Daily home life law identification method based on Internet of Things mode

Abstract

The invention relates to the technical field of daily home life law identification, in particular to a daily home life law identification method based on an Internet of Things mode, which comprises the following steps of Internet of Things data acquisition, data analysis, scene restoration, law extraction, model establishment and anomaly detection. Omnibearing data acquisition of a user behavior mode is realized, and the accuracy of data analysis is improved; a non-contact and non-wearing environment sensing mode is adopted, a user does not need to change habits or wear any equipment, experience is natural and free of burden, long-term and stable data collection is facilitated, deployment is simple, dependence on specific house types or expensive instruments is avoided, and the system is easy to popularize and install in various common family environments; by establishing a personal historical model, a periodic trend model and an attenuation model, subtle changes and deviation trends of life laws are intelligently analyzed, so that early warning signals are provided for abnormal behaviors, and the health and safety guarantee for users is greatly improved.

Description

Technical Field

[0001] This invention relates to the technical field of identifying daily home life patterns, and in particular to a method for identifying daily home life patterns based on the Internet of Things. Background Technology

[0002] Existing home health management methods mostly rely on wearable devices, such as clothing for measuring biological signals disclosed in patent CN116261426B and armband-type heart rate detection belt disclosed in patent CN204133455U, or manual recording methods. Although these methods can achieve health management for users' family members, they are not the only ones that can be used.

[0003] However, during use, it was found that the above methods are difficult to continuously and comprehensively reflect the daily routines of family members, such as their daily living habits, diet, sleep patterns, and activity patterns. While traditional smart home systems can collect some data such as sleep and electricity consumption (as disclosed in patent application CN117643454A, which describes a smart pillow, a sleep monitoring method based on a smart pillow, a system, and a storage medium), they lack cross-device information fusion capabilities and cannot accurately determine the changing trends of daily routines or health risks, resulting in poor practicality. Therefore, there is an urgent need for a method for identifying daily home life patterns based on the Internet of Things to improve the above problems. Summary of the Invention

[0004] To address the aforementioned technical issues, this invention provides a method for identifying daily home life patterns based on the Internet of Things (IoT). This method integrates data from multiple heterogeneous sensors, including sleep, infrared, water usage, electricity consumption, and door / motor sensors, overcoming the limitations of single-sensor monitoring. This collaborative analysis can accurately reconstruct complex daily life scenarios such as cooking, toileting, bathing, and living, achieving comprehensive data acquisition of user behavior patterns and improving the accuracy of data analysis. Employing a non-contact, non-wearable environmental sensing method, users do not need to change their habits or wear any devices, providing a natural and burden-free experience. It facilitates long-term and stable data collection, is simple to deploy, does not rely on specific apartment layouts or expensive equipment, and is easy to promote and install in various ordinary home environments. By establishing personal history models, periodic trend models, and decay models, it intelligently analyzes subtle changes and deviations in daily life patterns, providing early warning signals for abnormal behavior and greatly improving the health and safety protection of users.

[0005] The present invention provides a method for identifying daily home life patterns based on the Internet of Things, comprising the following steps: S1. Internet of Things (IoT) data acquisition: Collect user behavior data by deploying IoT devices in the home; S2, Data Analysis: The collected data is parsed to obtain information on the location, type, frequency, duration, and severity of the behavior. S3, Scene Recreation: Based on the analysis results, the user's daily behavior scenarios are inferred through a scene reconstruction algorithm; S4. Pattern Extraction: Perform statistical analysis on historical behavioral data; S5. Model Building: Based on the statistical analysis data in S4, a user lifestyle pattern model is constructed. S6. Anomaly Detection: Real-time data is matched with models to detect and warn of abnormal behavior.

[0006] Preferably, the IoT devices in the S1 IoT data acquisition specifically include: Sleep detectors, human infrared detectors, water sensors, toilet flush detectors, smart sockets, thermometers and hygrometers, entry infrared sensors, and door magnetic sensors.

[0007] Preferably, the sleep detector detects: in bed, out of bed, sleep initiation time, wake-up time, heart rate changes, respiratory rate changes, sleep stages, snoring, and body turning movements; The human infrared detector: detects whether there is a human infrared source; The water sensor detects whether water is being used. The toilet flushing detector: detects whether the toilet is flushing; The smart socket: detects whether there is a power consumption action; The thermometer and hygrometer are used to detect changes in ambient temperature and humidity. The in-home infrared sensor detects the presence of human infrared sources. The door magnetic sensor detects the opening and closing of doors and windows.

[0008] Preferably, the scene restoration in S3 specifically includes: Based on data from sleep monitors, the sleep initiation and wake-up times can be calculated in reverse, or scenarios involving prolonged bed rest can be identified. Based on the data reporting order of the entrance infrared sensor and the door magnetic sensor, determine whether the scene is homecoming or going out; Based on the coordinated triggering of human infrared detectors, smart sockets and water sensors in the kitchen area, combined with the duration of electricity use, the system determines whether the cooking or post-meal cleaning scenario is being performed. Based on humidity fluctuation data from infrared human body detectors, door magnetic sensors, toilet flushing detectors, and thermometers and hygrometers within the bathroom area, the system determines whether the person is using the toilet or taking a shower.

[0009] Preferably, the S4 pattern extraction specifically includes: The daily routines, meals, hygiene, outings, and spatial activities of families or individuals are statistically analyzed, and the statistical results are normalized according to a preset cycle to obtain a periodic behavioral pattern benchmark.

[0010] Preferably, the S5 model building specifically includes: Construct a historical full-scale model: Based on all historical data, extract the center value and offset range of time-related indicators, the mean of vital sign indicators, the mean of duration-related indicators, the mean of count-related indicators, and the mean of frequency-related indicators. Cyclical trend model: Based on cyclical data, analyze the changing trends of various indicators over time or seasons; Decay model: Fits the decay trend curves of various indicators based on historical data.

[0011] Preferably, the S6 anomaly detection specifically includes: Using the historical full-volume model as a benchmark, determine whether the real-time data deviates from its indicator range by more than a preset threshold; Using the aforementioned periodic change trend model as a benchmark, determine whether the trend of real-time data changes conforms to the expected pattern; Using the aforementioned attenuation model as a benchmark, determine whether the real-time data exceeds the expected attenuation range.

[0012] Compared with the prior art, the beneficial effects of the present invention are as follows: 1. This invention breaks through the limitations of single sensor monitoring by integrating data from multiple heterogeneous sensors such as sleep, infrared, water usage, electricity usage, and door magnets. This collaborative analysis can accurately reconstruct complex daily life scenarios such as cooking, toileting, bathing, and living, achieving comprehensive data acquisition of user behavior patterns and improving the accuracy of data analysis. 2. It adopts a non-contact, non-wearable environmental sensing method, so users do not need to change their habits or wear any devices. The experience is natural and burden-free, which facilitates long-term and stable data collection. It is also simple to deploy, does not depend on specific house types or expensive equipment, and is easy to promote and install in various ordinary home environments. 3. By establishing personal history models, cyclical trend models, and decay models, the system intelligently analyzes subtle changes and deviations from daily routines, providing early warning signals for abnormal behavior and greatly improving the health and safety of users. Attached Figure Description

[0013] Figure 1 This is a flowchart illustrating the present invention; Figure 2 This is a schematic diagram of the overall architecture of the present invention; Figure 3 This is a schematic diagram of the home deployment structure of the present invention; Figure 4This is a schematic diagram of the bathroom deployment structure of the present invention. Detailed Implementation

[0014] 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.

[0015] Example: Figures 1 to 4 As shown, a method for identifying daily home life patterns based on the Internet of Things includes the following steps: S1. Internet of Things (IoT) data acquisition: Collect user behavior data by deploying IoT devices in the home; IoT devices involve: Human infrared detector: detects the presence or absence of human infrared sources; Water sensor: detects whether water is being used; Toilet flush detector: Detects whether the toilet is flushing; Smart socket: detects whether there is any electrical activity; Thermometer and hygrometer: detects changes in ambient temperature and humidity; Indoor infrared sensor: detects the presence of human infrared sources; Door magnetic sensor: detects the opening and closing of doors and windows; S2, Data Analysis: The data collected is processed in terms of location (in which space), qualitative analysis (what happened), quantitative analysis (how many times, for how long, how much), and severity (degree). S3, Scene Recreation: By combining time and space dimensions to infer multiple data, we can obtain everyday life scenarios, such as sleeping, being sick, cooking, using the toilet, taking a bath, and going out. like: A. Based on the data from the sleep detector, the sleep initiation and wake-up determination scenarios are deduced in reverse. In addition, the bed state that spans multiple days is determined as a long-term bed rest scenario due to illness. B. When the magnetic sensor and infrared sensor at the entrance door are triggered, the system will reconstruct the scene to determine whether someone has returned home or gone out based on the order of data reporting. C. The home reports data from the kitchen infrared detector, triggering electrical equipment and water usage in the kitchen. In the meantime, actions such as opening and closing the refrigerator are present. The system reconstructs and determines the cooking and post-meal cleaning scenarios based on kitchen traces and electricity usage duration. D. The home reports data from the infrared human body detector in the bathroom. In conjunction with the opening and closing of the bathroom door magnetic sensor, if there is a toilet flushing action, it is determined to be a toilet use; if the humidity value reported by the thermometer and hygrometer jumps in a short period of time, it is determined to be a bathing activity. S4. Pattern Extraction: A. The system compiles daily statistics on the daily routines, meals, home life, hygiene, outings, and spatial activities of families and individuals. B. The system performs normalization processing (statistics, averaging, etc.) on a periodic basis (monthly / quarterly / yearly). S5. Model Building: Create profiles of family members and extract models of various indicators; A. Constructing a historical full-scale model: A-1. Obtain all data after the equipment is installed in a household or by an individual; A-2. For time-related indicators (such as sleep initiation and wake-up time), take the center value and offset range; A-3. For vital signs indicators (blood pressure, blood sugar, uric acid, etc.), the average value is taken based on the number of valid days; A-4. For time-related indicators (sleep duration, time spent in various spaces, time spent outside, cooking time, water usage time, TV viewing time, etc.), take the average value based on the number of valid days; A-5. For count-related indicators (number of sleeps, number of cooking sessions, number of toilet visits, etc.), take the average value based on the number of valid days; A-6. For frequency-related indicators (bathing frequency, bedridden frequency), take the average value over a period of time; B. Constructing a cyclical trend model: B-1. Obtain periodic (monthly / quarterly / yearly) data for households or individuals; B-2. Statistically analyze the changing trends of each indicator over time to determine the overall trend of health changes; B-3. ​​Compare various indicators seasonally to determine the seasonal trend of health changes. C. Constructing the decay model: C-1. Obtain all data after the equipment is installed in a household or by an individual; C-2. For each indicator, a decay trend curve is then fitted to guide subsequent anomaly detection. S6. Anomaly Detection: Monitor daily activities, match them using a periodic trend model, and provide early warnings for deviations; A. Using the historical full-volume model as a benchmark, verify whether the deviation of real-time measurement data from the benchmark range exceeds the threshold; B. Using the periodic trend model as a benchmark, verify whether the positive and negative trends of real-time measurement data conform to the expected pattern through the periodic trend model; C. Based on the attenuation model, determine whether the real-time measurement data exceeds the expected attenuation range.

[0016] The main functions achieved by this invention are: 1. This invention breaks through the limitations of single sensor monitoring by integrating data from multiple heterogeneous sensors such as sleep, infrared, water usage, electricity usage, and door magnets. This collaborative analysis can accurately reconstruct complex daily life scenarios such as cooking, toileting, bathing, and living, achieving comprehensive data acquisition of user behavior patterns and improving the accuracy of data analysis. 2. It adopts a non-contact, non-wearable environmental sensing method, so users do not need to change their habits or wear any devices. The experience is natural and burden-free, which facilitates long-term and stable data collection. It is also simple to deploy, does not depend on specific house types or expensive equipment, and is easy to promote and install in various ordinary home environments. 3. By establishing personal history models, cyclical trend models, and decay models, the system intelligently analyzes subtle changes and deviations from daily routines, providing early warning signals for abnormal behavior and greatly improving the health and safety of users.

[0017] 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 home life patterns based on the Internet of Things (IoT), characterized in that, Includes the following steps: S1. Internet of Things (IoT) data acquisition: Collect user behavior data by deploying IoT devices in the home; S2, Data Analysis: The collected data is parsed to obtain information on the location, type, frequency, duration, and severity of the behavior. S3, Scene Recreation: Based on the analysis results, the user's daily behavior scenarios are inferred through a scene reconstruction algorithm; S4. Pattern Extraction: Perform statistical analysis on historical behavioral data; S5. Model Building: Based on the statistical analysis data in S4, a user lifestyle pattern model is constructed. S6. Anomaly Detection: Real-time data is matched with models to detect and warn of abnormal behavior.

2. The method for identifying daily home life patterns based on the Internet of Things as described in claim 1, characterized in that, The IoT devices used in the S1 IoT data acquisition specifically include: Sleep detectors, human infrared detectors, water sensors, toilet flush detectors, smart sockets, thermometers and hygrometers, entry infrared sensors, and door magnetic sensors.

3. The method for identifying daily home life patterns based on the Internet of Things as described in claim 2, characterized in that, The sleep detector detects: in bed, out of bed, sleep initiation time, wake-up time, heart rate changes, respiratory rate changes, sleep stages, snoring, and body turning movements; The human infrared detector: detects whether there is a human infrared source; The water sensor detects whether water is being used. The toilet flushing detector: detects whether the toilet is flushing; The smart socket: detects whether there is a power consumption action; The thermometer and hygrometer are used to detect changes in ambient temperature and humidity. The in-home infrared sensor detects the presence of human infrared sources. The door magnetic sensor detects the opening and closing of doors and windows.

4. The method for identifying daily home life patterns based on the Internet of Things as described in claim 1, characterized in that, The scene restoration in S3 specifically includes: Based on data from sleep monitors, the sleep initiation and wake-up times can be calculated in reverse, or scenarios involving prolonged bed rest can be identified. Based on the data reporting order of the entrance infrared sensor and the door magnetic sensor, determine whether the scene is homecoming or going out; Based on the coordinated triggering of human infrared detectors, smart sockets and water sensors in the kitchen area, combined with the duration of electricity use, the system determines whether the cooking or post-meal cleaning scenario is being performed. Based on humidity fluctuation data from infrared human body detectors, door magnetic sensors, toilet flushing detectors, and thermometers and hygrometers within the bathroom area, the system determines whether the person is using the toilet or taking a shower.

5. The method for identifying daily home life patterns based on the Internet of Things as described in claim 1, characterized in that, The S4 pattern extraction specifically includes: The daily routines, meals, hygiene, outings, and spatial activities of families or individuals are statistically analyzed, and the statistical results are normalized according to a preset cycle to obtain a periodic behavioral pattern benchmark.

6. The method for identifying daily home life patterns based on the Internet of Things as described in claim 1, characterized in that, The S5 model establishment specifically includes: Construct a historical full-scale model: Based on all historical data, extract the center value and offset range of time-related indicators, the mean of vital sign indicators, the mean of duration-related indicators, the mean of count-related indicators, and the mean of frequency-related indicators. Cyclical trend model: Based on cyclical data, analyze the changing trends of various indicators over time or seasons; Decay model: Fits the decay trend curves of various indicators based on historical data.

7. The method for identifying daily home life patterns based on the Internet of Things as described in claim 1, characterized in that, The S6 anomaly detection specifically includes: Using the historical full-volume model as a benchmark, determine whether the real-time data deviates from its indicator range by more than a preset threshold; Using the aforementioned periodic change trend model as a benchmark, determine whether the trend of real-time data changes conforms to the expected pattern; Using the aforementioned attenuation model as a benchmark, determine whether the real-time data exceeds the expected attenuation range.

Images