermined room is the bathroom R1, but the present invention is not limited to this configuration. For example, the predetermined room may be a room other than the bathroom in a house or a facility. Further, a configuration in which an image of a target person in a semi-outdoor space, an outdoor space, etc. is captured by a photographing means and the control unit executes a biological information detection process for detecting the biological information of the target person is also included in the present invention.

[0095] In the embodiment, the room entry detection means is the door switch 6S, but the present invention is not limited to this configuration. For example, as the room entry detection means, a human sensor provided in the bathroom R1 separately from the camera 40 may be used. Further, the water level sensor 93 may be used as the room entry detection means. In this case, when the water level sensor 93 detects that the bather P1 is in the bathtub 3, it can be considered that the bather P1 has entered the bathroom R1. Furthermore, the camera 40 may be used as the room entry detection means. In this case, the camera 40 can detect whether or not the bather P1 has entered the bathroom R1 based on a change in the pixel information of the captured image due to the entry of some object into the shooting range. At this time, the image processing unit 11A cooperates with the camera 40, analyzes the captured image of the camera 40, and determines whether or not the bather P1 is in a undressed state, so that it is possible to accurately detect whether or not the bather P1 has entered the bathroom R1. Note that when using a photographing means as the room entry detection means for detecting whether or not a target person has entered a predetermined room, even if the shooting range of the photographing means corresponds only to a specific location in the predetermined room, it can be considered that the target person has entered the predetermined room due to a frame-in to the shooting range.

[0096] In the embodiment, the camera 40 is integrally provided in the bathroom remote controller 10, and the control unit 11 controls the camera 40 to execute the biological information detection process. However, the present invention is not limited to this configuration. For example, a configuration in which the camera 40 is provided separately from the bathroom remote controller 10 and the control unit 91 controls the camera 40 to execute the biological information detection process is also included in the present invention. Further, a configuration in which the camera 40 is provided separately from the bathroom remote controller 10, has a control unit similar to the control unit 11, and the control unit cooperates with the control unit 91 to execute at least a part of the biological information detection process is also included in the present invention.

[0097] The configuration of eliminating the wireless communication unit 13 according to the embodiment from the bathroom remote controller 10 and connecting the control unit 11 to the external network NW1 via the control unit 91 of the hot water supply device 90, the wireless communication unit of the kitchen remote controller 20, and the wireless router 7 to execute network communication with the external server 9 or the like is also included in the present invention.

[0098] In the embodiment, the control unit 11 executes the biological information detection process program shown in FIG. 3 when the power supplies of the hot water supply device 90 and the bathroom remote controller 10 are turned on. However, the present invention is not limited to this configuration. For example, the control unit 11 does not execute the biological information detection process program shown in FIG. 3 when the power supplies of the hot water supply device 90 and the bathroom remote controller 10 are turned on, and may execute the biological information detection process program when the resident or bather P1 of the house H1 performs an operation instructing the start of the biological information detection process program on the input unit 12B of the bathroom remote controller 10.

Industrial Applicability

[0099] The present invention can be used, for example, in houses and facilities where bathrooms are installed, facilities where bathrooms are not installed, semi-outdoor spaces and outdoor spaces used by people, and the like.

Explanation of Signs

[0100] 1... Biological information detection system P1... Subject (bather) 40... Photographing means (camera) 11... Control unit A1... Entire area of the image B1…Partition area S106…First process S107…Second process C1…First muscle area S108…Third process S111…Fourth process C2…Second muscle area S112, S113…Fifth process R1…Predetermined room (bathroom) 6S…Room entry detection means (door switch)

Claims

1. Imaging means for imaging an image of a subject, A control unit that controls the imaging means and executes a biological information detection process for detecting the biological information including at least the vital state of the subject based on the image and digitized biological information, Comprising, The biological information detection process includes a first process of dividing the entire region of the image into a plurality of divided regions, A second process of calculating a signal-to-noise ratio based on image information that varies over time in each of the divided regions, A third process of comparing each of the signal-to-noise ratios of the divided regions with a predetermined value, and extracting the divided regions where the signal-to-noise ratio is equal to or greater than the predetermined value and setting them as at least one first skin region, A fourth process of setting a detection hue range based on the hue information extracted from the first skin region, A fifth process of setting, as a second skin region, a region included in the detection hue range in the entire region, and detecting the biological information from the second skin region, A biological information detection system characterized by having the above.

2. The imaging means images the image of the subject who has entered a predetermined room, It is provided with entry detection means for detecting whether or not the subject has entered the room, The control unit starts the biological information detection process every time the entry detection means detects that the subject has entered the room, executes from the first process to the fifth process, and then repeatedly executes the fifth process. The biological information detection system according to Claim 1.

3. After executing the first process to the fourth process, the control unit repeatedly executes the fifth process while the second skin region can be set in the fifth process, and when the second skin region cannot be set in the fifth process, the control unit stops repeating the fifth process and executes from the first process. The biological information detection system according to Claim 1.

4. The imaging means images the image of the subject who has entered a predetermined room, The room is a bathroom, The subject is a bather. The biological information detection system according to any one of Claims 1 to 3.

Images