Airflow control system for air shower, and airflow control method for air shower

Abstract

An airflow control system (10) is provided with: an acquisition unit (34) for acquiring image data about an image in which a person appears; an identification unit (35) for identifying a feature pertaining to the external appearance of the person on the basis of the acquired image data; an estimation unit (36) for estimating the amount of dust generated from the person on the basis of the feature pertaining to the external appearance of the person; and a control unit (37) for controlling, on the basis of the estimated amount of dust generated, an airflow generation device (40) that generates an airflow for reducing the dust.

Description

Technical Field

[0001] The present invention relates to an airflow control system and an airflow control method.

Background Art

[0002] Various techniques related to ventilation or air conditioning of indoor spaces have been proposed. Patent Document 1 discloses an air conditioner that performs blowing control by automatically selecting a part to which air is blown according to the elapsed time without the user changing the operation mode.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] The present invention provides an airflow control system and an airflow control method capable of controlling airflow based on an estimated result of the amount of dust generated.

Means for Solving the Problems

[0005] An airflow control system according to an aspect of the present invention includes an acquisition unit that acquires image data of an image in which a person appears, a specification unit that specifies features related to the appearance of the person based on the acquired image data, an estimation unit that estimates the amount of dust generated from the person based on the specified features related to the appearance of the person, and a control unit that controls an airflow generation device that generates an airflow for reducing the dust based on the estimated amount of dust generated.

[0006] The airflow control method according to one aspect of the present invention includes an acquisition step of acquiring image data of an image in which a person is reflected, a specification step of specifying characteristics related to the appearance of the person based on the acquired image data, an estimation step of estimating the amount of dust generated from the person based on the specified characteristics related to the appearance of the person, and a control step of controlling an airflow generation device that generates an airflow for reducing the dust based on the estimated amount of dust generated.

[0007] The program according to one aspect of the present invention is a program for causing a computer to execute the airflow control method.

Advantages of the Invention

[0008] The airflow control system and the airflow control method of the present invention can control the airflow based on the estimation result of the amount of dust generated.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

MODE FOR CARRYING OUT THE INVENTION

[0010] Hereinafter, embodiments will be described with reference to the drawings. Note that each of the embodiments described below shows comprehensive or specific examples. Numerical values, shapes, materials, components, arrangement positions and connection forms of components, steps, order of steps, etc. shown in the following embodiments are merely examples and are not intended to limit the present invention. In addition, among the components in the following embodiments, components not described in the independent claims are described as optional components.

[0011] Note that each figure is a schematic diagram and is not necessarily drawn precisely. Also, in each figure, substantially the same configuration is denoted by the same reference numeral, and duplicate descriptions may be omitted or simplified.

[0012] (Embodiment 1) [Configuration] First, the configuration of the airflow control system according to Embodiment 1 will be described. FIG. 1 is a block diagram showing a functional configuration of the airflow control system according to Embodiment 1. FIG. 2 is a diagram showing a schematic configuration of an airflow generation device included in the airflow control system according to Embodiment 1.

[0013] As shown in FIGS. 1 and 2, the airflow control system 10 is a system that controls an airflow as an air shower in a sub-space 50 (antechamber) provided at the entrance of a main space (such as the main space 60 described later), such as an office space. A camera 20 and an airflow generator 40 are provided in the sub-space 50. The airflow control system 10 is a system that acquires image data of an image in which a person appears, output by the camera 20, and controls the airflow (air shower) generated by the airflow generator 40 based on the acquired image data. The airflow control system 10 includes a camera 20, a control device 30, and an airflow generator 40.

[0014] The camera 20 is installed, for example, on the ceiling or wall of the sub-space 50, and photographs an image (a moving image composed of a plurality of images) including a person located in the sub-space 50 as a subject. Further, the camera 20 transmits the image data of the photographed image to the control device 30. The camera 20 may be a camera using a CMOS (Complementary Metal Oxide Semiconductor) image sensor, or may be a camera using a CCD (Charge Coupled Device) image sensor.

[0015] Further, the camera 20 may be a camera using an image sensor capable of detecting infrared rays (infrared light). That is, the camera 20 may be an infrared camera. Thereby, the camera 20 can photograph an image (infrared image) even when the sub-space 50 is dark. Note that the airflow control system 10 may include two or more cameras 20.

[0016] The control device 30 receives image data from the camera 20 and controls the airflow generation device 40 based on the received image data. The control device 30 is, for example, a local controller (i.e., an edge computer or the like) installed in the same facility where the sub-space 50 is provided, but may also be a server device (i.e., a cloud computer or the like) installed outside the facility. The control device 30 includes a communication unit 31, an information processing unit 32, and a storage unit 33.

[0017] The communication unit 31 is a communication module (communication circuit) for the control device 30 to communicate with the camera 20 and the airflow generation device 40. The communication unit 31, for example, receives image data from the camera 20 and transmits a control signal to the airflow generation device 40. The communication performed by the communication unit 31 may be wireless communication or wired communication. The communication standard used for communication is not particularly limited either.

[0018] The information processing unit 32 acquires the image data of the image received by the communication unit 31 and performs information processing for controlling the airflow generation device 40 based on the acquired image data. Specifically, the information processing unit 32 is realized by a processor or a microcomputer. The information processing unit 32 includes an acquisition unit 34, a specification unit 35, an estimation unit 36, and a control unit 37. The functions of the acquisition unit 34, the specification unit 35, the estimation unit 36, and the control unit 37 are realized by a processor or a microcomputer constituting the information processing unit 32 executing a computer program stored in the storage unit 33. Details of the functions of the acquisition unit 34, the specification unit 35, the estimation unit 36, and the control unit 37 will be described later.

[0019] The storage unit 33 is a storage device that stores the image data received by the communication unit 31 and the computer programs executed by the information processing unit 32, etc. The storage unit 33 also stores a machine learning model, an estimation model, etc., which will be described later. Specifically, the storage unit 33 is realized by a semiconductor memory or an HDD (Hard Disk Drive), etc.

[0020] The airflow generation device 40 is installed in the sub-space 50 and forms an airflow in the sub-space 50. The airflow generation device 40 includes a blower fan 41.

[0021] The blower fan 41 generates an airflow in the sub-space 50 by rotating. Note that the airflow generation device 40 may include a plurality of blower fans 41. The air volume of the blower fan 41 (that is, the strength of the airflow) and the period during which the air is blown are changed based on a control signal transmitted from the control device 30.

[0022] [Operation Example 1] Next, Operation Example 1 of the airflow control system 10 will be described. FIG. 3 is a flowchart of Operation Example 1 of the airflow control system 10.

[0023] When a person enters the sub-space 50, the camera 20 captures an image of the person and transmits the image data of the image to the control device 30. The communication unit 31 of the control device 30 receives the image data of the image of the sub-space 50 from the camera 20 (S11), and the information processing unit 32 stores the received image data in the storage unit 33 (S12).

[0024] Next, the acquisition unit 34 acquires the image data received by the communication unit 31 and stored in the storage unit 33 (S13). The identification unit 35 identifies the texture of the clothing worn by the person reflected in the image based on the acquired image data (S14). The identification unit 35 identifies the texture of the clothing using, for example, a machine learning model. FIG. 4 is a conceptual diagram of such a machine learning model.

[0025] As shown in FIG. 4, the machine learning model for identifying the texture of clothing is a machine learning model configured to be able to identify the texture of clothing using a large number of images of a person wearing clothing as learning data, and is stored in the storage unit 33 in advance. Note that identification information (such as texture A) of the texture of the clothing reflected in the image is given as a label to the images used as learning data. Note that the texture of the clothing can be rephrased as, for example, the surface state of the clothing.

[0026] The machine learning model specifically outputs a classification score based on machine learning such as a convolutional neural network (CNN: Convolutional Neural Network). The classification score is, for example, a score indicating the likelihood of the texture of the clothing shown in the image, such as texture A (smooth): 0.60, texture B (rough): 0.20, etc. The specific part 35 specifies the texture with the highest classification score as the texture of the clothing of the person shown in the image.

[0027] Next, the estimation unit 36 estimates the amount of dust generated based on the specified texture of the clothing (S15). For example, the estimation unit 36 estimates the amount of dust generated by referring to table information associating the texture of the clothing with the amount of dust expected to be generated from the clothing having the texture. FIG. 5 is a diagram showing an example of such table information. Such table information is stored in advance in the storage unit 33. Note that the amount of dust generated shown in the table information shown in FIG. 5 is appropriately determined empirically or experimentally by, for example, the designer of the airflow control system 10 or the like.

[0028] Next, the control unit 37 controls the airflow generation device 40 based on the estimated amount of dust generated (S16). The control of the airflow generation device 40 is performed by transmitting a control signal from the communication unit 31 to the airflow generation device 40. For example, when the airflow generation device 40 generates an airflow toward a person in the sub-space 50 for a certain period, the control unit 37 controls the airflow generation device 40 so that the air volume of the blower fan 41 increases (the wind becomes stronger) as the amount of dust generated increases. Also, when the airflow generation device 40 generates an airflow toward a person with a constant air volume in the sub-space 50, the control unit 37 controls the airflow generation device 40 so that the airflow generation time becomes longer as the amount of dust generated increases.

[0029] As described above, the airflow control system 10 identifies the texture of a person's clothing based on image data. The airflow control system 10 estimates the amount of dust generated from a person based on the identified texture of the person's clothing, and controls an airflow generation device 40 that generates an airflow toward the person based on the estimated amount of dust generated. Thereby, the airflow control system 10 can effectively reduce the dust expected to be generated from the person's clothing (reduce the dust around the person).

[0030] Note that the airflow control system 10 may identify the amount of clothing based on image data. Here, the amount of clothing includes not only the number of clothing items but also the amount of fabric depending on the type of clothing (thick fabric, thin fabric, sleeves and hems, length, etc.). That is, the airflow control system 10 may estimate the amount of dust generated from a person based on the identified amount of the person's clothing, and control the airflow generation device 40 that generates an airflow toward the person based on the estimated amount of dust generated.

[0031] In this case, in step S14, the specifying unit 35 specifies the amount of clothing using, for example, a machine learning model. The machine learning model for specifying the amount of clothing is a machine learning model configured to be able to identify the amount of clothing using a large number of images showing a person wearing clothing as learning data, and is stored in the storage unit 33 in advance. Note that identification information (very much, much, normal, little, very little, etc.) about the amount of clothing shown in the image is given as a label to the images used as learning data.

[0032] Specifically, the machine learning model outputs a classification score based on machine learning such as a convolutional neural network. The classification score is a score indicating the degree (the high probability of the degree) of the amount of clothing shown in the image, for example, something like considerably much: 0.60, much: 0.20 ···. The specifying unit 35 specifies the amount of clothing having the highest classification score as the amount of the person's clothing shown in the image.

[0033] When the amount of clothing is estimated by the specific unit 35, in step S15, the estimation unit 36 estimates the amount of dust generated based on the specified amount of clothing. For example, the estimation unit 36 estimates the amount of dust generated by referring to the table information that associates the amount of clothing with the amount of dust expected to be generated from the clothing. Such table information is stored in the storage unit 33 in advance. In the table information, for example, the amount of dust generated is determined such that the greater the amount of clothing, the greater the amount of dust generated. The amount of dust generated shown in the table information is appropriately determined empirically or experimentally by, for example, the designer of the airflow control system 10 or the like.

[0034] As described above, in operation example 1, the airflow control system 10 identifies features related to a person's clothing (at least one of the texture of the person's clothing and the amount of the person's clothing) based on the image data. The airflow control system 10 estimates the amount of dust generated from the person based on the identified features related to the person's clothing, and controls the airflow generation device 40 that generates an airflow (an airflow directed toward the person) for reducing dust based on the estimated amount of dust generated. Thereby, the airflow control system 10 can effectively reduce (keep away from the vicinity of the person) the dust expected to be generated from the person's clothing.

[0035] Note that the airflow control system 10 (estimation unit 36) may comprehensively estimate the amount of dust based on both the texture of the clothing and the amount of the clothing.

[0036] [Operation Example 2] Next, operation example 2 of the airflow control system 10 will be described. FIG. 6 is a flowchart of operation example 2 of the airflow control system 10.

[0037] When a person enters the sub-space 50, the camera 20 captures an image of the person and transmits the image data of the image to the control device 30. The communication unit 31 of the control device 30 receives the image data of the image of the sub-space 50 from the camera 20 (S21), and the information processing unit 32 stores the received image data in the storage unit 33 (S22).

[0038] Next, the acquisition unit 34 acquires the image data received by the communication unit 31 and stored in the storage unit 33 (S23), and the identification unit 35 identifies the hairstyle of the person shown in the image based on the acquired image data (S24). The identification unit 35 identifies the hairstyle of a person using, for example, a machine learning model. FIG. 7 is a conceptual diagram of such a machine learning model.

[0039] As shown in FIG. 7, the machine learning model for identifying a person's hairstyle is a machine learning model configured to be able to identify a person's hairstyle using a large number of images in which a person appears as learning data, and is stored in advance in the storage unit 33. Note that identification information (such as straight short hair, straight long hair, permed short hair, etc.) of the hairstyle of the person shown in the image is given as a label to the images used as learning data.

[0040] Specifically, the machine learning model outputs a classification score based on machine learning such as a convolutional neural network. The classification score is a score indicating which hairstyle the person shown in the image is likely to have, for example, straight short hair: 0.60, straight long hair: 0.20, etc. The identification unit 35 identifies the hairstyle of the person with the highest classification score as the hairstyle of the person shown in the image.

[0041] Next, the estimation unit 36 estimates the amount of dust generated based on the identified hairstyle of the person (S25). For example, the estimation unit 36 estimates the amount of dust generated by referring to table information associating a person's hairstyle with the amount of dust expected to be generated from the person's hair. FIG. 8 is a diagram showing an example of such table information. Such table information is stored in advance in the storage unit 33. Note that the amount of dust generated shown in the table information shown in FIG. 8 is appropriately determined empirically or experimentally by, for example, the designer of the airflow control system 10. In the table information, for example, the amount of dust generated is determined such that the more likely the hairstyle is to accumulate dust, the greater the amount of dust generated.

[0042] Next, the control unit 37 controls the airflow generation device 40 based on the estimated amount of dust generated (S26). The control of the airflow generation device 40 is performed by transmitting a control signal from the communication unit 31 to the airflow generation device 40. For example, when the airflow generation device 40 generates an airflow toward a person in the sub-space 50 for a certain period of time, the control unit 37 controls the airflow generation device 40 such that the air volume of the blower fan 41 increases (the wind becomes stronger) as the amount of dust generated increases. Also, when the airflow generation device 40 generates an airflow toward a person with a constant air volume in the sub-space 50, the control unit 37 controls the airflow generation device 40 such that the generation time of the airflow becomes longer as the amount of dust generated increases.

[0043] As described above, the airflow control system 10 identifies a person's hairstyle based on the image data. The airflow control system 10 estimates the amount of dust generated from the person based on the identified hairstyle of the person, and controls the airflow generation device 40 that generates an airflow toward the person based on the estimated amount of dust generated. Thereby, the airflow control system 10 can effectively reduce (keep away from the vicinity of the person) the dust expected to be generated from the person's hair.

[0044] Note that the airflow control system 10 may identify the amount of a person's hair based on the image data. That is, the airflow control system 10 may estimate the amount of dust generated from the person based on the identified amount of the person's hair, and control the airflow generation device 40 that generates an airflow toward the person based on the estimated amount of dust generated.

[0045] In this case, in step S24, the specifying unit 35 specifies the amount of a person's hair using, for example, a machine learning model. The machine learning model for specifying the amount of a person's hair is a machine learning model configured to be able to identify the amount of a person's hair using a large number of images in which the person appears as learning data, and is stored in advance in the storage unit 33. Note that identification information (very much, much, normal, little, very little, etc.) about the amount of the person's hair reflected in the image used as the learning data is given as a label.

[0046] The machine learning model specifically outputs a classification score based on machine learning such as a convolutional neural network. The classification score is, for example, a score indicating the amount of a person's hair in an image (how likely it is to be a certain amount), such as "quite a lot: 0.60, a lot: 0.20...". The specifying unit 35 specifies the amount of the person's hair with the highest classification score as the amount of the person's hair in the image.

[0047] When the amount of a person's hair is estimated by the specifying unit 35, in step S25, the estimating unit 36 estimates the amount of dust generation based on the specified amount of the person's hair. For example, the estimating unit 36 estimates the amount of dust generation by referring to table information associating the amount of a person's hair with the amount of dust expected to be generated from the person's hair. Such table information is stored in the storage unit 33 in advance. In the table information, for example, the amount of dust generation is determined such that the greater the amount of a person's hair, the greater the amount of dust generation. The amount of dust generation shown in the table information is appropriately determined empirically or experimentally by, for example, the designer of the airflow control system 10.

[0048] As described above, in operation example 2, the airflow control system 10 identifies features related to a person's hair (at least one of the person's hairstyle and the amount of the person's hair) based on the image data. The airflow control system 10 estimates the amount of dust generated from a person based on the identified features related to the person's hair, and controls the airflow generator 40 that generates an airflow (an airflow directed toward the person) for reducing dust based on the estimated amount of dust generation. Thereby, the airflow control system 10 can effectively reduce (keep away from the vicinity of the person) the dust expected to be generated from the person's hair.

[0049] Note that the airflow control system 10 (estimating unit 36) may comprehensively estimate the amount of dust based on both the person's hairstyle and the amount of the person's hair.

[0050] (Embodiment 2) [Configuration] Next, the configuration of the airflow control system according to Embodiment 2 will be described. FIG. 9 is a block diagram showing the functional configuration of the airflow control system according to Embodiment 2. FIG. 10 is a diagram showing the schematic configuration of the airflow generation device included in the airflow control system according to Embodiment 2.

[0051] As shown in FIGS. 9 and 10, the airflow control system 70 is a system that acquires the image data of the main space 60 output by the camera 20 and controls the airflow in the main space 60 based on the acquired image data. The main space 60 is, for example, an office space, but may also be an indoor space in other facilities such as a space in a commercial facility or a space in a house. As shown in FIGS. 9 and 10, the airflow control system 70 includes a camera 20, a control device 30, and an airflow generation device 80.

[0052] The camera 20 and the control device 30 are substantially the same devices as those in Embodiment 1 except that they are targeted at the main space 60 and the airflow generation device 80, so the description thereof will be omitted. The airflow generation device 80 is installed in the main space 60 and forms an airflow in the main space 60. The airflow generation device 80 includes a blower fan 81 and a louver 82.

[0053] The blower fan 81 generates an airflow in the main space 60 by rotating. Note that the airflow generation device 80 may include a plurality of blower fans 81. The air volume of the blower fan 81 (that is, the strength of the airflow) and the period during which the blowing is performed are changed based on the control signal transmitted from the control device 30.

[0054] The louver 82 is a structure for changing the direction of the airflow. In other words, the louver 82 is a guide structure for guiding the airflow. The louver 82 is, for example, a feather-shaped structure whose posture (angle of the blades) is changed based on the control signal transmitted by the control device 30.

[0055] Note that the airflow generation device 80 is, for example, a dedicated device of the airflow control system 70. However, as the airflow generation device 80, a ventilation device or an air conditioner installed in the main space 60 in advance may be diverted and used.

[0056] [Operation Example 1] Next, Operation Example 1 of the airflow control system 70 will be described. FIG. 11 is a flowchart of Operation Example 1 of the airflow control system 70.

[0057] The specifying unit 35 of the control device 30 performs a process of specifying the amount of clothing worn by a person reflected in the image based on the image data output by the camera 20 (S31). The process of step S31 is, in more detail, as described in Operation Example 1 of Embodiment 1, and is performed periodically.

[0058] Next, the estimating unit 36 determines whether there has been a change in the clothing (S32). For example, the estimating unit 36 determines whether the amount of clothing worn specified last time is different from the amount of clothing worn specified this time. If there is no change in the clothing, it is considered that the person has not put on or taken off clothes and no dust is generated. Therefore, when the estimating unit 36 determines that there is no change in the clothing (No in S32), it estimates that the person has not put on or taken off clothes and no dust is generated (the amount of dust generated is zero) (S33). In this case, the process of step S31 continues to be performed periodically.

[0059] On the other hand, when the estimating unit 36 determines that there has been a change in the clothing (Yes in S32), it estimates that dust is generated when the person puts on or takes off clothes (for example, the amount of dust generated is a predetermined amount) (S34). In this case, the control unit 37 performs control to strengthen the airflow generated by the airflow generation device 80 (S35). The control of the airflow generation device 80 is performed by transmitting a control signal from the communication unit 31 to the airflow generation device 80. Specifically, the control unit 37 controls the air volume of the blower fan 81 so that the airflow toward the person reflected in the image becomes stronger.

[0060] As a result, the airflow control system 70 can effectively reduce (move away from the vicinity of a person) the dust generated around the person when the person puts on or takes off clothing. For example, the airflow control system 70 can guide the generated dust to a predetermined area of the main space 60. Note that the control for strengthening the airflow here may be control for operating the airflow generation device 80 that is stopped, or may be control for further increasing the air volume of the airflow generation device 80 that is operating.

[0061] Note that when the person's position is generally determined, such as when the person is facing a desk and the airflow generation device 80 is installed so as to face the person's position, the control unit 37 does not need to control the direction of the airflow. On the other hand, when the person's position varies, the control unit 37 may estimate the person's position based on the image data and generate an airflow toward the estimated person's position by controlling the louver 82.

[0062] The control in step S35 is performed for a certain period, and when the control ends, the process of step S31 is performed again.

[0063] As described above, the airflow control system 70 identifies changes in a person's clothing based on image data. Changes in a person's clothing are an example of features related to a person's clothing. The airflow control system 70 estimates the amount of dust generated from the person (the presence or absence of dust generation) based on the identified changes in the person's clothing, and controls the airflow generation device 40 that generates an airflow (an airflow toward the person) for reducing dust based on the estimated amount of dust generated. As a result, the airflow control system 70 can effectively reduce (move away from the vicinity of the person) the dust expected to be generated based on the person putting on or taking off clothing.

[0064] Note that in step S31, the specific part 35 performs a process of specifying the texture of the clothing worn by the person shown in the image based on the image data. In step S32, the estimation unit 36 may determine whether there is a change in the clothing by determining whether the texture of the clothing specified this time is different from the texture of the clothing specified last time. That is, as a means for determining whether there is a change in the clothing, in addition to the method based on the amount of clothing, a method based on the texture of the clothing is also conceivable.

[0065] Further, similar to the airflow control system 10, the airflow control system 70 can also estimate the amount of dust based on the texture of the clothing or the amount of clothing. Therefore, the airflow control system 70 may comprehensively estimate the amount of dust based on two or more of the texture of the clothing, the amount of clothing, and the change in the clothing.

[0066] [Operation Example 2] Next, an operation example 2 of the airflow control system 70 will be described. FIG. 12 is a flowchart of the operation example 2 of the airflow control system 70.

[0067] The specific part 35 of the control device 30 performs a process of specifying the movement of the person's hair shown in the image based on the moving image data (image data of a plurality of images) output by the camera 20 (S41). For example, the specific part 35 can perform a process of extracting the contour of the person's head for each of the image data of the plurality of images, and specify the temporal change of the contour of the head as the movement of the person's hair. Further, the specific part 35 may specify the movement of the person's hair using a machine learning model.

[0068] Next, the estimation unit 36 determines whether the movement of the specified person's hair is large (S42). For example, the estimation unit 36 determines whether the amount of change in the shape or size of the contour is equal to or greater than a predetermined value. When the movement of the person's hair is small, it is considered that no dust is generated by the movement of the person's hair. Therefore, when the estimation unit 36 determines that the movement of the person's hair is small (No in S42), it is estimated that no dust is generated (the amount of dust generation is zero) (S43). In this case, the process of step S41 is continuously performed periodically.

[0069] On the other hand, when the estimation unit 36 determines that the movement of a person's hair is large (Yes in S42), it estimates that dust is generated due to the movement of the person's hair (for example, the amount of dust generated is a predetermined amount) (S44). In this case, the control unit 37 performs control to strengthen the airflow generated by the airflow generation device 80 (S45). The control of the airflow generation device 80 is performed by transmitting a control signal from the communication unit 31 to the airflow generation device 80. Specifically, the control unit 37 controls the air volume of the blower fan 81 so that the airflow toward the person reflected in the image becomes stronger.

[0070] Thereby, the airflow control system 70 can effectively reduce (move away from the vicinity of the person) the dust generated around the person due to the movement of the person's hair. For example, the airflow control system 70 can guide the generated dust to a predetermined area of the main space 60. Note that the control to strengthen the airflow here may be control to operate the stopped airflow generation device 80, or may be control to further increase the air volume of the operating airflow generation device 80.

[0071] Note that when the person's position is generally determined, such as when the person is facing the desk and the airflow generation device 80 is installed to face the person's position, the control unit 37 does not need to control the direction of the airflow. On the other hand, when the person's position varies, the control unit 37 may estimate the person's position based on the image data and generate an airflow toward the estimated person's position by controlling the louver 82.

[0072] The control in step S45 is performed for a certain period. When the control ends, the process of step S41 is performed again.

[0073] Note that in steps S42 to S44, the estimation unit 36 simply estimates the presence or absence of dust generation. However, the estimation unit 36 may divide the movement of the person's hair into multiple stages for determination and finely estimate the amount of dust generation in multiple stages according to the determination result.

[0074] As described above, the airflow control system 70 identifies the movement of a person's hair based on image data. The movement of a person's hair is an example of a feature related to the person's hair. The airflow control system 70 estimates the amount of dust generated from the person (the presence or absence of dust generation) based on the identified movement of the person's hair, and based on the estimated amount of dust generation, controls an airflow generation device 80 that generates an airflow (an airflow directed toward the person) for reducing dust. Thereby, the airflow control system 70 can effectively reduce (keep away from the vicinity of the person) the dust expected to be generated based on the movement of the person's hair.

[0075] Note that, similar to the airflow control system 10, the airflow control system 70 can also estimate the amount of dust based on the hairstyle or the amount of a person's hair. Therefore, the airflow control system 70 may comprehensively estimate the amount of dust based on two or more of the person's hairstyle, the amount of the person's hair, and the movement of the person's hair.

[0076] [Operation Example 3] Next, operation example 3 of the airflow control system 70 will be described. FIG. 13 is a flowchart of operation example 3 of the airflow control system 70.

[0077] A specifying unit 35 of the control device 30 specifies time-series data of a skeletal model of a person shown in an image based on image data of a moving image output by a camera 20 (image data of a plurality of images) (S51). The skeletal model is a model in which spheres indicating the positions of joints are connected by links. FIG. 14 is a diagram showing an example of the skeletal model. The time-series data of the skeletal model is, in other words, data indicating the change over time of the coordinates of the joints. Any existing method may be used for specifying the skeletal model.

[0078] Next, an estimating unit 36 determines whether the movement of the person shown in the image corresponds to a predetermined movement based on the specified time-series data of the skeletal model (S52). In other words, the estimating unit 36 detects a predetermined movement of the person shown in the image.

[0079] The predetermined movement is specifically at least one of movements expected to generate dust, such as a movement to brush off dust, a movement to touch hair, and a movement to put on or take off clothes. The estimation unit 36 can determine whether the movement of the person reflected in the image corresponds to a predetermined movement based on whether the time-series data of the skeletal model (the change over time of the joint coordinates) is similar to a predetermined change pattern corresponding to the predetermined movement. Note that in steps S51 and S52, instead of the method using the skeletal model, a method using pattern matching or a method using a machine learning model may be used to detect a predetermined movement.

[0080] When the estimation unit 36 determines that the movement of the person reflected in the image does not correspond to a predetermined movement (No in S52), it is estimated that no dust is generated (the amount of dust generated is zero) (S53). In this case, the process of step S51 is continuously performed periodically.

[0081] On the other hand, when the estimation unit 36 determines that the movement of the person reflected in the image corresponds to a predetermined movement (Yes in S52), it is estimated that dust is generated by the movement of the person (for example, the amount of dust generated is a predetermined amount) (S54). In this case, the control unit 37 performs control to strengthen the airflow generated by the airflow generation device 80 (S55). The control of the airflow generation device 80 is performed by transmitting a control signal from the communication unit 31 to the airflow generation device 80. Specifically, the control unit 37 controls the air volume of the blower fan 81 so that the airflow toward the person reflected in the image becomes stronger.

[0082] Thereby, the airflow control system 70 can effectively reduce (keep away from the vicinity of the person) the dust generated around the person when the person moves. For example, the airflow control system 70 can guide the generated dust to a predetermined area of the main space 60. Note that the control to strengthen the airflow here may be control to operate the stopped airflow generation device 80 or control to further increase the air volume of the operating airflow generation device 80.

[0083] Incidentally, when a person is facing the desk and the position of the person is generally determined, and the airflow generation device 80 is installed so as to face the position of the person, the control unit 37 does not need to control the direction of the airflow. On the other hand, when the position of the person varies, the control unit 37 may estimate the position of the person based on the image data and generate an airflow toward the estimated position of the person by controlling the louver 82.

[0084] The control in step S55 is performed for a certain period, and when the control ends, the process of step S51 is performed again.

[0085] In steps S52 to S54, the estimation unit 36 simply estimated the presence or absence of dust generation. However, the estimation unit 36 may finely estimate the amount of dust generation according to which of a plurality of predetermined movements the movement of the person corresponds to.

[0086] As described above, the airflow control system 70 identifies the movement of a person based on image data. The movement of the person is an example of a feature related to the movement of the person. The airflow control system 70 estimates the amount of dust generated from the person (the presence or absence of dust generation) based on the identified movement of the person, and controls the airflow generation device 80 that generates an airflow (an airflow toward the person) for reducing dust based on the estimated amount of dust generation. Thereby, the airflow control system 70 can effectively reduce (keep away from the vicinity of the person) the dust that is expected to be generated based on the movement of the person.

[0087] (Modification example) The processes described in Embodiments 1 and 2 may be combined as appropriate. For example, the airflow control system 10 may use the method for estimating the amount of dust generation described in Embodiment 2. More specifically, the airflow control system 10 can also perform an operation of strengthening the airflow from the airflow generation device 40 toward the person by instructing the person standing in front of the airflow generation device 40 to perform a specific operation. Similarly, the airflow control system 70 may use the method for estimating the amount of dust generation described in Embodiment 2.

[0088] In addition, each of the airflow control system 10 and the airflow control system 70 may estimate the amount of dust generated by combining two or more of the characteristics related to a person's clothing, the characteristics related to a person's hair, and the characteristics related to a person's movement. Each of the airflow control system 10 and the airflow control system 70 may control the airflow based on the total amount of dust generated, which is the sum of the amount of dust generated based on the characteristics related to a person's clothing, the amount of dust generated based on the characteristics related to a person's hair, and the amount of dust generated based on the characteristics related to a person's movement.

[0089] In addition, each of the airflow generating device 40 and the airflow generating device 80 was a device that generates an airflow from the device toward a person, but may be a device that generates an airflow from the person toward the device. That is, each of the airflow generating device 40 and the airflow generating device 80 may include a suction fan instead of a blower fan.

[0090] (Effect, etc.) As described above, the airflow control system 10 (or the airflow control system 70) includes an acquisition unit 34 that acquires image data of an image in which a person is reflected, a specification unit 35 that specifies characteristics related to a person's appearance based on the acquired image data, an estimation unit 36 that estimates the amount of dust generated from the person based on the specified characteristics related to the person's appearance, and a control unit 37 that controls an airflow generating device 40 (or an airflow generating device 80) that generates an airflow for reducing dust based on the estimated amount of dust generated.

[0091] Such an airflow control system 10 can control the airflow based on the estimation result of the amount of dust generated.

[0092] In addition, for example, the characteristics related to a person's appearance are the characteristics related to a person's clothing.

[0093] Such an airflow control system 10 can estimate the amount of dust generated from a person based on the characteristics related to the person's clothing.

[0094] In addition, for example, the characteristics related to a person's clothing include the texture of the person's clothing.

[0095] Such an airflow control system 10 can estimate the amount of dust generated from a person based on the texture of the person's clothing.

[0096] Also, for example, the characteristics related to a person's clothing include the amount of the person's clothing.

[0097] Such an airflow control system 10 can estimate the amount of dust generated from a person based on the amount of the person's clothing.

[0098] Also, for example, the characteristics related to a person's clothing include changes in the person's clothing.

[0099] Such an airflow control system 10 can estimate the amount of dust generated from a person based on changes in the person's clothing.

[0100] Also, for example, the characteristics related to a person's appearance are characteristics related to the person's hair.

[0101] Such an airflow control system 10 can estimate the amount of dust generated from a person based on the characteristics related to the person's appearance.

[0102] Also, for example, the characteristics related to a person's hair include the amount of the person's hair.

[0103] Such an airflow control system 10 can estimate the amount of dust generated from a person based on the amount of the person's hair.

[0104] Also, for example, the characteristics related to a person's hair include the hairstyle of the person's hair.

[0105] Such an airflow control system 10 can estimate the amount of dust generated from a person based on the hairstyle of the person's hair.

[0106] Also, for example, the characteristics related to a person's hair include the movement of the person's hair.

[0107] Such an airflow control system 10 can estimate the amount of dust generated from a person based on the movement of the person's hair.

[0108] Also, for example, features related to a person's appearance are features related to the person's movement.

[0109] Such an airflow control system 10 can estimate the amount of dust generated from a person based on features related to the person's movement.

[0110] Also, for example, features related to the person's movement include the movement of the person brushing off dust.

[0111] Such an airflow control system 10 can estimate the amount of dust generated from a person based on the movement of the person brushing off dust.

[0112] Also, an airflow control method executed by a computer such as the airflow control system 10 (or the airflow control system 70) includes an acquisition step of acquiring image data of an image in which a person is reflected, a specification step part of specifying features related to the person's appearance based on the acquired image data, an estimation step of estimating the amount of dust generated from the person based on the specified features related to the person's appearance, and a control step of controlling an airflow generation device 40 (or the airflow generation device 80) that generates an airflow for reducing dust based on the estimated amount of dust generated.

[0113] Such an airflow control method can control the airflow based on the estimation result of the amount of dust generated.

[0114] (Other embodiments) As described above, the airflow control system and the airflow control method according to the embodiments have been described, but the present invention is not limited to the above embodiments.

[0115] In the above-described embodiment, the airflow control system is implemented by a plurality of devices, but it may also be implemented as a single device. For example, the airflow control system may be implemented as a single device corresponding to a control device. When the airflow control system is implemented by a plurality of devices, each component included in the airflow control system may be distributed among the plurality of devices in any manner.

[0116] In the above embodiment, the processing executed by a specific processing unit may be executed by another processing unit. Also, the order of a plurality of processes may be changed, or a plurality of processes may be executed in parallel.

[0117] In the above embodiment, each component may also be realized by executing a software program suitable for each component. Each component may be realized by a program execution unit such as a CPU or a processor reading and executing a software program recorded on a recording medium such as a hard disk or a semiconductor memory.

[0118] Each component may also be realized by hardware. For example, each component may be a circuit (or an integrated circuit). These circuits may form one circuit as a whole, or may be separate circuits respectively. Also, each of these circuits may be a general-purpose circuit or a dedicated circuit.

[0119] Furthermore, the general or specific aspects of the present invention may be realized by a system, a device, a method, an integrated circuit, a computer program, or a recording medium such as a computer-readable CD-ROM. Also, it may be realized by any combination of a system, a device, a method, an integrated circuit, a computer program, and a recording medium. For example, the present invention may be realized as a program for causing a computer to execute the airflow control method of the above embodiment, or may be realized as a computer-readable non-transitory recording medium storing such a program.

[0120] In addition, forms obtained by applying various modifications that occur to those skilled in the art to each embodiment, or forms realized by arbitrarily combining the components and functions in each embodiment without departing from the gist of the present invention are also included in the present invention.

Explanation of Reference Numerals

[0121] 10, 70 airflow control system 20 camera 30 control device 31 communication unit 32 information processing unit 33 storage unit 34 acquisition unit 35 identification unit 36 estimation unit 37 control unit 40, 80 airflow generation device 41, 81 blower fan 50 sub - space 60 main space 82 louver

Claims

1. An acquisition unit that acquires image data of an image in which a person appears; An identification unit that identifies features related to the appearance of the person based on the acquired image data; An estimation unit that estimates the amount of dust generated from the person based on the identified features related to the appearance of the person; A control unit that controls an airflow generation device that generates an airflow for an air shower directed toward the person in order to move the dust away from around the person based on the estimated amount of dust generated; and The features related to the appearance of the person include at least one of the texture of the person's clothing, the amount of the person's clothing, changes in the person's clothing, the amount of the person's hair, the hairstyle of the person, the movement of the person's hair, the movement of the person brushing dust off, the movement of the person touching their hair, and the movement of the person putting on or taking off clothing An airflow control system for an air shower.

2. The features related to the appearance of the person include the texture of the person's clothing The airflow control system for an air shower according to claim 1.

3. The features related to the appearance of the person include the amount of the person's clothing The airflow control system for an air shower according to claim 1 or 2.

4. The features related to the appearance of the person include changes in the person's clothing The airflow control system for an air shower according to claim 1 or 2.

5. The features related to the appearance of the person include the amount of the person's hair The airflow control system for an air shower according to claim 1 or 2.

6. The features related to the appearance of the person include the hairstyle of the person The airflow control system for an air shower according to claim 1 or 2.

7. The features related to the appearance of the person include the movement of the person's hair The airflow control system for an air shower according to claim 1 or 2.

8. The features related to the appearance of the person include the movement of the person brushing dust off The airflow control system for an air shower according to claim 1 or 2.

9. The features related to the appearance of the person include the movement of the person touching their hair The airflow control system for an air shower according to claim 1 or 2.

10. The features related to the appearance of the person include the movement of the person putting on or taking off clothing The airflow control system for an air shower according to claim 1 or 2.

11. An acquisition step of acquiring image data of an image in which a person appears; An identification step of identifying features related to the appearance of the person based on the acquired image data; An estimation step of estimating the amount of dust generated from the person based on the identified features related to the appearance of the person; A control step of controlling an airflow generation device that generates an airflow for an air shower directed toward the person in order to move the dust away from around the person based on the estimated amount of the generated dust is included. The characteristics regarding the appearance of the person include at least one of the texture of the person's clothing, the amount of the person's clothing, the change in the person's clothing, the amount of the person's hair, the hairstyle of the person, the movement of the person's hair, the movement of the person to brush off dust, the movement of the person to touch the hair, and the movement of the person to put on or take off clothing. An airflow control method for an air shower.

12. A program for causing a computer to execute the airflow control method for an air shower according to Claim 11.

Images