Heatstroke monitoring system

Abstract

We provide a heatstroke monitoring system that can properly monitor whether workers are suffering from heatstroke. [Solution] The system includes a monitoring computer 3 that stores a work plan including the date, time, and location where the work will be performed, and a monitoring drone 2 equipped with a camera 21 that flies to the location at the time the work will be performed based on the work plan and photographs the worker W at the location with the camera 21. The monitoring computer 3 determines whether the worker W is at risk of heatstroke based on the images taken by the camera 21, and if it determines that there is a risk of heatstroke, it issues a warning to the worker W.

Description

[Technical Field]

[0001] The present invention relates to a heat stroke monitoring system for monitoring whether a worker is suffering from heat stroke. [Background Art]

[0002] In recent years, intense heat has continued every year, and the number of casualties caused by heat stroke remains unceasing. For this reason, for example, a system is known in which a camera is installed at a venue used for events and the like, the health condition of visitors is analyzed based on image analysis of camera images, and an alarm is issued as soon as a person who is likely to suffer from poor physical condition such as heat stroke is found (see, for example, Patent Document 1). [Prior Art Documents] [Patent Documents]

[0003] [Patent Document 1] Japanese Unexamined Patent Application Publication No. 2019-219844 [Summary of the Invention] [Problem to be Solved by the Invention]

[0004] By the way, when working outdoors, the risk of heat stroke is particularly increased, so it is desirable to monitor the condition of workers. However, in a large construction site such as a power plant, the work location and the number of workers change every day, and people do not always gather in one place like in an event venue, so the system described in Patent Document 1 is difficult to properly monitor the condition of workers.

[0005] Accordingly, an object of the present invention is to provide a heat stroke monitoring system capable of properly monitoring whether a worker is suffering from heat stroke. [Means for Solving the Problem]

[0006] To solve the above problems, the invention of claim 1 is a heatstroke monitoring system characterized by comprising: a work plan storage means for storing a work plan including the date and time and location where work is to be performed; an unmanned aerial vehicle equipped with a photography means, which flies to the location at the date and time when work is to be performed based on the work plan, and photographs workers at the location with the photography means; a determination means for determining whether or not the worker is at risk of heatstroke based on the image taken by the photography means; and a measure for issuing a warning to the worker when the determination means determines that there is a risk of heatstroke.

[0007] The invention of claim 2 is characterized in that, in the heatstroke monitoring system described in claim 1, the unmanned aerial vehicle is equipped with a temperature sensor for measuring the body temperature of the worker, and the determination means determines whether or not the worker is at risk of heatstroke based on the image and the body temperature measured by the temperature sensor.

[0008] The invention of claim 3 is a heatstroke monitoring system according to claim 1, wherein the determination means determines whether or not the worker has responded to the warning based on the image captured by the imaging means, and the action means calls for help from people in the vicinity when the determination means determines that the worker has not responded to the warning.

[0009] The invention of claim 4 is characterized in that, in the heatstroke monitoring system described in claim 1, the action means transmits alarm information to a designated supervisor's mobile terminal when the determination means determines that there is a risk of heatstroke.

[0010] The invention of claim 5 is a heatstroke monitoring system according to claim 1, characterized in that the unmanned aerial vehicle periodically photographs the worker with the photographing means until the work is completed, and the determination means determines whether or not the worker is at risk of heatstroke each time the worker is photographed by the photographing means.

[0011] The invention of claim 6 is a heatstroke monitoring system according to claim 1, wherein the work plan includes the name and photograph of the worker in charge of the work, the determination means identifies the name of the worker at risk of heatstroke based on the work plan and the photograph, and the action means announces the name of the worker identified by the determination means when issuing a warning to the worker.

[0012] The invention of claim 7 is a heatstroke monitoring system according to claim 4, wherein the work plan includes the name and photograph of the worker in charge of the work, the determination means identifies the name of the worker at risk of heatstroke based on the work plan and the photograph, and the alarm information includes the name of the worker identified by the determination means.

[0013] The invention of claim 8 is characterized in that, in the heatstroke monitoring system described in claim 1, the determination means uses a determination learning model that has been machine-trained based on past performance data, so that when the image is input, it outputs whether or not the worker is at risk of heatstroke. [Effects of the Invention]

[0014] According to the invention described in claim 1, based on the work plan, an unmanned aerial vehicle flies to the work site at the time the work is to be performed, photographs the workers, and determines whether the workers are at risk of heatstroke. If there is a risk of heatstroke, the workers are alerted and heatstroke prevention measures are promoted. In this way, even if the work site changes from day to day, the unmanned aerial vehicle flies to the work site to monitor and determine whether the workers are at risk of heatstroke and alerts them, making it possible to properly monitor the occurrence of heatstroke even in large construction sites and ensure and support the safety of workers.

[0015] According to the invention described in claim 2, it becomes possible to properly determine whether a worker is at risk of heatstroke based not only on the worker's appearance but also on the worker's body temperature, using images of the worker. As a result, it becomes possible to properly monitor the onset of heatstroke and properly ensure and support the safety of workers.

[0016] According to the invention described in claim 3, if a worker does not respond to a warning, a call for help from those around them is made, thereby encouraging those around to rescue the unconscious worker and ensuring and supporting the worker's safety.

[0017] According to the invention described in claim 4, not only is a warning issued to the worker, but alarm information is also transmitted to the supervisor's mobile device, making it possible to prompt action and rescue for workers at risk of heatstroke, thereby ensuring and supporting the safety of the workers.

[0018] According to the invention described in claim 5, workers are photographed periodically until the work is completed to determine whether or not they are at risk of heatstroke. This makes it possible to properly monitor for the onset of heatstroke until the work is completed, thereby ensuring and supporting the safety of the workers.

[0019] According to the invention described in claim 6, when a worker at risk of heatstroke is warned, the worker's name is announced, thereby increasing the worker's attention and, as a result, ensuring and supporting the worker's safety.

[0020] According to the invention described in claim 7, alarm information including the name of a worker at risk of heatstroke is transmitted to the supervisor's mobile terminal, enabling appropriate and prompt response and rescue of the worker, thereby ensuring and supporting the worker's safety.

[0021] According to the invention of claim 8, since whether a worker is at risk of heat stroke is output using a machine-learned judgment learning model, it is possible to appropriately determine whether there is a risk of heat stroke, and appropriately ensure and support the safety of workers. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] [Figure 1] It is a schematic configuration diagram showing the heat stroke monitoring system according to an embodiment of the present invention. [Figure 2] It is a schematic configuration block diagram showing a monitoring computer of the heat stroke monitoring system of Fig. 1. [Figure 3] It is a diagram showing an example of a work plan described in a work plan database of the monitoring computer of Fig. 2. [Figure 4] It is a functional block diagram showing a schematic configuration of a judgment learning model of the monitoring computer of Fig. 2. DETAILED DESCRIPTION OF THE INVENTION

[0023] Hereinafter, the present invention will be described based on the illustrated embodiments.

[0024] Figs. 1 to 4 show an embodiment of the present invention, and Fig. 1 is a schematic configuration diagram showing a heat stroke monitoring system 1 according to this embodiment. This heat stroke monitoring system 1 is a system that monitors whether a worker W has contracted heat stroke. In this embodiment, the case where the work site is a power plant will be mainly described, but it goes without saying that the system may also be applied to other work sites.

[0025] The heat stroke monitoring system 1 mainly includes a monitoring drone (unmanned aerial vehicle) 2 that flies over a work site, and a monitoring computer (judgment means, measure means) 3 disposed in a management center C, wherein the monitoring drone 2 and the monitoring computer 3 are communicably connected to each other.

[0026] The surveillance drone 2 is an unmanned aerial vehicle that flies to the work site at the time and date the work is to be performed and takes photographs of the workers W at that site. In this embodiment, it flies automatically from a designated landing area within the power plant. Specifically, it is equipped with GPS (Global Positioning System) and, based on the work plan described later, receives flight commands from the surveillance computer 3, including location information (latitude, latitude) of the work site and the number of workers. It then autonomously controls its flight to the work site while confirming its own position using GPS. At this time, it is equipped with a function to avoid obstacles to flight depending on the flight environment.

[0027] This surveillance drone 2 is equipped with a camera (shooting means) 21 and a thermosensor (temperature sensor) 22. The camera 21 detects workers W, and photographs workers W (at least their faces) and measures their body temperature. That is, once it flies to the work area, it patrols and circles the area, sequentially detecting workers W, photographing them with the camera 21 and thermosensor 22, measuring their temperature, and transmitting the captured images, measured body temperature, and location information (latitude, longitude) of the shooting and temperature measurement locations, along with the identification information of the surveillance drone 2, to the monitoring computer 3 in real time. By shooting and measuring temperature while patrolling, workers W can be photographed and their temperature measured from various positions, heights, and directions, enabling proper monitoring.

[0028] In this way, all workers W at the work site are photographed and their temperatures are measured, and the images and body temperatures are transmitted to the monitoring computer 3. After all workers W have been photographed and their temperatures have been measured, and a return command is received from the monitoring computer 3 as described later, the aircraft flies to the designated landing area and lands.

[0029] Furthermore, the monitoring drone 2 is equipped with a speaker (measurement means) 23 and outputs a measure message sound corresponding to the type of measure received from the monitoring computer 3, as will be described later. That is, it is equipped with a sound source that has multiple types of measure message sounds stored in advance and outputs the measure message sound corresponding to the type of measure from the speaker 23. The types of measures and measure message sounds will be described later. In addition, it is equipped with text-to-speech conversion software and, as will be described later, when the name of worker W is transmitted from the monitoring computer 3 along with the type of measure, it outputs (speaks) the name of worker W as voice.

[0030] The monitoring computer 3 is a computer that determines whether or not worker W is at risk of heatstroke, and as shown in Figure 2, it mainly comprises an input unit 31, a display unit 32, a communication unit 33, a storage unit 34, a monitoring support task (determination means, action means) 35, a learning task 36, and a central processing unit 37 that controls these.

[0031] The input unit 31 is an interface for inputting various information and commands, specifically for inputting activation commands for the monitoring support task 35 and inputting work plans into the work plan database 341, which will be described later. The display unit 32 is a display for displaying various data and information, specifically for displaying judgment results determined by the monitoring support task 35 and images received from the monitoring drone 2. The communication unit 33 is an interface for communicating with the outside world, specifically for receiving images and temperature measurement results from the monitoring drone 2 and transmitting flight commands and action types to the monitoring drone 2. It also transmits alarm information to the mobile terminal K1 of a designated supervisor K, as will be described later.

[0032] The memory unit 34 stores images and temperature measurement results received from the monitoring drone 2 in chronological order for each worker W identified as described later, and also includes a work plan database (work plan storage means) 341, a judgment learning model 342, and a judgment performance database 343. Here, the work plan database 341 will be described, and the judgment learning model 342 and the judgment performance database 343 will be described later.

[0033] The work plan database 341 is a database that stores work plans, including the date, time, and location where work will be performed. For example, as shown in Figure 3, for a large-scale construction project S1, the work plan stores, for each task, the start date and time of the work and the planned end date and time of the work S2, the identification information and location information (latitude, longitude) of the work site S3, the department that manages and supervises the work and the name of the supervisor K (designated supervisor) for this work, contact information such as the telephone number and IP address of the mobile terminal K1 S4, the name of the construction company that will carry out the work and the number of workers S5. Furthermore, the names S6 and facial photographs S7 of each worker W in charge of this work are also stored.

[0034] In this embodiment, the work plan stores location information of the work site, but the following is also possible. That is, a map containing identification information and location information of each work site in the entire work site (the entire power plant) may be stored separately from the work plan, and location information may be obtained from the identification information of the work site stored in the work plan and the map. In addition, the name of the worker W and the facial photograph S7 may be stored in a separate database, and the facial photograph S7 may be obtained from this database.

[0035] The monitoring support task 35 is a task program that determines whether or not worker W is at risk of heatstroke at each work site. In this embodiment, it is activated for each task stored in the work plan at the start date and time of that task (or a predetermined time after that), but it may also be activated when an activation command and work site identification information are input to the input unit 31.

[0036] Upon activation, a flight command including the location information (latitude, latitude) of the work site and the number of workers stored in the work plan is first sent to the monitoring drone 2. As described above, the monitoring drone 2 is then flown to the work site to photograph and measure the temperature of worker W. If there are multiple tasks to be performed simultaneously, a flight command including the location information of the work site and the number of workers may be sent to multiple monitoring drones 2, or multiple flight commands may be sent to a single monitoring drone 2, allowing that single monitoring drone 2 to move between multiple work sites and photograph and measure temperature sequentially and alternately.

[0037] In this embodiment, the monitoring drone 2 is periodically flown to the work site until the work is completed to photograph and measure the temperature of the worker W. Specifically, as described later, a return command is sent to the monitoring drone 2 at a predetermined timing to land at a predetermined landing site. However, from the work start date and time stored in the work plan until the scheduled work end date and time (or when work completion information is received from an external source), a flight command is sent to the monitoring drone 2 at predetermined intervals (for example, every hour) to photograph and measure the temperature of the worker W. Alternatively, depending on the work environment and work content, the monitoring drone 2 may be kept flying and hovering above the work site until the scheduled work end date and time to periodically photograph and measure the temperature of the worker W.

[0038] Next, upon receiving images and body temperature data of worker W from the surveillance drone 2, the system first identifies the name of worker W, whose image and temperature were captured. Specifically, the name of worker W is determined based on the facial photograph stored in the work plan and the received image of worker W. Then, for each identified worker W, the received images and body temperature data are stored chronologically in the storage unit 34.

[0039] Next, based on the image of worker W received from the surveillance drone 2 and their body temperature, it is determined whether worker W is at risk of heatstroke. Specifically, it is determined whether worker W is at risk of heatstroke based on whether the image of worker W shows symptoms or movements characteristic of heatstroke (e.g., excessive sweating, frequently wiping away sweat, dizziness, etc.) and whether their body temperature is higher than normal. Here, "at risk of heatstroke" includes cases where the worker is already suffering from heatstroke and cases where they are at risk of developing heatstroke if left untreated. In making this determination, the heat index (WBGT) may be taken into consideration, or past images (e.g., from one hour ago) and body temperature stored in the memory unit 34 may be taken into consideration.

[0040] If it is determined that worker W is at risk of heatstroke, a warning is issued to worker W. Specifically, the type of action to be taken and the name of worker W are transmitted to the monitoring drone 2, and a warning message sound, such as "Mr. / Ms. XX, you are at risk of heatstroke, so please take a break and drink some water," is output from speaker 23. Furthermore, an alert information regarding the risk of heatstroke and the name of worker W are transmitted to supervisor K's mobile terminal K1, which is stored in the work plan.

[0041] When this action message sound is emitted, the monitoring drone 2 hovers in place for a predetermined time (for example, several minutes) to continue photographing and measuring the temperature of worker W, and transmits the captured images and temperature data to the monitoring computer 3. Based on the images of worker W received from the monitoring drone 2, it is determined whether or not worker W has responded to the warning. In other words, it is determined whether or not worker W has responded to the warning, that is, whether or not worker W is conscious, based on whether or not there are any gestures or movements in the image of worker W that indicate a reaction or response to the action message sound (for example, stopping work, drinking water, nodding, etc.).

[0042] If it is determined that worker W is not responding, the system alerts those nearby to seek help. Specifically, it transmits the type of action required to request rescue and the name of worker W to the monitoring drone 2, at which point a rescue message sound, such as "Mr. / Ms. XX appears to be suffering from heatstroke. Please rescue them," is output from speaker 23. In this case, the monitoring drone 2 circles the area for a predetermined time (for example, several minutes), emitting the rescue message sound. Furthermore, it transmits an alert requesting rescue information and the name of worker W to supervisor K's mobile terminal K1, which is stored in the work plan.

[0043] This process is performed for all workers W at the work site. In other words, the above judgment and measures are made for all worker W images and body temperatures received from the monitoring drone 2. If it is determined that there is no risk of heatstroke for any of the workers W, no measures are taken. After that, a return command is sent to the monitoring drone 2, and as described above, the monitoring drone 2 is flown to the designated landing area and landed. Subsequently, after a predetermined time (for example, 1 hour), a flight command is sent to the monitoring drone 2 again to photograph and measure the temperature of the workers W, and the same judgment and measures are made each time images and body temperatures of the workers W are received from the monitoring drone 2. This process is repeated until the scheduled work completion date and time stored in the work plan (or when work completion information is received from an external source).

[0044] This monitoring support task 35 uses a machine learning model 342 that has been trained based on past performance data to output whether or not worker W is at risk of heatstroke (hereinafter referred to as "presence or absence of heatstroke") when an image and body temperature of worker W are input. This machine learning model 342 for judgment is created by the learning task 36.

[0045] In other words, the learning task 36 creates a learning model 342 for judgment using a known machine learning algorithm such as a neural network, based on past performance data recorded and stored in the judgment performance database 343. This judgment performance database 343 is a database in which performance data, including whether or not heatstroke is present, is determined by experts and skilled personnel in heatstroke diagnosis and image analysis, based on the image and body temperature of worker W as input information. The past performance data includes data created based on the actual image and body temperature of worker W, as well as data created based on the actual determination of whether or not heatstroke is present by experts and skilled personnel, and data created through pre-training, etc.

[0046] As shown in Figure 4, this learning task 36 uses machine learning and deep learning with a neural network to create a neural network based on the actual data recorded in the judgment performance database 343. For example, it uses the image and body temperature of worker W as the input layer, the presence or absence of heatstroke as the output layer, and the analysis processing from the input layer to the output layer as the hidden layer. Then, the learning task 36 uses the actual data of the judgment learning model 342 as training data to learn various parameters in the hidden layer. In other words, the learning task 36 learns various parameters in the hidden layer so that the presence or absence of heatstroke is output appropriately based on the image and body temperature of worker W.

[0047] Furthermore, as described above, when determining whether worker W responded to a warning based on an image of worker W, a machine learning model based on past performance data may also be used so that when an image of worker W is input, it outputs whether or not worker W responded to the warning.

[0048] As described above, according to this heatstroke monitoring system 1, based on the work plan, a monitoring drone 2 flies to the work site at the time the work is to be performed, photographs the worker W, measures their temperature, and determines whether the worker W is at risk of heatstroke. If there is a risk of heatstroke, the worker W is alerted, and heatstroke prevention measures are promoted. In this way, even if the work location changes from day to day, the monitoring drone 2 flies to the work site to monitor and determine whether the worker W is at risk of heatstroke and issues an alert, making it possible to properly monitor the onset of heatstroke even in large construction sites and ensure and support the safety of the worker W.

[0049] Furthermore, by using images of worker W—that is, not only their appearance but also their body temperature—it becomes possible to accurately determine whether worker W is at risk of heatstroke. As a result, it becomes possible to properly monitor for the onset of heatstroke and ensure and support the safety of worker W.

[0050] Furthermore, if worker W does not respond to the warning, the system will alert those around him to seek help, encouraging them to rescue worker W who has lost consciousness, thereby ensuring and supporting worker W's safety.

[0051] Furthermore, not only is worker W alerted, but warning information is also sent to supervisor K's mobile device K1, enabling them to prompt action and rescue for worker W who is at risk of heatstroke, thereby ensuring and supporting worker W's safety.

[0052] Furthermore, by periodically photographing and measuring the temperature of worker W until the work is completed, it is possible to determine whether or not worker W is at risk of heatstroke. This allows for proper monitoring of the onset of heatstroke until the work is completed, ensuring and supporting the safety of worker W.

[0053] Furthermore, when a warning is issued to worker W who is at risk of heatstroke, the worker W's name is announced, which helps to increase worker W's attention, and as a result, makes it possible to ensure and support worker W's safety.

[0054] Similarly, since warning information including the name of worker W, who is at risk of heatstroke, is transmitted to supervisor K's mobile terminal K1, it becomes possible to respond to and rescue worker W appropriately and quickly, thereby ensuring and supporting worker W's safety.

[0055] On the other hand, using the machine learning-developed judgment model 342, it is possible to output whether or not worker W is at risk of heatstroke, thereby making it possible to properly determine whether or not there is a risk of heatstroke and to properly ensure and support the safety of worker W.

[0056] Although embodiments of this invention have been described in detail above, the specific configuration is not limited to these embodiments, and any design changes, etc., that do not depart from the gist of this invention are also included. For example, in the above embodiment, the monitoring computer 3 is equipped with a monitoring support task 35, but the monitoring drone 2 may be equipped with the monitoring support task 35, and the monitoring drone 2 may determine whether or not worker W is at risk of heatstroke. Also, although the monitoring drone 2 is equipped with a speaker 23, speakers may be placed around each work area, and action message sounds may be output from these speakers. Furthermore, if it is determined that worker W is at risk of heatstroke, a drone other than the monitoring drone 2 may be used to transport heatstroke prevention supplies such as oral rehydration solution to worker W, or a drone other than the monitoring drone 2 may be used to guide an emergency vehicle to worker W. [Explanation of symbols]

[0057] 1. Heatstroke monitoring system 2. Surveillance drones (unmanned aerial vehicles) 21. Camera (Method of taking pictures) 22. Thermosensor (temperature sensor) 23. Speaker (measure) 3. Surveillance computer 341 Work plan database (work plan storage means) 342 Classification Learning Models 35. Monitoring and support tasks (determination means, action means) W Worker K Supervisor K1 Mobile Device

Claims

1. A work plan storage means that stores a work plan including the date, time, and location where the work will be performed, An unmanned aerial vehicle equipped with a camera, which flies to the location at the date and time when the work is to be carried out according to the work plan, and photographs the workers at the location using the camera, A determination means that determines whether or not the worker is at risk of heatstroke based on the image captured by the aforementioned photographic means, If the determination means determines that there is a risk of heatstroke, measures are provided to alert the worker, A heatstroke monitoring system characterized by being equipped with the following features.

2. The unmanned aircraft is equipped with a temperature sensor to measure the body temperature of the worker. The determination means determines whether the worker is at risk of heatstroke based on the image and the body temperature measured by the temperature sensor. The heatstroke monitoring system according to feature 1.

3. The determination means determines, based on the image captured by the photographing means, whether or not the worker responded to the warning. The aforementioned means of action, when the determination means determines that the worker does not respond to the warning, will call for help from people nearby. The heatstroke monitoring system according to feature 1.

4. The aforementioned measures, when the determination means determines that there is a risk of heatstroke, transmit warning information to a designated supervisor's mobile terminal. The heatstroke monitoring system according to feature 1.

5. The unmanned aerial vehicle periodically photographs the worker with the photographic means until the work is completed. The determination means determines whether the worker is at risk of heatstroke each time they are photographed by the photography means. The heatstroke monitoring system according to feature 1.

6. The aforementioned work plan includes the names and photographs of the workers responsible for the aforementioned work. The determination means identifies the name of the worker at risk of heatstroke based on the work plan and the image. The aforementioned means, when alerting the worker, pronounces the name of the worker identified by the determination means. The heatstroke monitoring system according to feature 1.

7. The aforementioned work plan includes the names and photographs of the workers responsible for the aforementioned work. The determination means identifies the name of the worker at risk of heatstroke based on the work plan and the image. The alarm information includes the name of the worker identified by the determination means, The heatstroke monitoring system according to feature 4.

8. The determination means uses a machine learning model for determination, which is based on past performance data, so that when the image is input, it outputs whether or not the worker is at risk of heatstroke. The heatstroke monitoring system according to feature 1.

Images