CN116434497A - Outdoor exercise temperature loss early warning method, system and medium - Google Patents

Abstract

An outdoor exercises losing temperature early warning method comprises the following steps: acquiring monitoring data, wherein the monitoring data comprises human body state data of monitored personnel, outdoor activity plan data of the monitored personnel, environment state data of an area where the monitored personnel is located, weather forecast data of the area where the monitored personnel is located, human body state data of reference personnel in the same area and positioning information of the area where the monitored personnel is located; based on the processing of the monitoring data by the temperature-losing danger prediction model, predicting the danger probability of the temperature-losing danger of the monitored personnel in the future target time; the temperature loss danger prediction model is a machine learning model; and determining prompt information based on the risk probability, and sending an early warning prompt to a corresponding user terminal.

Description

A method, system and medium for early warning of hypothermia in outdoor sports

technical field

The invention relates to the field of smart wearable devices, in particular to a method, system and medium for early warning of hypothermia during outdoor sports.

Background technique

As people's enthusiasm for outdoor sports has been increasing in recent years, more and more people have participated in outdoor activities such as outdoor hiking, field adventures, jungle shuttles, and field marathons. But there is a big difference between outdoor activities and ordinary leisure sports. The outdoor terrain is complex and the environment is changeable, and severe natural weather may be encountered at any time. Improper handling may result in personal injury or even death directly. There are great requirements for personnel's outdoor knowledge reserve and practical experience.

Body temperature, and more critically the body's core temperature, is a key indicator of safely performing outdoor activities. Preserving body temperature, avoiding a large loss of body temperature, and replenishing and recovering in time after a large amount of body temperature loss is a principle that outdoor personnel should keep in mind. However, news of people losing their lives outdoors due to hypothermia is often reported.

Generally speaking, hypothermia refers to a disease in which the heat loss of the human body is greater than the heat supply, resulting in a decrease in the temperature of the core area of the human body, resulting in a series of symptoms such as shivering, confusion, cardiopulmonary failure, and eventually death. If the body temperature is lower than 35 degrees, people will lose consciousness or even die in a coma.

Among the accidents caused by hypothermia, some are caused by ignorance of the hazards of hypothermia, some are caused by sudden bad weather, but there are also experienced outdoor personnel who are careless or misjudged the trend of physical fitness changes. And there is a situation where a large number of teams in the team lose their temperature collectively. This is because hypothermia usually has two notable characteristics of "behindsight" and "too late".

1. "Lost awareness". When you are in a state of hypothermia at the beginning, due to the body temperature regulation mechanism, your body data will not change significantly, but you are already in danger of hypothermia. And even professional outdoor athletes cannot judge with absolute accuracy whether they are in a hypothermic state or how long they can stay in a hypothermic environment. When people realize that they are in a state of hypothermia, they are often already in a state of mild hypothermia. At this time, the body has begun to tremble and tremble, and the ability to move is greatly affected. If you cannot get out of the hypothermic environment, seek shelter, and recover your body temperature in time, the hypothermic state will be further aggravated until life-threatening.

Two, "it's too late." Personnel are unable to sustain hypothermia for extended periods of time. Especially when the personnel realize that they have lost temperature and are in the middle and late stages, they need to stay away from the environment as soon as possible. Once a person enters a state of hypothermia, their bodily functions are greatly weakened. If you do not get out of the hypothermic environment, seek shelter, and regain body temperature as soon as possible, the hypothermic state will further aggravate and become life-threatening. However, because the outdoor environment has the characteristics of changeable climate and extreme weather at any time, and the outdoor environment is far away from the city, it is difficult to find a warm and comfortable shelter, and personnel cannot improve the environmental conditions in time, which eventually leads to tragedy. The affected physical state will have a great influence on the implementation of these measures. At this time, whether it is self-rescue or calling for rescue, it is a race against time.

It can be seen that the best way to prevent the hazards of hypothermia is to predict the possibility of hypothermia in advance and remind personnel to take measures before hypothermia, such as taking insulation measures, changing or canceling action plans, etc., to avoid being in a state of hypothermia in a vicious circle.

Therefore, it is necessary to provide an outdoor sports hypothermia early warning method, system and medium that can predict the hypothermia risk in advance.

Contents of the invention

The purpose of the present invention is to provide a method, system and medium for early warning of hypothermia during outdoor sports. In order to realize a system that can issue an early warning of the risk of hypothermia before personnel enter an area with a risk of hypothermia, and remind personnel to take measures in advance to avoid danger.

In order to achieve the aforementioned object, the present invention adopts the following technical solutions:

A method for early warning of hypothermia in outdoor sports, comprising: acquiring monitoring data, the monitoring data including the human body state data of the monitored person, the outdoor activity plan data of the monitored person, and the environmental state data of the area where the monitored person is located , the weather forecast data of the area where the monitored person is located, the human body state data of the reference personnel in the same area, and the positioning information of the area where the monitored person is located; Describe the danger probability of hypothermia danger of the monitored person within the future target time; the hypothermia risk prediction model is a machine learning model; based on the danger probability, determine the prompt information, and send an early warning prompt to the corresponding user terminal.

In some embodiments, the human body state data includes blood oxygen concentration data, heartbeat frequency data, temperature data of the core area of the human body, and humidity data of the core area of the human body.

In some embodiments, the environmental status data includes environmental temperature data, humidity data, air pressure data, and air velocity; the outdoor activity planning data includes outdoor activity time, outdoor activity location, and outdoor activity route.

In some embodiments, the prompt information includes: at least one of adding clothes in advance, changing the course of action, finding a shelter in advance, and canceling outdoor activities.

At the same time, the present invention also discloses an outdoor exercise hypothermia warning system, which includes: an acquisition module for acquiring monitoring data, the monitoring data includes the human body state data of the monitored person, and the outdoor activity plan data of the monitored person , the environmental state data of the area where the monitored person is located, the weather forecast data of the area where the monitored person is located, the human body state data of the reference personnel in the same area, and the positioning information of the area where the monitored person is located; the prediction module uses Based on the processing of the monitoring data based on the hypothermia risk prediction model, the risk probability of the hypothermia risk of the monitored personnel within the future target time is predicted; the hypothermia risk prediction model is a machine learning model; the early warning module uses Based on the risk probability, the prompt information is determined, and an early warning prompt is issued to a corresponding user terminal.

In some embodiments, the acquisition module includes a human body sensor measurement unit and an environmental sensor measurement unit, an interaction unit, a positioning unit, an external data acquisition unit, a central processing unit, and a communication unit;

The human body sensor measurement unit includes a blood oxygen concentration sensor for obtaining blood oxygen concentration data, a heartbeat frequency sensor for obtaining heartbeat frequency data, a temperature sensor for obtaining temperature data of the core region of the human body, and a temperature sensor for obtaining the temperature data of the human body core area. Humidity sensor for humidity data in the core area;

The environmental sensor measurement unit includes a temperature sensor for acquiring ambient temperature data, a humidity sensor for acquiring ambient humidity data, an air pressure sensor for acquiring ambient air pressure data, and an air flow rate for acquiring ambient air velocity sensor;

The interaction unit is used to obtain the outdoor activity plan data of the monitored person;

The positioning unit is used to obtain the positioning information of the area where the monitored person is located;

The external data acquisition unit is used for the weather forecast data of the area where the monitored person is located, and the human body state data of the reference person in the same area;

The central processing unit is used to summarize the monitoring data;

The communication unit is used to transmit the monitoring data collected by the central processing unit to a prediction module.

In some embodiments, the hypothermia risk prediction model is an LSTM deep learning model, and the LSTM deep learning model is obtained based on model training;

The model training includes:

Obtain sample training data, the sample training data includes multiple sets of sample monitoring data and corresponding label data, the sample monitoring data includes sample human body state data of sample personnel, sample outdoor activity plan data of the sample personnel, the sample The sample environmental state data of the area where the person is located, the sample weather forecast data of the area where the sample person is located, the sample human body state data of the sample reference person in the same area, and the sample positioning information of the area where the sample person is located; the sample monitoring data is based on historical data acquisition;

The tag data is the actual hypothermia state of the sample personnel during the sample time period, and the value of the tag data is 1 if they are in the hypothermia state, otherwise, the value of the tag data is 0;

dividing the sample training data into a test set and a training set;

Training the initial hypothermia risk prediction model based on the training set; obtaining the trained initial hypothermia risk prediction model;

Test the trained initial hypothermia risk prediction model based on the test set, and judge whether the test result satisfies the preset condition, and if so, use the trained initial hypothermia risk prediction model as the hypothermia a risk prediction model; otherwise, repeatedly use the training set to train the initial hypothermia risk prediction model.

In some embodiments, the early warning module is further used for:

judging whether the risk probability exceeds a probability threshold,

In response to the danger probability exceeding the probability threshold, the corresponding prompt information is generated based on the danger probability, and the warning prompt is sent to a corresponding user terminal.

In some embodiments, the early warning module further includes a display screen and a speaker; the early warning prompt is issued based on the display screen and the speaker.

At the same time, the present invention also discloses an outdoor sports hypothermia warning device, which includes at least one processor and at least one memory; the at least one memory is used to store computer instructions; the at least one processor is used to execute the At least some of the instructions in the computer instructions are used to realize the aforementioned method for early warning of hypothermia during outdoor sports.

At the same time, the present invention also discloses a computer-readable storage medium, which stores computer instructions, and when the computer reads the computer instructions, the computer executes the above-mentioned outdoor exercise hypothermia warning method.

Beneficial effect

Compared with the prior art, the present invention has the remarkable advantages of:

(1) The present invention can predict the probability that a person will face the risk of hypothermia within a certain period of time in the future, rather than judging whether the person is currently in a state of hypothermia. Hypothermia symptoms have obvious physical symptoms, and mild hypothermia will cause trembling symptoms. Personnel are often belatedly aware of hypothermia, and when they feel that they are in a state of hypothermia, it is often too late, the mobility of personnel is limited, the state of hypothermia is aggravated, and even life-threatening. Although clothing with good thermal insulation effect can delay or delay the arrival of hypothermia, it cannot fundamentally solve the risk of hypothermia. Therefore, in the face of the danger of hypothermia, the present invention can predict the arrival of the danger of hypothermia in advance, and make corresponding treatment in advance, so as to avoid the occurrence of hypothermia hazard and prevent problems before they happen.

(2) The present invention can detect the data of a period of time during the movement of the person, so as to realize the prediction of hypothermia, instead of simply relying on the state of the person at the current moment. The detection of data during a period of time during the movement of personnel not only includes the status data of the personnel at the current point in time, but also includes the status change data of the personnel in the past period of time, which increases the dimensionality of the representation of the data on the status of personnel, thereby improving the prediction results accuracy. The status of personnel is a process of dynamic change, and it is difficult to predict the risk of hypothermia in the future from the current status alone. Adding the historical data of personnel status can obtain the change trend of personnel status and improve the accuracy of prediction.

(3) The present invention can make comprehensive use of personnel state change data, outdoor activity plan data, environmental state change data, weather forecast data, and other personnel state data in the surrounding area to jointly predict hypothermia, rather than solely relying on the personnel's own data, thereby improving the prediction the accuracy of the results. Personnel face the risk of hypothermia. In addition to factors such as their own physical condition and thermal insulation measures, it is also related to various and extremely severe weather. Adding environmental state change data and weather forecast data improves the accuracy of prediction. Most of the outdoor activities are multi-person actions. Individuals in the team will suffer from functional decline or even disability due to hypothermia, which will affect the mobility of the entire team and cause the symptoms of hypothermia to spread in the team; increase the status data of other personnel in the surrounding area to improve the accuracy of the prediction results.

Description of drawings

Fig. 1 is a schematic flow chart of a method for early warning of outdoor sports hypothermia involved in the present embodiment;

FIG. 2 is a schematic diagram of an outdoor sports hypothermia warning system according to this embodiment;

3 is a schematic diagram of an acquisition module of the outdoor sports hypothermia warning system involved in this embodiment;

FIG. 4 is a schematic diagram of a hypothermia risk prediction model involved in this embodiment.

Detailed ways

In order to make the purpose, technical solution and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present application, and are not intended to limit the present application.

On the contrary, this application covers any alternatives, modifications, equivalent methods and schemes within the spirit and scope of this application as defined by the claims. Further, in order to make the public have a better understanding of the application, some specific details are described in detail in the detailed description of the application below. The present application can be fully understood by those skilled in the art without the description of these detailed parts.

An intelligent planning and production scheduling system involved in the embodiment of the present application will be described in detail below with reference to FIGS. 1-3 . It should be noted that the following examples are only used to explain the present application, and do not constitute a limitation to the present application.

Example 1

As shown in FIG. 1 , a schematic flowchart of a method for early warning of hypothermia during outdoor sports, the process may be executed by the early warning system 200 for outdoor sports hypothermia, including:

Step 110, obtaining monitoring data, the monitoring data includes the human body state data of the monitored person, the outdoor activity plan data of the monitored person, the environmental status data of the area where the monitored person is located, and the area where the monitored person is located. The weather forecast data, the human body status data of the reference personnel in the same area, and the location information of the area where the monitored personnel are located.

In some embodiments, the human body state data includes blood oxygen concentration data, heartbeat frequency data, temperature data of the core area of the human body, and humidity data of the core area of the human body.

Step 120, based on the processing of the monitoring data by the hypothermia risk prediction model, predict the risk probability of the hypothermia risk of the monitored person within the future target time; the hypothermia risk prediction model is a machine learning model.

In some embodiments, the environmental status data includes environmental temperature data, humidity data, air pressure data, and air velocity; the outdoor activity planning data includes outdoor activity time, outdoor activity location, and outdoor activity route.

In some embodiments, based on the processing of the monitoring data based on the hypothermia risk prediction model, the danger probability of the hypothermia risk of the monitored person within the future target time (such as the next 2 hours) can be predicted. The danger probability can be a value between 0 and 1. The larger the value, the greater the probability that the monitored person will be in danger of hypothermia within the target time in the future. Therefore, the more advanced protection is required.

Step 130, based on the risk probability, determine prompt information, and issue an early warning prompt to a corresponding user terminal.

In some embodiments, the prompt information includes: at least one of adding clothes in advance, changing the course of action, finding a shelter in advance, and canceling outdoor activities.

In some embodiments, the probability threshold (such as 0.6) can be preset in advance. Based on the comparison between the probability value output by the model and the probability threshold, it can be judged whether an early warning is required and what kind of early warning is required. For example, the probability value output by the model is 0.4, The prompt information can be adding clothes in advance, and if the probability value output by the model is 0.9, the prompt information can be cancellation of outdoor activities, etc.

Example 2

As shown in Figure 2, an outdoor exercise hypothermia warning system 200 includes:

The obtaining module 210 is used to obtain monitoring data, the monitoring data includes the human body state data of the monitored person, the outdoor activity plan data of the monitored person, the environmental state data of the area where the monitored person is located, the monitored The weather forecast data of the area where the person is located, the human body status data of the reference person in the same area, and the positioning information of the area where the monitored person is located;

The prediction module 220 is used to process the monitoring data based on the hypothermia risk prediction model, and predict the risk probability of the hypothermia risk of the monitored person within the future target time; the hypothermia risk prediction model is a machine learning model ;

The early warning module 230 is configured to determine prompt information based on the danger probability, and send a warning prompt to a corresponding user terminal.

In some embodiments, as shown in FIG. 3 , the acquisition module 210 includes a human body sensor measurement unit 211 and an environment sensor measurement unit 212, an interaction unit 213, a positioning unit 214, an external data acquisition unit 215, a central processing unit 216, a communication Unit 217.

The human body sensor measurement unit includes a blood oxygen concentration sensor for obtaining blood oxygen concentration data, a heartbeat frequency sensor for obtaining heartbeat frequency data, a temperature sensor for obtaining temperature data of the core region of the human body, and a temperature sensor for obtaining the temperature data of the human body core area. Humidity sensor for humidity data in the core area.

For example, the human body sensor measurement module can be worn on the core area, back, and limbs of the human body to collect temperature and humidity data in the core human body area; the temperature sensor is connected to the central processing unit, and the collected temperature data is transmitted to the central processing unit; the humidity sensor is connected to the central processing unit, and transmits the collected humidity data to the central processing unit.

The environmental sensor measurement unit includes a temperature sensor for acquiring ambient temperature data, a humidity sensor for acquiring ambient humidity data, an air pressure sensor for acquiring ambient air pressure data, and an air flow rate for acquiring ambient air velocity sensor.

The environmental sensor measuring unit can be set on the outside of the clothes, the outside of the backpack, etc., to collect changes in the temperature, humidity, atmospheric pressure, and air velocity of the environment where the person is moving; the temperature sensor is connected to the central processing unit , and transmit the collected temperature data to the central processing unit; the humidity sensor is connected to the central processing unit, and transmits the collected humidity data to the central processing unit; the air pressure sensor is connected to the central processing unit The central processing unit is connected, and the collected air pressure data is transmitted to the central processing unit; the air velocity sensor is connected to the central processing unit, and the collected air velocity data is transmitted to the central processing unit device.

The interaction unit is used to acquire the outdoor activity plan data of the monitored person. It may be an interactive device such as a button. The button is connected to the central processing unit to realize the switch function of the human body sensor measurement module, and to realize the frequency binding operation between the Bluetooth data transceiver and the Bluetooth data transceiver in the central control module.

The positioning unit is used to acquire positioning information of the area where the monitored person is located. The external data acquisition unit is used for the weather forecast data of the area where the monitored person is located, and the human body state data of the reference person in the same area. The central processing unit is used for summarizing the monitoring data. The communication unit is used to transmit the monitoring data collected by the central processing unit to a prediction module.

The communication unit can be a bluetooth data transceiver, and correspondingly, a corresponding bluetooth data transceiver can be set in the prediction module, and the bluetooth data transceiver of the acquisition module is connected with the central processing unit for converting the measured temperature . Humidity data is transmitted to the Bluetooth data transceiver in the prediction module by using Bluetooth data.

In some embodiments, the acquisition module can also set a power supply unit to supply power to each unit of the acquisition module, such as a battery can be added, and the battery can communicate with the temperature sensor, the humidity sensor, the central processing unit, and the Bluetooth data. connected to provide electrical energy for the operation of each component.

In some embodiments, the above-mentioned key can also realize the switch function of the environmental sensor measurement module, and realize the frequency binding operation between the Bluetooth data transceiver and the Bluetooth data transceiver in the central control module.

In some more specific embodiments, the outdoor exercise hypothermia warning system 200 includes a temperature sensor, a humidity sensor, an air pressure sensor, a blood oxygen concentration sensor, a heart rate sensor, an air velocity sensor, a Bluetooth data transceiver, a central processing unit, a display screen , speaker, button, mobile data signal transceiver, satellite data signal receiver, positioning module, battery, server.

There are multiple aforementioned temperature sensors, aforementioned humidity sensors, aforementioned barometric pressure sensors, aforementioned blood oxygen concentration sensors, aforementioned Bluetooth data transceivers, aforementioned batteries, and aforementioned buttons;

The aforementioned blood oxygen concentration sensor, the aforementioned heartbeat frequency sensor, the aforementioned Bluetooth data transceiver, the aforementioned central processing unit, the aforementioned display screen, the aforementioned speaker, the aforementioned buttons, the aforementioned mobile data signal transceiver, the aforementioned satellite data signal receiver, the aforementioned positioning module, The aforementioned battery constitutes a central control module; the aforementioned battery supplies power to the module, and the aforementioned buttons are used to control the switch state of the module, as well as perform functions such as parameter setting, device binding, data viewing, and message confirmation of the module.

The aforementioned temperature sensor, the aforementioned humidity sensor, the aforementioned Bluetooth data transceiver, the aforementioned key, the aforementioned battery, and the aforementioned key are combined into a human body sensor measurement module for obtaining human body temperature and humidity data; the aforementioned battery supplies power to the module, and the aforementioned key is used to control The switching state of the module, and controlling the device binding between the module and the aforementioned central control module.

The aforementioned central control module is worn on the user's wrist, and uses the aforementioned blood oxygen concentration sensor and the aforementioned heartbeat frequency sensor to measure the user's blood oxygen concentration and heartbeat frequency data.

The aforementioned human body sensor measurement module has structures such as straps, tethers, buckles, etc., which are convenient for users to wear on the limbs, core areas, back, etc., to collect changes in temperature and humidity of the human body during exercise, and pass The bluetooth data transceiver sends data to the aforementioned central control module. Multiple sensor modules are spread over the main areas of the body, which can well monitor the body state and obtain timely and accurate body state data.

The aforementioned temperature sensor, the aforementioned humidity sensor, the aforementioned air pressure sensor, the aforementioned air velocity sensor, the aforementioned Bluetooth data transceiver, the aforementioned button, the aforementioned battery, and the aforementioned buttons are combined into an environmental sensor measurement module for obtaining ambient temperature, humidity, air pressure, and air velocity Data; the aforementioned battery supplies power to the module, and the aforementioned button is used to control the switch state of the module, and to control the module to bind with the aforementioned central control module.

The aforementioned environmental sensor measurement module has straps, tethers, buckles and other structures, which are convenient for users to wear on the outside of clothing, backpacks, etc., and collect the temperature, humidity, atmospheric pressure, air The change of the flow rate, and send the data to the aforementioned central control module through the aforementioned Bluetooth data transceiver; the state of the environment greatly affects the loss of human body temperature, and at the same time, sudden changes in the weather can also be monitored in advance through this module. The prediction of hypothermia provides strong data support.

The aforementioned central control module receives the data of a plurality of aforementioned human body sensor measurement modules and the aforementioned environmental sensor measurement modules through the aforementioned Bluetooth data transceiver; the aforementioned central control module measures the human body blood oxygen concentration data and Heart rate data.

The aforementioned central control module obtains the current location through the aforementioned positioning module, and uploads to the aforementioned server through the aforementioned mobile data signal transceiver and the aforementioned satellite data signal receiver; The device receives the weather forecast data of the current area and the physical condition data of the people around the current area from the aforementioned server through the medium of the mobile data signal and the satellite data signal. When actually doing outdoor sports, it is usually in virgin forests and deserts. This area is far away from the city, and mobile signal base stations are rarely built. Therefore, the aforementioned mobile data signal transceiver and the aforementioned satellite data signal receiver are used together to avoid people from entering without moving. The problem that the data from the aforementioned server cannot be received in an outdoor environment covered by data signals.

The weather forecast data transmitted by the aforementioned server can observe the trend of weather changes in a larger space range, and can greatly predict the arrival of extreme severe weather in the future, providing good data support for the prediction of hypothermia. At the same time, pay attention to the physical condition of the members of the same team in the area, as well as the physical condition of other teams in the area. The physical condition of other teams around, predict and deal with the danger in advance.

The above-mentioned display screen and the above-mentioned buttons realize the input function of outdoor activity plan data such as the time, place, and route of the personnel's outdoor activities, and transmit the data to the aforementioned central processing unit;

The aforementioned central control module combines the data of the aforementioned modules and the entered outdoor activity plan data, and through the aforementioned central processing unit, uses a deep learning algorithm to obtain the probability of facing the risk of hypothermia in the future, and through the aforementioned display screen and the aforementioned loudspeaker, in the form of images and sounds, prompting the user to deal with the danger in advance;

During the operation of the system, the above-mentioned server will continuously receive the actual data of the user. After sorting out, it will continue to train the above-mentioned LSTM deep learning model, optimize the model prediction results, and transmit it to the user through the system software upgrade. In the outdoor sports hypothermia early warning system; so as to achieve the effect of continuous upgrading and improvement of the prediction ability of the aforementioned outdoor sports hypothermia early warning system;

The aforementioned server is deployed in the cloud and can provide weather forecast data and physical status data of people in the area for the aforementioned outdoor sports hypothermia warning system, and can collect user data uploaded by the aforementioned outdoor sports hypothermia warning system.

As shown in FIG. 4 , in some embodiments, the prediction module can predict the risk probability 430 of hypothermia risk of the monitored person within the future target time based on the processing of the monitoring data 410 by the hypothermia risk prediction model 420 .

In some embodiments, the hypothermia risk prediction model is an LSTM deep learning model, and the LSTM deep learning model is obtained based on model training;

The model training includes:

Obtain sample training data, the sample training data includes multiple sets of sample monitoring data and corresponding label data, the sample monitoring data includes sample human body state data of sample personnel, sample outdoor activity plan data of the sample personnel, the sample The sample environmental state data of the area where the person is located, the sample weather forecast data of the area where the sample person is located, the sample human body state data of the sample reference person in the same area, and the sample positioning information of the area where the sample person is located; the sample monitoring data is based on historical data acquisition;

The tag data is the actual hypothermia state of the sample personnel during the sample time period, and the value of the tag data is 1 if they are in the hypothermia state, otherwise, the value of the tag data is 0;

dividing the sample training data into a test set and a training set;

Training the initial hypothermia risk prediction model based on the training set; obtaining the trained initial hypothermia risk prediction model;

Test the trained initial hypothermia risk prediction model based on the test set, and judge whether the test result satisfies the preset condition, and if so, use the trained initial hypothermia risk prediction model as the hypothermia a risk prediction model; otherwise, repeatedly use the training set to train the initial hypothermia risk prediction model.

For example, the steps of model training are as follows:

S1. Collect human body status data, outdoor activity planning data, environmental status data, weather forecast data, other personnel status data around the same area, and mark data corresponding to the data whether they are in a state of hypothermia in multiple time ranges in the future.

Specifically, the above-mentioned human body state data include: blood oxygen concentration data, heartbeat frequency data, temperature data and humidity data of the core area of the person's body; the above-mentioned environmental state data include: temperature data, humidity data, and air pressure data of the person's environment , air velocity data; a sample piece of data in the original data is as follows:

The "h_temperature" data item is the temperature data of the core area of the person's body, in degrees Celsius (°C);

The "h_humidity" data item is the humidity data of the core area of the person's body, and the unit is the percentage relative humidity (%rh);

The "h_blood_oxygen_level" data item is the blood oxygen concentration data of the personnel, and the unit is percentage (%);

The "h_heart_rate" data item is the heartbeat frequency data of the person, and the unit is times/minute (bpm);

Among them, the "e_temperature" data item is the temperature data of the environment where the personnel are located, and the unit is Celsius (°C);

The "e_humidity" data item is the humidity data of the environment where the personnel live, and the unit is the percentage relative humidity (%rh);

The "e_atmospheric_pressure" data item is the air pressure data of the environment in which the person lives, and the unit is kilopascal (KPa);

Among them, the "e_air_velocity" data item is the air velocity data of the environment where the personnel are located, and the unit is minute per second (m/s);

The "weather_report" data item is the name of the main meteorological conditions that will occur in the future in the weather forecast data;

The "weather_report_time" data item is the time between the appearance of future meteorological conditions in the weather forecast data and the present time, in minutes (m);

The "area_people_status" data item is the physical status data of other people in the area where the person is located; among them, "good", "normal", "bad", and "worst" respectively represent four levels: good, general, poor, and extremely poor. Numbers represent the number of people at each level;

Among them, the "hypothermia_time" data item is the marked data of whether the person is facing the danger of hypothermia in the future; if there is no danger of hypothermia, the value of this item is 0, and if there is a danger of hypothermia, this value is the danger distance The current time in minutes (m).

S2. Perform vectorization processing on the data.

Specifically, the above sample data is vectorized and processed as follows:

[35.7,80,92,110,8,72,91,9.3,7,75,5,7,3,0,1,0,0,0,0,0]

Among them, the first 14 items of data are the temperature data of the core area of the person's body, the humidity data of the core area of the person's body, the blood oxygen concentration data of the person, the heartbeat frequency data of the person, the temperature data of the environment where the person lives, and the humidity of the environment where the person lives data, the air pressure data of the environment where the personnel live, the air velocity data of the environment where the personnel live, the numerical labels corresponding to the names of the main meteorological conditions that will occur in the future in the weather forecast data, the time from the present weather conditions in the weather forecast data to the future meteorological conditions, and the location of the personnel Data on the number of people in good physical condition in the area, data on the number of people in the area where the person is in general physical condition, data on the number of people in the area where the person is in poor physical condition, data on the number of people in the area where the person is in extremely poor physical condition;

Among them, the data of the last 6 items are the danger of hypothermia after 0-30 minutes, the danger of hypothermia after 30-60 minutes, the danger of hypothermia after 60-90 minutes, the danger of hypothermia after 90-120 minutes, and the danger of hypothermia after 120-150 minutes. The label data of the risk of hypothermia after 150-180 minutes; the risk of hypothermia occurred after 21 minutes in the original data, so the data item of the risk of hypothermia after 0-30 minutes is 1, and the other data Item is 0.

S3. Normalize the data.

Specifically, the above-mentioned vectorized sample data is normalized as follows:

[0.12,0.93,0.31,0.91,0.26,0.67,0.91,0.48,0.7,0.33,0.33,0.47,0.2,0,1,0,0,0,0,0]

The data of the first 10 items is consistent with the data meaning of the above-mentioned vectorized sample data; the normalization of the first 10 items of data is calculated according to the following formula:

Where X _norm is the normalized data, X is the original data of the data, X _max and X _min are the maximum and minimum values of the data in the original data set respectively;

Among them, the meanings of items 11 to 14 become: the proportion of the number of people in good physical condition in the area where the person is located to the total number of people in the area, the proportion of the number of people in good physical condition in the area where the person is located in the total number of people in the area, the physical condition of the area where the person is located The ratio of the number of poor people to the total number of people in the area, the ratio of the number of people in extremely poor physical condition to the total number of people in the area; the normalization of items 11 to 14 is calculated according to the following formula:

Among them, X _norm is the normalized data, X is the original data of the data, _Xi is the data of the number of people in good physical condition in the area where the personnel are located, the data of the number of people in general physical condition in the area where the personnel are located, and the area where the personnel are located. Data on the number of people in poor physical condition, and data on the number of people in extremely poor physical condition in the area where the person is located.

S4. Divide the data into training set data and test set data according to a certain ratio.

Specifically, for example: divide the data into training set data and test set data according to a certain ratio; divide the normalized data randomly into training set data and test set data according to the ratio of 3:1.

S5. Using the training set data to train the LSTM deep learning model.

S6. Use the test set data to test the LSTM deep learning model, but repeat the steps from S2 to S6 when the prediction effect of the model is not satisfactory.

In some embodiments, the model prediction phase includes the following steps:

S1. Collect current human body status data, outdoor activity plan data, environmental status data, weather forecast data, and other personnel status data around the same area;

S2. Use the same vectorization processing method and normalization processing method as the above-mentioned model training stage for processing;

S3. Input the above-mentioned LSTM deep learning model to obtain the probability of being in a hypothermia state within multiple time periods in the future;

S4. If the probability exceeds the set threshold, the above-mentioned display screen and the above-mentioned speaker will issue a warning.

In some embodiments, the early warning module is further used for:

It is judged whether the danger probability exceeds the probability threshold, and in response to the danger probability exceeding the probability threshold, the corresponding prompt information is generated based on the danger probability, and the warning prompt is sent to the corresponding user terminal.

In some embodiments, the early warning module further includes a display screen and a speaker; the early warning prompt is issued based on the display screen and the speaker.

At the same time, the present invention also discloses an outdoor sports hypothermia warning device, which includes at least one processor and at least one memory; the at least one memory is used to store computer instructions; the at least one processor is used to execute the At least some of the instructions in the computer instructions are used to realize the aforementioned method for early warning of hypothermia during outdoor sports.

At the same time, the present invention also discloses a computer-readable storage medium, which stores computer instructions, and when the computer reads the computer instructions, the computer executes the above-mentioned outdoor exercise hypothermia warning method.

In summary, the technical solution of the present invention utilizes the sensors installed close to the human body’s core area, back, and limbs to collect changes in temperature, humidity, heart rate, and blood oxygen concentration of the human body during exercise; Collect changes in temperature, humidity, atmospheric pressure, and air velocity of the environment during the movement of personnel; enter the time, location, route and other outdoor activity planning data of personnel's outdoor activities through the display screen and buttons; through mobile data signals, satellites, etc. Receive the current regional weather change forecast information issued by the server and the physical status information of other people around the area by means of data signals; collect the above data of a large number of people, use artificial intelligence algorithms to establish a hypothermia risk prediction model, and predict the temperature of people in the area. The probability of facing the risk of hypothermia in the future; when the user is using, collect the above data at the current moment of the user, input the above-mentioned hypothermia risk prediction model, and obtain the probability that the user will face the risk of hypothermia in the outdoor area for a period of time in the future Probability, prompting users to deal with the risk of hypothermia in advance, so as to solve the problem that people will face the risk of hypothermia and endanger their lives during outdoor sports.

The above descriptions are only preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present invention should be included in the protection of the present invention. within range.

Claims (10)

1. The outdoor exercises lose temperature early warning method is characterized by comprising the following steps:

acquiring monitoring data, wherein the monitoring data comprises human body state data of monitored personnel, outdoor activity plan data of the monitored personnel, environment state data of an area where the monitored personnel is located, weather forecast data of the area where the monitored personnel is located, human body state data of reference personnel in the same area and positioning information of the area where the monitored personnel is located;

based on the processing of the monitoring data by the temperature-losing danger prediction model, predicting the danger probability of the temperature-losing danger of the monitored personnel in the future target time; the temperature loss danger prediction model is a machine learning model;

and determining prompt information based on the risk probability, and sending an early warning prompt to a corresponding user terminal.

2. The outdoor exercises temperature loss pre-warning method according to claim 1, wherein the human body state data comprises blood oxygen concentration data, heartbeat frequency data, temperature data of a human body core area and humidity data of the human body core area.

3. The outdoor exercises temperature loss pre-warning method according to claim 1, wherein the environmental status data comprises environmental temperature data, humidity data, air pressure data and air flow rate; the outdoor activity planning data comprises the time of an outdoor activity, the place of the outdoor activity and the route of the outdoor activity.

4. The outdoor exercises temperature loss early warning method according to claim 1, wherein the prompting information comprises: at least one of adding clothes in advance, changing the action route, searching shelter in advance, and canceling outdoor activities.

5. An outdoor exercises loses warm early warning system, characterized in that includes:

the system comprises an acquisition module, a monitoring module and a control module, wherein the acquisition module is used for acquiring monitoring data, wherein the monitoring data comprises human body state data of monitored personnel, outdoor activity planning data of the monitored personnel, environment state data of an area where the monitored personnel is located, weather forecast data of the area where the monitored personnel is located, human body state data of reference personnel in the same area and positioning information of the area where the monitored personnel is located;

the prediction module is used for predicting the risk probability of the monitored personnel at the future target time based on the processing of the monitoring data by the temperature-losing risk prediction model; the temperature loss danger prediction model is a machine learning model;

and the early warning module is used for determining prompt information based on the risk probability and sending early warning prompts to the corresponding user terminals.

6. The outdoor exercises temperature loss warning system according to claim 5, characterized in that,

the acquisition module comprises a human body sensor measurement unit, an environment sensor measurement unit, an interaction unit, a positioning unit, an external data acquisition unit, a central processing unit and a communication unit;

the human body sensor measuring unit comprises an blood oxygen concentration sensor for acquiring blood oxygen concentration data, a heartbeat frequency sensor for acquiring heartbeat frequency data, a temperature sensor for acquiring temperature data of a human body core area and a humidity sensor for acquiring humidity data of the human body core area;

the environment sensor measuring unit comprises a temperature sensor for acquiring temperature data of the environment, a humidity sensor for acquiring humidity data of the environment, an air pressure sensor for acquiring air pressure data of the environment and an air flow rate sensor for acquiring air flow rate of the environment;

the interaction unit is used for acquiring outdoor activity plan data of the monitored personnel;

the positioning unit is used for acquiring positioning information of the area where the monitored personnel are located;

the external data acquisition unit is used for weather forecast data of the area where the monitored personnel are located and human body state data of reference personnel in the same area;

the central processing unit is used for summarizing the monitoring data;

the communication unit is used for transmitting the monitoring data summarized by the central processing unit to the prediction module.

7. The outdoor exercises temperature loss warning system according to claim 5, characterized in that,

the temperature loss danger prediction model is an LSTM deep learning model, and the LSTM deep learning model is obtained based on model training;

the model training comprises:

acquiring sample training data, wherein the sample training data comprises a plurality of groups of sample monitoring data and corresponding tag data, and the sample monitoring data comprises sample human body state data of a sample person, sample outdoor activity plan data of the sample person, sample environment state data of an area where the sample person is located, sample weather forecast data of an area where the sample person is located, sample human body state data of a sample reference person in the same area and sample positioning information of an area where the sample person is located; the sample monitoring data is acquired based on historical data;

the label data is in an actual temperature losing state of the sample personnel in a sample time period, the temperature losing state is that the value of the label data is 1, otherwise, the value of the label data is 0;

dividing the sample training data into a test set and a training set;

training an initial temperature-loss danger prediction model based on the training set; obtaining a trained initial temperature loss danger prediction model;

testing the trained initial temperature-loss risk prediction model based on the test set, judging whether a test result meets a preset condition, and if so, taking the trained initial temperature-loss risk prediction model as the temperature-loss risk prediction model; otherwise, the training set is used for training the initial temperature loss danger prediction model repeatedly.

8. The outdoor exercises temperature loss warning system of claim 5, wherein the warning module is further configured to:

determining whether the risk probability exceeds a probability threshold,

and responding to the dangerous probability exceeding the probability threshold, generating corresponding prompt information based on the dangerous probability, and sending the early warning prompt to a corresponding user terminal.

9. The outdoor exercises temperature loss warning system according to claim 5, characterized in that,

the early warning module also comprises a display screen and a loudspeaker; the early warning prompt is sent out based on the display screen and the loudspeaker.

10. A computer-readable storage medium storing computer instructions that, when read by a computer, perform the outdoor exercises temperature loss warning method of any one of claims 1 to 4.

Images