Method for identifying daily unhealthy living habits

Abstract

The invention relates to the technical field of smart home and health management, in particular to a method for identifying daily unhealthy living habits, which comprises a data acquisition layer, a data processing and analysis layer and an application layer. The data acquisition layer comprises an intelligent socket, a human body infrared detector, a sleep sensor, a door magnetic opening and closing sensor, a hygrothermograph, a water sensor, an air quality detector, a closestool flushing detector, an SOS button and an intelligent night lamp; the intelligent socket, the human body infrared detector, the sleep sensor, the door magnetic opening and closing sensor, the hygrothermograph, the water sensor, the air quality detector, the closestool flushing detector, the SOS button and the intelligent night lamp are installed at the corresponding positions in a living room, a horizontal room, a kitchen, a dining room and a bathroom in a matched mode respectively. By fusing data of a plurality of Internet of Things sensors, a system capable of automatically identifying, analyzing and early warning specific unhealthy living habits is constructed, so that active management of user behavior health is realized.

Description

Technical Field

[0001] This invention relates to the technical field of smart home and health management, and in particular to a method for identifying unhealthy daily habits. Background Technology

[0002] With increasing health awareness, various health management systems and wearable devices are becoming more and more popular. However, existing technologies mainly focus on monitoring and alerting for physiological indicators (such as heart rate, blood pressure, blood oxygen, and steps). These belong to the "result" level of health problems. In fact, many chronic diseases (such as obesity, gastrointestinal diseases, and insomnia) are caused by long-term unhealthy habits (i.e., "triggers"). For example: Going to bed immediately after meals can easily cause gastroesophageal reflux, affecting digestion and sleep quality; Taking a shower immediately after a meal can increase blood flow to the skin and reduce blood flow to the digestive tract, thus affecting digestive function. Eating while watching TV can easily lead to unconscious overeating and distraction, affecting digestion. Frequent consumption of leftovers may lead to nutrient loss, bacterial growth, and irregular eating patterns.

[0003] Currently, the monitoring of these daily habits mainly relies on users' subjective records (such as health diary apps), which has problems such as dependence on user self-awareness, inaccurate data, and difficulty in long-term adherence. Although some smart home devices can record single behaviors (such as sleep and turning the TV on and off), the data is isolated and lacks the ability to correlate and logically analyze multiple behaviors over time, making it impossible to accurately identify the aforementioned complex behavioral patterns with potential health risks.

[0004] Therefore, there is an urgent need in this field for a technical solution that can identify users' unhealthy daily habits in a non-invasive, automated, and multi-dimensional manner. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides a method for identifying unhealthy daily habits by integrating data from multiple IoT sensors to construct a system capable of automatically identifying, analyzing, and issuing early warnings about specific unhealthy lifestyle habits, thereby enabling proactive management of user behavior and health.

[0006] The present invention provides a method for identifying unhealthy daily habits, comprising a data acquisition layer, a data processing and analysis layer, and an application layer; The data acquisition layer includes: smart socket, human infrared detector, sleep sensor, door magnetic opening and closing sensor, thermometer and hygrometer, water consumption sensor, air quality detector, toilet flushing detector, SOS button, smart night light and temperature sensor; Smart sockets, infrared human body detectors, sleep sensors, door magnetic opening and closing sensors, thermometers and hygrometers, water usage sensors, air quality detectors, toilet flushing detectors, SOS buttons, and smart night lights are installed in corresponding locations in the living room, bedroom, kitchen, dining room, and bathroom, respectively. Smart sockets are used to monitor the start and stop status of various electrical appliances in different spaces to determine the start and end times of dining and entertainment activities; Human infrared detectors are used to sense the activities and movements of people in the living room, bedroom, kitchen, dining room, and bathroom areas; Sleep sensors are used to monitor a person's bedtime, bedtime, and sleep status. Door magnetic opening and closing sensors are used to record the opening and closing events of refrigerator doors, windows, and bathroom doors, as well as their duration. Thermometers and hygrometers are used to monitor changes in temperature and humidity in the bathroom environment, serving as supplementary evidence for judging bathing behavior. Water sensors are used to detect water usage in bathrooms and kitchens, serving as supplementary evidence for determining whether cooking or washing is being done. Air quality detectors are used to monitor air quality in kitchens; A toilet flush detector is used to monitor the frequency of toilet flushing. The SOS button is used to trigger an alarm when the user feels unwell; The smart night light provides illumination to the head of the bed in a sleeping position and automatically turns off after the person falls asleep; The temperature sensor is used to detect the shower water temperature; The data processing and analysis layer is used to analyze and process the data collected by the data acquisition layer, identify simple behavioral events, and determine whether unhealthy lifestyle habits have been triggered. The application layer is used to generate targeted health improvement suggestions for unhealthy behaviors and to generate early warning signals to send to the user terminal.

[0007] Preferably, the data processing and analysis layer includes a data queue module, an IoT data-driven engine module, an IoT algorithm determination module, and a database module. Data queue module: Used to receive real-time data from various devices in the data acquisition layer, and to align and store the data with a unified timestamp; The IoT data-driven engine module receives data from the data queue module, performs noise reduction, missing value imputation, and feature extraction on the data, and identifies simple behavioral events based on preset logic. The IoT algorithm judgment module is used to associate the basic events identified by the IoT data-driven engine on the timeline, mine long-term habit patterns, and determine whether unhealthy lifestyle habits have been triggered according to preset habit judgment rules. It also counts the number of times a specific habit is triggered and the total duration within the statistical period. When the frequency or duration threshold is exceeded, it is determined that the unhealthy habit has been formed. DB module: Used to store the data after the data queue module is aligned, supports backtracking analysis, and stores the behavioral events identified by the IoT data driving engine and the habitual patterns identified by the IoT algorithm judgment module.

[0008] Preferably, the application layer includes a family elderly life model building engine module, an abnormal scene recognition and early warning module, and a user terminal; Family-based elderly living model building engine module: Based on identified unhealthy behaviors, it generates targeted health improvement suggestions; Abnormal Scene Recognition and Early Warning Module: When the system identifies that a user has frequent unhealthy habits, it generates an early warning signal and sends it to the user terminal. When the air quality detector detects a gas leak in the kitchen and the toilet flushing detector issues an alarm, it generates an alarm signal and actively sends the alarm information to the specific user terminal. User terminal: Used to receive warning signals from the abnormal scene identification and early warning module, and to display habit analysis reports and improvement suggestions to the user.

[0009] Preferably, the habit determination rules include the habit of going to bed shortly after meals, the habit of taking a short shower after meals, the habit of watching TV while eating, the habit of leaving leftovers, and the habit of drinking less water; Short bedtime after meals: When the smart socket detects that the kitchen appliances are running, the human infrared detector detects that a person is in the dining area, which is considered the start of the meal. When the human infrared detector detects that a person leaves the dining area, it is considered the end of the meal. If the sleep sensor detects the bedtime event again within the preset first time threshold T1 after the end of the meal is detected, it is considered that the habit has been repeated once. Short shower after meals: If, after the end of the meal is detected, within the preset second time threshold T2, the temperature and humidity meter detects that the temperature and humidity in the bathroom continue to rise and exceed the threshold, and at the same time the water sensor and temperature sensor detect shower water, then this habit is determined to have occurred once. Meal-watching TV habit: When the start of meal is detected and the TV is turned on via smart plug, if the start of meal and the TV turn-on event overlap in time and the overlap time exceeds the preset third time threshold T3, it is determined that the habit has been observed once. Leftovers habit: If the number of times the door magnetic sensor detects the refrigerator door opening event exceeds the preset threshold T4 in a single day, and the smart socket detects an association with the start-up event of a microwave oven or rice cooker, then this habit is determined to have occurred once. Low water intake habit: The number of times a user flushes the toilet in a single day is identified by a toilet flushing detector. If the number of flushes is less than a preset threshold T5, the user is determined to have this habit.

[0010] Preferably, it includes the following steps: S1: Collect multi-dimensional behavioral data of users through multiple IoT devices deployed in the environment; S2: Perform time alignment and preprocessing on the collected data; S3: Based on the preprocessed data, identify independent behavioral events related to lifestyle habits; S4: Link multiple independent behavioral events on a timeline and identify specific unhealthy lifestyle habits based on preset habit judgment rules; S5: Perform a statistical evaluation of the frequency and duration of the identified habits; S6: Based on the assessment results, generate and output health advice or early warning information.

[0011] Preferably, it also includes an IoT gateway; The IoT gateway receives data collected by the data acquisition layer, converts the data into a standard protocol, filters, compresses, and performs preliminary analysis on the raw data to reduce the transmission pressure on the cloud, and then sends the processed data to the data queue module.

[0012] Compared with the prior art, the beneficial effects of the present invention are: non-invasive and non-intrusive: monitoring is carried out using existing or deployed IoT devices in the environment, without requiring users to wear additional devices or actively record, resulting in a good user experience and easy long-term adherence; Multi-dimensional data fusion: By fusing data from different devices and dimensions, the problem of isolated data from a single sensor is solved, greatly improving the accuracy and reliability of behavior recognition; Contextualization and correlation analysis: Innovatively, different behaviors are logically linked on a timeline (e.g., "eating" is followed immediately by "sleeping"), which enables the identification of complex behavioral patterns with health risks, rather than just isolated events; Proactive early warning and personalization: The system can not only identify habits, but also provide personalized improvement suggestions based on data analysis, realizing a closed-loop health management from "monitoring" to "intervention", which is more practical. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of the structure of the horizontal smart socket of the present invention; Figure 2 This is a schematic diagram of the structure of the human infrared detector in the kitchen of the present invention; Figure 3 This is a schematic diagram of the structure of the water sensor in the bathroom of the present invention; Figure 4 This is a structural diagram of the IoT data-driven engine module and the IoT algorithm determination module; Figure 5 This is a schematic diagram of the steps of a method for identifying unhealthy daily habits according to the present invention.

[0014] The following labels are used in the attached diagram: 101, Smart Socket; 102, Human Infrared Detector; 103, Sleep Sensor; 104, Door Magnetic Opening / Closing Sensor; 105, Thermohygrometer; 106, Water Use Sensor; 107, Air Quality Detector; 108, Toilet Flushing Detector; 109, SOS Button; 110, Smart Night Light; 111, Temperature Sensor. Detailed Implementation

[0015] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. The present invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0016] Example 1 like Figures 1 to 5 As shown, a method for identifying unhealthy daily habits according to the present invention includes a data acquisition layer, a data processing and analysis layer, and an application layer. The data acquisition layer includes: a smart socket 101, a human infrared detector 102, a sleep sensor 103, a door magnetic opening and closing sensor 104, a thermometer and hygrometer 105, a water sensor 106, an air quality detector 107, a toilet flushing detector 108, an SOS button 109, a smart night light 110, and a temperature sensor 111. The smart socket 101, human infrared detector 102, sleep sensor 103, door magnetic opening and closing sensor 104, thermometer and hygrometer 105, water sensor 106, air quality detector 107, toilet flushing detector 108, SOS button 109 and smart night light 110 are respectively installed in the corresponding positions in the living room, bedroom, kitchen, dining room and bathroom. The smart socket 101 is used to monitor the start and stop status of various electrical appliances in different spaces to determine the start and end times of dining and entertainment activities; The human infrared detector 102 is used to detect the activities and movements of people in the living room, bedroom, kitchen, dining room and bathroom areas; The sleep sensor 103 is used to monitor the time a person goes to and from bed, as well as their sleep state. The door magnetic opening and closing sensor 104 is used to record the opening and closing events of the refrigerator door, window, and bathroom door, as well as their duration. The thermometer and hygrometer 105 are used to monitor changes in temperature and humidity in the bathroom environment as supplementary evidence for judging bathing behavior; The water sensor 106 is used to detect water usage in the bathroom and kitchen, serving as supplementary evidence for determining whether cooking or washing is being done. Air quality detector 107 is used to monitor air quality in the kitchen; Toilet flush detector 108 is used to monitor the frequency of toilet flushing; SOS button 109 is used to trigger an alarm when a person feels unwell; The smart night light 110 provides illumination for the head of the bed in a sleeping position and automatically turns off after the person falls asleep; Temperature sensor 111 is used to detect the shower water temperature; The data processing and analysis layer is used to analyze and process the data collected by the data acquisition layer, identify simple behavioral events, and determine whether unhealthy lifestyle habits have been triggered. The application layer is used to generate targeted health improvement suggestions for unhealthy behaviors and to generate early warning signals to send to the user terminal; The data processing and analysis layer includes a data queue module, an IoT data-driven engine module, an IoT algorithm determination module, and a database module. Data queue module: Used to receive real-time data from various devices in the data acquisition layer, and to align and store the data with a unified timestamp; The IoT data-driven engine module receives data from the data queue module, performs noise reduction, missing value imputation, and feature extraction on the data, and identifies simple behavioral events based on preset logic. The IoT algorithm judgment module is used to associate the basic events identified by the IoT data-driven engine on the timeline, mine long-term habit patterns, and determine whether unhealthy lifestyle habits have been triggered according to preset habit judgment rules. It also counts the number of times a specific habit is triggered and the total duration within the statistical period. When the frequency or duration threshold is exceeded, it is determined that the unhealthy habit has been formed. DB module: Used to store the data after the data queue module is aligned, supports backtracking analysis, and stores the behavioral events identified by the IoT data driving engine and the habitual patterns identified by the IoT algorithm judgment module. The application layer includes a family elderly life model building engine module, an abnormal scene recognition and early warning module, and a user terminal; Family-based elderly living model building engine module: Based on identified unhealthy behaviors, it generates targeted health improvement suggestions; Abnormal Scene Recognition and Early Warning Module: When the system recognizes that the user has frequent unhealthy habits, it generates an early warning signal and sends it to the user terminal. When the air quality detector 107 detects a gas leak in the kitchen and the toilet flush detector 108 issues an alarm, it generates an alarm signal and actively sends the alarm information to the specific user terminal. User terminal: Used to receive warning signals from the abnormal scene identification and early warning module, and to display habit analysis reports and improvement suggestions to the user; The rules for determining habits include the habit of going to bed shortly after meals, the habit of taking a short shower after meals, the habit of watching TV while eating, the habit of leaving leftovers, and the habit of drinking less water. Short bedtime habit after meals: After the smart socket 101 detects that the kitchen appliance is running, the human infrared detector 102 detects that a person is in the dining area, and then it is determined that the meal has started. When the human infrared detector 102 detects that a person has left the dining area, it is determined that the meal has ended. If the sleep sensor 103 detects the bedtime event again within the preset first time threshold T1 after the meal end event is detected, it is determined that the habit has been repeated once. Short shower after meals: If, after the end of the meal is detected, within the preset second time threshold T2, the temperature and humidity meter 105 detects that the temperature and humidity in the bathroom continue to rise and exceed the threshold, and at the same time the water sensor 106 and the temperature sensor 111 detect shower water, then this habit is determined to have occurred once. Meal-watching TV habit: When the start of meal is detected and the TV is turned on via smart plug 101, the meal start and TV turn-on events overlap in time and the overlap time exceeds the preset third time threshold T3, then it is determined that this habit has occurred once. Leftovers habit: If the number of times the door magnetic sensor 104 detects the refrigerator door opening event exceeds the preset threshold T4 within a single day, and the smart socket 101 detects an association with the start event of a microwave oven or rice cooker, then it is determined as one instance of this habit. Drinking less water habit: In a single day, the toilet flushing detector 108 identifies the number of times the user flushes the toilet. If the number of flushes is less than the preset threshold T5, it is determined that the user has this habit. In this embodiment, it is non-invasive and unobtrusive: monitoring is performed using existing or deployed IoT devices in the environment, without requiring users to wear additional devices or actively record data, resulting in a good user experience and easy long-term adherence; Multi-dimensional data fusion: By fusing data from different devices and dimensions, the problem of isolated data from a single sensor is solved, greatly improving the accuracy and reliability of behavior recognition; Scene-based and Relevance Analysis: Innovatively conduct logical association of different behaviors on the timeline (such as "going to bed" immediately after "eating"), so as to identify complex behavior patterns with health risks, rather than just isolated events; Active Warning and Personalization: The system can not only identify habits, but also provide personalized improvement suggestions based on data analysis, achieving a closed-loop health management from "monitoring" to "intervention", which has more practical value.

[0017] Embodiment 2 Based on Embodiment 1, a method for identifying daily unhealthy living habits of the present invention includes the following steps: S1: Collect multi-dimensional behavior data of users through multiple Internet of Things devices deployed in the environment; S2: Align the collected data in time and perform preprocessing; S3: Based on the preprocessed data, identify independent behavior events related to living habits; S4: Associate multiple independent behavior events on the time axis and identify specific unhealthy living habits according to preset habit determination rules; S5: Conduct statistical evaluation of the frequency and duration of the identified habits; S6: Generate and output health suggestions or warning information based on the evaluation results; It also includes an IOT gateway; The IOT gateway is used to receive the collected data of the data acquisition layer, uniformly convert the data into a standard protocol, and at the same time filter, compress and perform preliminary analysis on the original data to reduce the cloud transmission pressure, and then send the processed data to the data queue module; Embodiment 3 Based on Embodiment 1, the habit of going to bed shortly after a meal: The smart socket 101 detects that the kitchen range hood or rice cooker stops working at 18:30. Combining with the fact that the human infrared detector 102 detects the user in the dining area at this time, the system marks a potential "end of meal" event point A (timestamp T_A); The sleep sensor 103 detects that the user goes to bed at 19:15 and enters the ready-to-sleep state, and the system marks a "going to bed" event point B (timestamp T_B); The data processing and analysis layer calculates the time interval ΔT = T_B - T_A = 45 minutes; The first time threshold T1 preset by the system is 60 minutes. Since ΔT < T1, the system determines that this behavior triggers the habit of "going to bed shortly after a meal"; In the system's statistics for this week, this habit has been triggered 3 times. The system's preset weekly frequency threshold is 4 times. The system generates a prompt message: "You have had 3 instances this week where activities after a meal within an hour are likely to affect digestion. Please pay attention to improving this." and pushes it to the user via the mobile App.

[0018] Example 4 Based on Example 1, the habit of taking a bath shortly after a meal: The smart socket 101 detects that the kitchen range hood or rice cooker stops working at 07:00. Combining with the fact that the human infrared detector 102 detects the user in the dining area at this time, the system marks a potential "end of meal" event point C (timestamp T_C); The thermometer hygrometer 105, water usage sensor 106, and temperature sensor 111 detect that the user is taking a bath at 07:45. The system marks a "taking a bath" event point D (timestamp T_D); The data processing and analysis layer calculates the time interval ΔT2 = T_D - T_C = 45 minutes; The system's preset first time threshold T2 is 60 minutes. Since ΔT2 < T2, the system determines that this behavior has triggered the habit of "taking a bath shortly after a meal"; In the system's historical statistics, this habit accounts for 70% of the total number of bath events. The system generates a prompt message: "You are accustomed to taking a bath after a meal, which may affect digestion and cause discomfort. Please pay attention to improving this." and pushes it to the user via the mobile App.

[0019] Example 5 Based on Example 1, watching TV during a meal: The smart socket 101 detects that the kitchen range hood or rice cooker stops working at E (timestamp T_E). Combining with the fact that the human infrared detector 102 detects the user in the dining area at this time, the system marks a potential "end of meal" event point F (timestamp T_F), and at the same time calculates the meal time length ΔT3 = T_F - T_E; Detect the time span G (timestamp T_G) to H (timestamp T_H) of using the TV through the smart socket; The data processing and analysis layer determines that there is an intersection ΔT4 between the span of the meal time from E to F and the span of the TV watching time from G to H; The system's preset proportion of the intersection time ΔT4 of eating and watching TV in the meal time ΔT3 exceeds 80%. The system determines that this behavior has triggered the habit of "watching TV during a meal"; In the system's historical statistics, this habit accounts for 70% of the total number of meal events. The system generates a prompt message: "You are accustomed to watching TV during a meal, which may affect digestion and cause discomfort. Please pay attention to improving this." and pushes it to the user via the mobile App.

[0020] Example 6 Based on Example 1, leftover food: The smart socket 101 detects that the kitchen range hood or rice cooker stops working at 07:00. At the same time, the human infrared detector 102 detects that the user is in the dining area, and the system marks a potential "end of meal" time point I. The use of the refrigerator is detected by the opening and closing of the door magnetic sensor 104. The data processing and analysis layer pre-sets a window time period ΔT5 after the meal is finished, defining the time from the "end of meal" time point I through the ΔT5 time period to time point J; If the opening and closing of the refrigerator door magnet falls between time points I and J, the system determines that this behavior triggers the habit of "eating leftovers"; According to historical statistics, this habit accounts for 50% of all mealtime events. The system generates a notification message: "Improper storage or prolonged storage of leftover food may pose health risks due to bacterial growth or increased nitrite content. Please take precautions." This message is then pushed to the user via the mobile app.

[0021] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for identifying daily non-healthy life habits, characterized in that, It comprises a data collection layer, a data processing and analysis layer, and an application layer. The data collection layer comprises an intelligent socket (101), a human body infrared detector (102), a sleep sensor (103), a door magnetic opening and closing sensor (104), a hygrometer (105), a water use sensor (106), an air quality detector (107), a toilet flushing detector (108), an SOS button (109), a smart night light (110), and a temperature sensor (111); The intelligent socket (101), the human body infrared detector (102), the sleep sensor (103), the door magnetic opening and closing sensor (104), the hygrometer (105), the water use sensor (106), the air quality detector (107), the toilet flushing detector (108), the SOS button (109), and the smart night light (110) are respectively installed at corresponding positions in the living room, the bed, the kitchen, the dining room, and the bathroom; The intelligent socket (101) is used for monitoring the start and stop states of various electrical appliances in each space to determine the start and end times of meals and entertainment activities; The human body infrared detector (102) is used for sensing the activities and movements of personnel in the living room, the bed, the kitchen, the dining room, and the bathroom; The sleep sensor (103) is used for monitoring the time of getting into bed, getting out of bed, and the sleep state of personnel; The door magnetic opening and closing sensor (104) is used for recording the opening and closing events and their durations of the refrigerator door, the window, and the bathroom door; The hygrometer (105) is used for monitoring the temperature and humidity changes of the bathroom environment as auxiliary evidence for determining the bathing behavior; The water use sensor (106) is used for detecting the water use in the bathroom and the kitchen as auxiliary evidence for determining cooking and washing; The air quality detector (107) is used for monitoring the air quality in the kitchen; The toilet flushing detector (108) is used for monitoring the toilet flushing frequency; The SOS button (109) is used for triggering an alarm when the user is unwell; The smart night light (110) is used for providing illumination for the bed head and automatically turning off after the personnel falls asleep; The temperature sensor (111) is used for detecting the shower water temperature; The data processing and analysis layer is used for analyzing and processing the data collected by the data collection layer, identifying simple behavior events, and determining whether unhealthy living habits are triggered; The application layer is used for generating targeted health improvement suggestions for unhealthy behaviors and generating a warning signal to send to the user terminal.

2. The method for identifying daily non-healthy living habits of claim 1, wherein, The data processing and analysis layer comprises a data queue module, an IOT data driving engine module, an IOT algorithm determination module, and a DB module; The data queue module is used for receiving real-time data from each device of the data collection layer and aligning and storing the data with a unified timestamp; The IOT data driving engine module is used for receiving data from the data queue module, performing denoising, missing value filling, and feature extraction on the data, and identifying simple behavior events based on a preset logic; The IOT algorithm determination module is configured to associate the basic events identified by the IOT data driving engine on a time axis, mine long-term habit patterns, determine whether a non-healthy living habit is triggered according to preset habit determination rules, and count the number of times and total duration of a specific habit being triggered in a statistical period, and determine that the non-healthy habit has been formed when a set frequency or time threshold is exceeded; The DB module is configured to store the data aligned by the data queue module, support backtracking analysis, and store the behavior events identified by the IOT data driving engine and the habit patterns identified by the IOT algorithm determination module.

3. The method for identifying daily non-healthy living habits of claim 1, wherein, The application layer comprises a family old person life model construction engine module, an abnormal scene identification and early warning module, and a user terminal; The family old person life model construction engine module is configured to generate targeted health improvement suggestions based on the identified non-healthy behaviors; The abnormal scene identification and early warning module is configured to generate an early warning signal and send the early warning signal to the user terminal when the system identifies that the user has a high frequency of non-healthy habits, and generate an alarm signal and actively send alarm information to a specific user terminal when the air quality detector (107) detects that gas leakage occurs in the kitchen and the toilet flushing detector (108) sends an alarm information; The user terminal is configured to receive the early warning signal sent by the abnormal scene identification and early warning module, and display habit analysis reports and improvement and suggestions to the user.

4. The method for identifying unhealthy daily life habits of claim 2, wherein, The habit determination rules comprise a short-time bed after meal habit, a short-time bath after meal habit, a meal watching TV habit, a leftover meal habit, and a less water drinking habit; The short-time bed after meal habit is determined as follows: the smart socket (101) detects that a kitchen electrical appliance is running, the human body infrared detector (102) detects that a person is in the dining room, and then it is determined that the meal starts; when the human body infrared detector (102) detects that the person leaves the dining room, it is determined that the meal ends; if the meal end event is identified, and a sleep sensor (103) identifies a bed event within a preset first time threshold T1, then the habit is determined as one time; The short-time bath after meal habit is determined as follows: if the meal end event is identified, and a temperature and humidity meter (105) detects that the temperature and humidity in the bathroom continuously rise and exceed a threshold value, and a water sensor (106) and a temperature sensor (111) detect that water is used for bathing, then the habit is determined as one time; The meal watching TV habit is determined as follows: when the meal start event is identified, and the smart socket (101) identifies that the TV is turned on, the meal start event and the TV turn-on event overlap in time, and the overlapping time exceeds a preset third time threshold T3, then the habit is determined as one time; The leftover meal habit is determined as follows: within a single day, the door magnetic opening and closing sensor (104) identifies that the number of times of opening the refrigerator door exceeds a preset threshold T4, and the smart socket (101) detects that a microwave oven or an electric rice cooker is started, then the habit is determined as one time; The less water drinking habit is determined as follows: within a single day, the number of times of flushing the toilet is identified by the toilet flushing detector (108), and if the number of times of flushing the toilet is less than a preset threshold T5, then it is determined that the user has the habit.

5. The method for identifying unhealthy daily life habits of claim 1, wherein, The method comprises the following steps: S1: Collect multi-dimensional behavior data of users through a plurality of Internet of Things devices deployed in the environment; S2: Time align and preprocess the collected data; S3: Identify independent behavior events related to living habits based on the preprocessed data; S4: Correlate a plurality of independent behavior events on a time axis and identify specific unhealthy living habits according to preset habit determination rules; S5: Statistically evaluate the frequency and duration of the identified habits; S6: Generate and output health suggestions or warning information based on the evaluation results.

6. The method for identifying unhealthy daily life habits of claim 2, wherein, Also includes an IOT gateway; The IOT gateway is used to receive the collected data of the data collection layer, and uniformly convert the data into a standard protocol, filter, compress and preliminarily analyze the original data, reduce the transmission pressure of the cloud, and then send the processed data to the data queue module.

Images