Water intake estimation device and water intake estimation method

Abstract

To provide a water intake amount estimation program, a water intake amount estimation device, and a water intake amount estimation method capable of estimating a water intake amount even when used by an unspecified number of people without imposing a personal behavior restriction.SOLUTION: The program for estimating water intake acquires a group of temporally consecutive images of at least one person who drinks water from a camera disposed at a predetermined location, detects at least the person, a position of a body part of the person, and a container size of a container from the group of images, calculates an amount of change in the position of the body part of the person and a time from when the body part of the person changes from an initial position to when the body part returns to the initial position, and estimates water intake of the person based on the amount of change, the container size, and the time.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a water intake estimation program, a water intake estimation device, and a water intake estimation method. [Background technology]

[0002] Conventionally, many posture estimation methods based on machine learning have been proposed for estimating a person's posture, and they boast high accuracy. For example, Patent Document 1 discloses a technology in which a target image of a person is rotated in a direction that allows estimation of the person's posture, and then the posture of the person is estimated from the rotated image, generating a posture estimation result that indicates the positions of the person's joint points, and then performing a reverse rotation on this posture estimation result to return the rotated image to the target image before rotation, thereby estimating the posture. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Publication No. 2019-219989 Summary of the Invention [Problem to be solved by the invention]

[0004] Incidentally, most of the conventional technologies for predicting and estimating human fluid intake are intended for use in health management situations, mainly in the medical, nursing, and care industries. Known estimation methods in conventional technologies include those that record urination, those that use a container containing ingested fluid, and those that use physiological data collected by a sensor placed in contact with the skin surface.

[0005] However, the above-mentioned conventional technology is only intended for use in managing the health of an identified individual, and requires restrictions on the identified individual's behavior, such as specifying a place to urinate, specifying a container for drinking water, and keeping the individual still to attach a sensor, leaving room for improvement.

[0006] The present invention has been made in consideration of the above, and aims to provide a water intake estimation program, a water intake estimation device, and a water intake estimation method that can estimate water intake even when used by an unspecified number of people, without imposing restrictions on individual behavior. [Means for solving the problem]

[0007] In order to solve the above-mentioned problems and achieve the object, the water intake estimation program of the present invention is a water intake estimation program executed by a water intake estimation device, and executes the following steps: an acquisition step of acquiring a group of temporally consecutive images of one or more people drinking water from a camera placed at a predetermined location; a detection step of detecting at least the people, the positions of their body parts and the size of a container from the group of images; a calculation step of calculating the amount of change in the position of the people's body parts detected in the detection step and the time it takes for the people's body parts to change from their initial position and return to their initial position; and an estimation step of estimating the amount of water intake drank by the people based on the amount of change, the container size and the time.

[0008] In addition, the water intake estimation program according to the present invention, in the above invention, further executes an output step of outputting the water intake estimated in the estimation step to a preset device.

[0009] In addition, in the water intake estimation program of the present invention, in the above invention, the detection step further detects the characteristics of the person, and the output step refers to the representative device at the work site where the person is located, or an address information recording unit that associates the characteristics of the person with the addresses of devices carried by the person, and outputs the water intake estimated in the estimation step to the device corresponding to the characteristics of the person detected in the detection step.

[0010] In addition, the water intake estimation device of the present invention includes an acquisition unit that acquires a group of temporally consecutive images of one or more people drinking water from a camera placed at a predetermined location; a detection unit that detects at least the people, the positions of their body parts, and the size of a container from the group of images; a calculation unit that calculates the amount of change in the position of the people's body parts detected in the detection step and the time it takes for the people's body parts to change from their initial position and return to their initial position; and an estimation unit that estimates the amount of water intake drank by the people based on the amount of change, the container size, and the time.

[0011] Furthermore, the water intake estimation method of the present invention is a water intake estimation method executed by a water intake estimation device, and includes the steps of: acquiring a group of temporally consecutive images of one or more people drinking water from a camera arranged at a predetermined location; detecting at least the people, the positions of their body parts, and the size of a container from the group of images; calculating the amount of change in the position of the people's body parts detected in the detection step and the time it takes for the people's body parts to change from their initial position and return to their initial position; and estimating the amount of water intake drank by the people based on the amount of change, the container size, and the time. [Effects of the Invention]

[0012] According to the present invention, it is possible to estimate water intake even when used by an unspecified number of people, without imposing restrictions on individual behavior. [Brief explanation of the drawings]

[0013] [Figure 1] FIG. 1 is a block diagram showing the functional configuration of a water intake estimation device according to an embodiment of the present invention. [Figure 2] FIG. 2 is a flowchart showing an outline of the process executed by the water intake amount estimating device according to one embodiment of the present invention. [Figure 3]FIG. 3 is a diagram showing an example of a group of images acquired by an acquisition unit included in a water intake estimation device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0014] A water intake estimation device capable of executing a water intake estimation program and a water intake estimation method according to the present invention will be described in detail below with reference to the drawings. Note that the present disclosure is not limited to the following embodiments. Furthermore, the figures referred to in the following description merely show a schematic representation of the shape, size, and positional relationship to the extent that the contents of the present disclosure can be understood. In other words, the present disclosure is not limited to the shape, size, and positional relationship exemplified in each figure. Furthermore, in the drawings, the same parts are denoted by the same reference numerals.

[0015] [Functional configuration of the water intake estimation device] Figure 1 is a block diagram showing the functional configuration of a fluid intake estimation device according to one embodiment of the present invention. The fluid intake estimation device 10 shown in Figure 1 is a device that estimates the amount of fluid intake of a person who appears in a group of temporally consecutive images when that person drinks water. The fluid intake estimation device 10 includes a communication unit 11, an input unit 12, a display unit 13, a recording unit 14, and a control unit 15.

[0016] Under the control of the control unit 15, the communication unit 11 communicates with the camera 100 placed at a location where a person drinks water, sequentially acquires image files from the camera 100, each including a group of temporally consecutive images, the location where the camera 100 is placed, and identification information for identifying the camera 100, and outputs the acquired image files to the control unit 15. Here, the group of temporally consecutive images from the camera 100 refers to video data captured by the camera 100 or multiple temporally consecutive still image data captured by the camera 100 at a fixed time interval, for example, 30 FPS. The location includes latitude and longitude, etc. The communication unit 11 also outputs various information to an external mobile terminal, etc., under the control of the control unit 15. The communication unit 11 is configured using, for example, a communication module capable of 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark). The camera 100 is assumed to be, for example, a device capable of capturing image data or video data at predetermined intervals, such as a surveillance camera, video camera, camcorder, camera-equipped mobile phone, or camera-equipped tablet terminal device.

[0017] The input unit 12 receives input of various operations from the user and outputs information corresponding to the received operations to the control unit 15. The input unit 12 is configured using an input interface such as a keyboard, a mouse, or a touch panel.

[0018] The display unit 13 displays various types of information and images under the control of the control unit 15. The display unit 13 is configured using an organic electroluminescent display (OLED) or a liquid crystal display.

[0019] The recording unit 14 records various information related to the water intake estimation device 10, data currently being processed, etc. The recording unit 14 is configured using a hard disk drive (HDD), a solid state drive (SSD), flash memory, volatile memory, non-volatile memory, etc. The recording unit 14 has a program recording unit 141, a trained model recording unit 142, and an address information recording unit 143.

[0020] The program recording unit 141 records various programs including the water intake estimation program executed by the water intake estimation device 10.

[0021] The trained model recording unit 142 records multiple trained models trained using training data. The trained model recording unit 142 records trained models that have learned a person's posture through machine learning such as deep learning, trained models that have learned container sizes, trained models that have learned water intake, and the like. A trained model that has learned a person's posture is, for example, a posture estimation trained model. This posture estimation trained model learns training data that associates multiple still image data or video data depicting a person or other subject with annotation information that annotates the still image data or video data with the person's joint points and body parts. From the input still image data or video data, the trained model outputs, as output results, the presence or absence of a person appearing in a still image corresponding to the still image data, and the person's body parts based on the human posture connected to the person's joint points as output parameters. Here, the person's body parts refer to the coordinate positions of each of the joints, eyes, mouth, nose, chin, elbows, hands, and arms in the still image. Furthermore, a trained model that has learned container size (hereinafter simply referred to as the "container size trained model") learns training data that associates multiple still image data and video data with annotation information that annotates the still image data and video data with container size and container capacity, and outputs the container size that appears in a still image corresponding to the input still image data or video data as an output parameter as an output result from the input still image data or video data. Furthermore, a trained model that has learned water intake (hereinafter simply referred to as the "water intake trained model") learns training data that associates the amount of change in a person's body part, container size, the time it takes to return to the initial position, and the person's water intake, and outputs the water intake as an output parameter as an output result from the input amount of change in a person's body part, container size, and time it takes to return to the initial position.

[0022] The address information recording unit 143 records address information that associates the name of each of multiple people, features obtained by image detection or the like for each of the multiple people, the job title (monitoring position) of each of the multiple people, the affiliation of each of the multiple people, and the address (contact telephone number or email address) of a mobile phone or the like owned by each of the multiple people.

[0023] The control unit 15 is realized using a processor having hardware such as an FPGA (Field-Programmable Gate Array), GPU (Graphics Processing Unit), or CPU (Central Processing Unit), and a memory that serves as a temporary storage area used by the processor. The control unit 15 controls each component of the water intake estimation device 10. The control unit 15 has an acquisition unit 151, a detection unit 152, a calculation unit 153, an estimation unit 154, an output control unit 155, and a learning unit 156.

[0024] The acquisition unit 151 acquires, via the communication unit 11, a group of temporally consecutive images (a group of temporally consecutive image data captured at a predetermined frame rate) showing one or more people drinking water from a camera 100 placed at a predetermined location.

[0025] The detection unit 152 detects at least the position of the person, the body part of the person, and the size of the container from the group of images acquired by the acquisition unit 151.

[0026] The calculation unit 153 calculates the amount of change in the position of the person's body part and the time it takes for the person's body part to change from its initial position and return to its initial position, based on the group of images acquired by the acquisition unit 151 and the person's body part detected by the detection unit 152.

[0027] The estimation unit 154 estimates the amount of water intake that the person has drunk based on the amount of change in the person's body part calculated by the calculation unit 153, the size of the container, and the time it takes for the container to return to its initial position.

[0028] The output control unit 155 outputs the estimation result estimated by the estimation unit 154 to the display unit 13. Furthermore, the output control unit 155 refers to the address of a mobile phone or the like carried by the work site supervisor recorded in the address information recording unit 143 via the communication unit 11, and outputs the estimation result estimated by the estimation unit 154 to the mobile phone carried by the work site supervisor.

[0029] The learning unit 156 re-learns the water intake learned model based on the estimation result estimated by the estimation unit 154, the amount of change in the person's body part calculated by the calculation unit 153, the container size, the time until returning to the initial position, and the water intake learned model.

[0030] [Processing by the water intake estimation device] Next, we will explain the processing executed by the water intake estimation device 10. Figure 2 is a flowchart showing an outline of the processing executed by the water intake estimation device 10.

[0031] As shown in FIG. 2, first, the acquisition unit 151 acquires an image group from the camera 100 via the communication unit 11 (step S101). FIG. 3 is a diagram showing an example of an image group acquired by the acquisition unit 151. As shown in FIG. 3, the acquisition unit 151 acquires temporally consecutive images P from the camera 100 via the communication unit 11. t , Image P t+1 , Image P t+2 3, three images that are consecutive in time are described as an example of an image group, but the present invention is not limited to this, and there may be a plurality of images in the time between each image, and these can be selected appropriately depending on the imaging function of the camera 100 and the communication function of the communication unit 11.

[0032] Next, the detection unit 152 detects a person from the group of images acquired by the acquisition unit 151 (step S102), detects the body parts of the person (step S103), and detects the container size of the container from the group of images (step S104). Specifically, the detection unit 152 detects the posture estimation trained model and the container size trained model recorded by the trained model recording unit 142 from the image P t , Image Pt+1 , Image P t+2 3, the detection unit 152 detects the person O1 in the image P, the body parts of the person O1, and the size of the container D1 held by the person O1. t , Image P t+1 , Image P t+2 The detection unit 152 detects the person O1 appearing in each of the images P and detects the elbow H1, chin A1, and nose N1 as the body parts of the person O1. t , Image P t+1 , Image P t+2 The detection unit 152 detects the container D1 that appears in each of the images and detects the size of the container D1. Note that the detection unit 152 may detect the feature amount of the person O1 using a posture estimation trained model, a well-known pattern matching technique, or the like.

[0033] Thereafter, the calculation unit 153 calculates the amount of change in the position of the person's body part and the time it takes for the person's body part to change from its initial position and return to its initial position again, based on the group of images acquired by the acquisition unit 151 and the person's body part detected by the detection unit 152 (step S105). Specifically, as shown in FIG. 3, the calculation unit 153 calculates the amount of change in the position of the person's body part and the time it takes for the person's body part to change from its initial position and return to its initial position again, based on the group of images P t , Image P t+1 , Image P t+2 The calculation unit 153 calculates the amount of change in the elbow H1, for example, the amount of change in angle or the amount of movement, from the time change of the elbow H1 of the body part of the person O1 that appears in each of the images P. Of course, the calculation unit 153 may calculate the amount of change in the nose N1 and the chin A1 in addition to the elbow H1. Furthermore, as shown in FIG. 3, the calculation unit 153 calculates the amount of change in the nose N1 and the chin A1 from the time change of the elbow H1 of the body part of the person O1 that appears in each of the images P. t , Image P t+1 , Image P t+2 Specifically, the calculation unit 153 calculates the time it takes for the elbow H1 of the body part of the person O1 shown in each of the images P to move from the initial position and return to the initial position again. t Image P from the time t+2 Based on the time, the time taken for the elbow H1 of the body part of the person O1 to move from the initial position and return to the initial position is calculated.

[0034] Next, the estimation unit 154 estimates the amount of water intake drank by the person based on the amount of change in the person's body part calculated by the calculation unit 153, the container size, and the time until returning to the initial position (step S106). Specifically, the estimation unit 154 uses the water intake trained model recorded by the trained model recording unit 142 to input the amount of change (angle) in the person's body part calculated by the calculation unit 153, the container size, and the time until returning to the initial position as input parameters, and outputs the amount of water intake drank by the person as an output parameter to estimate the amount of water intake. That is, the estimation unit 154 inputs the amount of change in the person's body part calculated by the calculation unit 153 (see (A) in FIG. 3), the container size (see (B) in FIG. 3), and the time until returning to the initial position (see (C) in FIG. 3) as input parameters to the water intake trained model, and estimates the amount of water intake as an output parameter.

[0035] Thereafter, the output control unit 155 outputs the estimation result estimated by the estimation unit 154 (step S107). Specifically, the output control unit 155 outputs the estimation result estimated by the estimation unit 154 to the display unit 13. Furthermore, dehydration can be fatal depending on the severity. Therefore, at construction sites in the summer when thermal sweating and work-related sweating are likely to occur, it is necessary to actively intake fluids (and salt) as part of heatstroke prevention. For this reason, the output control unit 155 references the address of a mobile phone or other device carried by the work site supervisor recorded in the address information recording unit 143 via the communication unit 11, and outputs the estimation result estimated by the estimation unit 154 to the mobile phone carried by the work site supervisor. The output control unit 155 may also reference the address information recording unit 143 and output the estimation result estimated by the estimation unit 154 to the address or contact information of a mobile phone carried by a person who matches the feature values ​​of the person appearing in the image detected by the detection unit 152.

[0036] Next, the learning unit 156 re-learns the water intake trained model based on the estimation result estimated by the estimation unit 154, the amount of change in the person's body part calculated by the calculation unit 153, the container size, the time until returning to the initial position, and the water intake trained model (step S108). After step S108, the water intake estimation device 10 ends this process.

[0037] According to the embodiment described above, the estimation unit 154 estimates the amount of water intake that a person has drunk based on the amount of change in the person's body part calculated by the calculation unit 153, the size of the container, and the time it takes for the container to return to its initial position. Therefore, the amount of water intake can be estimated even when an unspecified number of people use the device, without imposing restrictions on individual behavior.

[0038] Furthermore, according to one embodiment, in order to prevent heatstroke among multiple workers working at a construction site, the amount of water intake of the workers can be automatically estimated without imposing restrictions on the workers' behavior.

[0039] Furthermore, according to one embodiment, the output control unit 155 refers to the features of the person recorded by the address information recording unit 143, and outputs the estimation result estimated by the estimation unit 154 to the mobile phone carried by the person whose features match those of the person appearing in the image detected by the detection unit 152, thereby making it possible to estimate the amount of water intake for each individual.

[0040] Furthermore, according to one embodiment, the output control unit 155 refers to the address of a mobile phone or the like carried by the supervisor of the work site, which is recorded in the address information recording unit 143, via the communication unit 11, and outputs the estimation result estimated by the estimation unit 154 to the mobile phone carried by the supervisor of the work site. This can contribute to the prevention of dehydration in the person being estimated by the specified supervisor.

[0041] While the embodiment has been described with respect to a single camera 100, the present invention is not limited to this configuration and images may be acquired from multiple cameras. For example, at a construction site, multiple cameras or video cameras may be installed to observe a wide area. In this case, the water intake estimation device 10 may estimate the water intake of each person appearing in each image from the images acquired from each camera, and if the person is identified, estimate the cumulative water intake of each person.

[0042] Various inventions can be formed by appropriately combining the multiple components disclosed in the water intake estimation device according to the embodiment of the present disclosure described above. For example, some components may be omitted from all the components described in the water intake estimation device according to the embodiment of the present disclosure. Furthermore, the components described in the information provision system according to the embodiment of the present disclosure described above may be appropriately combined.

[0043] Furthermore, in the water intake estimation device according to an embodiment of the present disclosure, the "unit" described above can be read as "means" or "circuit," etc. For example, the "control unit" can be read as a control means or a control circuit.

[0044] In addition, the program to be executed by the water intake estimation device according to one embodiment of the present disclosure is provided as file data in an installable or executable format recorded on a computer-readable recording medium such as a CD-ROM, flexible disk (FD), CD-R, DVD (Digital Versatile Disk), USB medium, or flash memory.

[0045] In addition, the program to be executed by the water intake estimation device according to one embodiment of the present disclosure may be configured to be stored on a computer connected to a network such as the Internet and provided by being downloaded via the network.

[0046] In the explanation of the flowcharts in this specification, the order of processing between steps is clearly indicated using expressions such as "first," "then," and "continue," but the order of processing required to implement the present invention is not uniquely determined by these expressions. In other words, the order of processing in the flowcharts described in this specification can be changed within a consistent range.

[0047] Although some of the embodiments of the present application have been described in detail above with reference to the drawings, these are merely examples, and the present invention can be implemented in other forms that have undergone various modifications and improvements based on the knowledge of those skilled in the art, including the aspects described in the disclosure of the present invention. [Explanation of symbols]

[0048] 10. Water intake estimation device 11 Communications Department 12 Input section 13 Display section 14 Recording section 15 Control Unit 100 cameras 141 Program Recording Section 142 Trained Model Recording Unit 143 Address information recording section 151 Acquisition Department 152 Detection unit 153 Calculation Unit 154 Estimation Department 155 Output control section 156 Learning Department

Claims

1. A water intake estimation program to be executed by a water intake estimation device, an acquisition step of acquiring a group of time-sequential images of one or more people drinking water from a camera arranged at a predetermined location; a detecting step of detecting at least the person, the position of a body part of the person, and the size of a container from the group of images; a calculation step of calculating a change amount in the position of the person's body part detected in the detection step and a time period from when the person's body part changes from its initial position until when the person's body part returns to its initial position; an estimation step of estimating a water intake amount of the person based on the amount of change, the container size, and the time; Execute Water intake estimation program.

2. 2. The water intake estimation program according to claim 1, an output step of outputting the water intake amount estimated in the estimation step to a predetermined device; Further execute Water intake estimation program.

3. 3. The water intake estimation program according to claim 2, The detecting step Further detecting the feature amount of the person; The output step includes: referencing a representative device at the work site where the person is located or an address information recording unit that associates the feature amount of the person with the address of a device carried by the person, and outputting the water intake amount estimated in the estimation step to a device corresponding to the feature amount of the person detected in the detection step; Water intake estimation program.

4. an acquisition unit that acquires a group of time-sequential images of one or more people drinking water from a camera disposed at a predetermined location; a detection unit that detects at least the person, the position of a body part of the person, and the size of a container from the group of images; a calculation unit that calculates the amount of change in the position of the person's body part detected by the detection unit and the time it takes for the person's body part to change from its initial position and return to its initial position; an estimation unit that estimates the amount of water intake that the person has drunk based on the amount of change, the container size, and the time; Equipped with Water intake estimation device.

5. A water intake estimation method executed by a water intake estimation device, comprising: an acquisition step of acquiring a group of time-sequential images of one or more people drinking water from a camera arranged at a predetermined location; a detecting step of detecting at least the person, the position of a body part of the person, and the size of a container from the group of images; a calculation step of calculating a change amount in the position of the person's body part detected in the detection step and a time period from when the person's body part changes from its initial position until when the person's body part returns to its initial position; an estimation step of estimating a water intake amount of the person based on the amount of change, the container size, and the time; Including, Methods for estimating fluid intake.

Images