Old people guarding method and system based on Internet of Things

Abstract

The invention provides an old people guarding method and system based on the Internet of Things, and the method comprises the steps: generating a control parameter distribution state of an Internet of Things node device according to the use data of the Internet of Things node device, the Internet of Things node device being a household intelligent device in communication connection with an Internet of Things central device; modeling based on the control parameter distribution state of the Internet of Things node equipment to obtain a behavioral habit model of a guarding object, monitoring daily use behaviors of the Internet of Things node equipment, and judging whether an abnormal use event which is not matched with the behavioral habit model exists or not; and when an abnormal use event which is not matched with the behavior habit model exists, calling video monitoring data to analyze the real-time state of the guarding object, and judging whether emergency treatment measures need to be executed or not according to the analysis result of the real-time state, so that the behavior data of the old people can be comprehensively obtained, and the safety of the old people is improved. Therefore, the physical state of the elderly can be analyzed and predicted more reasonably.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to an Internet of Things-based elderly care method and system. Background Art

[0002] With rapid economic development, influenced by various social factors and changing family values, the number of elderly people living alone has been growing over time, both in urban and rural areas. The self-care ability of elderly people living alone is closely linked to their physical condition. Since their physical condition often deteriorates with age, monitoring their health is crucial. Unlike in the past, with the increasing sophistication of my country's medical system, regular physical examinations for the elderly have become the norm, but this does not address the need for real-time monitoring. To address this, researchers have developed methods to monitor the physical condition of the elderly through wearable devices or mobile devices such as smart bracelets and smartphones, or through video surveillance. However, since the elderly are often not accustomed to wearing or carrying electronic devices like smart bracelets and smartphones, their practical monitoring capabilities are significantly reduced. Video surveillance can only provide simple analysis of the elderly's location, behavior, and physical condition. Without further data support, it is difficult to provide a more comprehensive analysis and prediction of their actual conditions. Summary of the Invention

[0003] Based on the above problems, the present invention proposes an Internet of Things-based elderly care method and system, which can comprehensively obtain the behavioral data of the elderly and conduct more reasonable analysis and prediction of the elderly's physical condition.

[0004] In view of this, the first aspect of the present invention proposes a method for protecting the elderly based on the Internet of Things, comprising:

[0005] Generate a control parameter distribution state of the IoT node device based on usage data of the IoT node device, wherein the IoT node device is a household smart device that is communicatively connected to the IoT hub device;

[0006] Modeling is performed based on the control parameter distribution state of the IoT node device to obtain a behavior habit model of the guarded object;

[0007] Monitor the daily usage behavior of IoT node devices;

[0008] Determining whether there are any abnormal usage events that do not match the behavioral habit model;

[0009] When there is an abnormal usage event that does not match the behavioral habit model, call the video surveillance data to analyze the real-time status of the guarded object;

[0010] Determine whether emergency measures need to be implemented based on the analysis results of the real-time status.

[0011] Furthermore, the step of generating a control parameter distribution state of the IoT node device according to the usage data of the IoT node device specifically includes:

[0012] Receive usage data reported by IoT node devices;

[0013] Identifying an event type of a usage event corresponding to the usage data, where the event type of the usage event includes an instantaneous event and a continuous event;

[0014] The usage data is parsed according to the event type of the usage event to obtain the control parameter distribution status of the Internet of Things node device, and the control parameter distribution status includes the numerical range of the control parameter of the Internet of Things node device and the distribution proportion of the control parameter of the Internet of Things node device corresponding to each numerical value in the numerical range.

[0015] Furthermore, the usage data includes an operation instruction executed by the IoT node device, and the step of identifying the event type of the usage event corresponding to the usage data specifically includes:

[0016] Determine an IoT node device that reports the usage data;

[0017] Extracting the operation instructions executed by the IoT node device from the usage data;

[0018] Determine whether the operation instruction causes the Internet of Things node device to enter a continuous working state from a non-working state, or causes the Internet of Things node device to enter another continuous working state from a continuous working state;

[0019] When the operation instruction causes the Internet of Things node device to enter a continuous working state from a non-working state, or causes the Internet of Things node device to enter another continuous working state from a continuous working state, determining that the usage event is a continuous event;

[0020] Otherwise, the usage event is determined to be a transient event.

[0021] Furthermore, the step of parsing the usage data according to the type of the usage event to obtain the control parameter distribution state of the Internet of Things node device specifically includes:

[0022] determining the control parameter of the usage event as a target control parameter;

[0023] Reading historical usage data of the guarded object corresponding to the usage event from a database;

[0024] determining a historical minimum value and a historical maximum value of the target control parameter from the historical usage data;

[0025] Determining a numerical range between the historical minimum value and the historical maximum value as a numerical range of the target control parameter;

[0026] Counting the number of occurrences of each value of the target control parameter within the value range of the target control parameter;

[0027] The ratio of the number of occurrences to the total number of occurrences of the target control parameter within the numerical range is determined as the distribution proportion of the corresponding numerical value.

[0028] Furthermore, the behavior habit model includes a plurality of usage event models connected in series on a time axis. The usage event model is composed of the numerical range and distribution ratio of the control parameters corresponding to each usage event. The steps of modeling the behavior habit model of the protected object based on the control parameter distribution state of the IoT node device specifically include:

[0029] Constructing the time axis of the behavioral habit model;

[0030] Counting the time range of occurrence of the usage event using the length of the time main axis as a period;

[0031] Constructing a usage event model based on the control parameter distribution state of the IoT node device;

[0032] The usage event model is associated with the time axis in chronological order using the occurrence time range as an association element.

[0033] Furthermore, the steps of constructing the time axis of the behavioral habit model specifically include:

[0034] Read the historical usage data of the guarded object from the database;

[0035] Performing a periodic feature analysis on the historical usage data in units of usage events to determine a usage feature period of the usage data of the guarded object on various IoT node devices, wherein the usage feature period is a minimum period with continuity;

[0036] The length of the time axis of the behavior habit model is determined according to the usage characteristic period.

[0037] Furthermore, the steps of performing periodic feature analysis on the historical usage data in units of usage events to determine the usage feature period of the usage data of the guarded object for various IoT node devices specifically include:

[0038] Constructing an occurrence time series of each usage event, wherein the occurrence time series is a discrete time series;

[0039] Extract the characteristic period T corresponding to the usage event based on the discrete time series i , where i is 1 to n event A positive integer between event The number of usage events of the guarded object;

[0040] Calculate the characteristic period T of all usage events of the guard object i The least common multiple of :

[0041] T LCM = lcm(T i );

[0042] The least common multiple T LCM The characteristic usage period is determined.

[0043] Furthermore, the step of constructing a usage event model based on the control parameter distribution state of the IoT node device specifically includes:

[0044] Determine the number of control parameters k for each usage event i ;

[0045] For the i-th usage event, construct a k i dimensional orthogonal space, the k i Each dimension in the dimensional orthogonal space corresponds to one of the control parameters of the i-th usage event;

[0046] Associate the numerical range and distribution weight of each control parameter to the k i The usage event model corresponding to the i-th usage event is obtained by the corresponding dimension in the dimensional orthogonal space.

[0047] Furthermore, the step of determining whether there is an abnormal usage event that does not match the behavioral habit model specifically includes:

[0048] When a usage event of any IoT node device is detected, the occurrence time and control parameters of the usage event are extracted;

[0049] Matching the occurrence time and control parameters with the usage event model on the time axis of the behavior habit model;

[0050] It is determined whether the usage event is an abnormal event based on the matching result.

[0051] The second aspect of the present invention proposes an Internet of Things-based elderly protection system, including an Internet of Things hub device and an Internet of Things node device communicatively connected to the networked hub device, and the Internet of Things hub device is configured to implement the Internet of Things-based elderly protection method described in any one of the first aspects of the present invention.

[0052] The present invention proposes a method and system for guarding the elderly based on the Internet of Things. The control parameter distribution status of the Internet of Things node device is generated according to the usage data of the Internet of Things node device. The Internet of Things node device is a household smart device that is communicatively connected to the Internet of Things hub device. The behavior habit model of the guarded object is obtained based on the control parameter distribution status of the Internet of Things node device. The daily usage behavior of the Internet of Things node device is monitored to determine whether there are abnormal usage events that do not match the behavior habit model. When there are abnormal usage events that do not match the behavior habit model, video surveillance data is called to analyze the real-time status of the guarded object. According to the analysis results of the real-time status, it is determined whether emergency treatment measures need to be implemented. The behavioral data of the elderly can be comprehensively obtained, and the physical condition of the elderly can be more reasonably analyzed and predicted based on this. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 This is a flowchart of a method for protecting the elderly based on the Internet of Things provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0054] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0055] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0056] In the description of the present invention, the term "plurality" refers to two or more. Unless otherwise specified, the terms "upper" and "lower" are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely to facilitate and simplify the description of the present invention. They are not intended to indicate or imply that the devices or components referred to must have a specific orientation, be constructed, or operate in a specific manner. Therefore, they should not be construed as limiting the present invention. The terms "connected," "mounted," and "fixed," etc., should be interpreted broadly. For example, "connected" can refer to fixed, removable, or integral connections; directly or indirectly through an intermediary. A person of ordinary skill in the art will understand the specific meanings of these terms in the present invention based on the specific circumstances. Furthermore, the terms "first," "second," etc., etc., are used for descriptive purposes only and should not be construed to indicate or imply relative importance or to implicitly specify the number of the technical features indicated. Therefore, a feature designated "first," "second," etc., may explicitly or implicitly include one or more of such features. In the description of the present invention, unless otherwise specified, "plurality" means two or more.

[0057] Throughout this specification, terms such as "one embodiment," "some implementations," and "specific examples" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0058] The following describes an elderly care method and system based on the Internet of Things according to some embodiments of the present invention with reference to the accompanying drawings.

[0059] like Figure 1 As shown, the first aspect of the present invention proposes a method for protecting the elderly based on the Internet of Things, comprising:

[0060] Generate a control parameter distribution state of the IoT node device based on usage data of the IoT node device, wherein the IoT node device is a household smart device that is communicatively connected to the IoT hub device;

[0061] Modeling is performed based on the control parameter distribution state of the IoT node device to obtain a behavior habit model of the guarded object;

[0062] Monitor the daily usage behavior of IoT node devices;

[0063] Determining whether there are any abnormal usage events that do not match the behavioral habit model;

[0064] When there is an abnormal usage event that does not match the behavioral habit model, call the video surveillance data to analyze the real-time status of the guarded object;

[0065] Determine whether emergency measures need to be implemented based on the analysis results of the real-time status.

[0066] Specifically, the IoT node devices are household smart devices networked through the IoT hub device, such as smart gas stoves, smart water dispensers, smart TVs, smart cups, smart curtains, and smart lamps. The IoT hub device can be a dedicated gateway device such as a router, or other portable or wearable smart device with sufficient storage and processing capabilities, such as a smartphone or smart bracelet. The IoT node devices establish a communication connection with the IoT hub device via Bluetooth or Wi-Fi, and report their usage data to the IoT hub device.

[0067] The usage data of the Internet of Things node device includes but is not limited to the power on and off of the Internet of Things node device, parameter adjustment of the Internet of Things node device and other usage data, such as the ignition data of the smart gas stove, the duration data in the ignition state and the gear adjustment data of the gas stove.

[0068] Furthermore, the step of calling the video surveillance data to analyze the real-time status of the guarded object specifically includes:

[0069] Performing human body recognition and / or facial recognition on the video surveillance image to determine the real-time location of the guarded object;

[0070] Performing human posture analysis on the monitoring image of the real-time location of the guarded object to determine whether there is any abnormality in the state of the guarded object.

[0071] After the step of calling the video surveillance data to analyze the real-time status of the guarded object, when emergency measures need to be implemented, measures such as sending a warning notification to the client device installed with the monitoring application through the cloud monitoring service, sending a warning text message to the emergency contact number, or making a phone call can be taken.

[0072] Furthermore, the step of generating a control parameter distribution state of the IoT node device according to the usage data of the IoT node device specifically includes:

[0073] Receive usage data reported by IoT node devices;

[0074] Identifying an event type of a usage event corresponding to the usage data, where the event type of the usage event includes an instantaneous event and a continuous event;

[0075] The usage data is parsed according to the event type of the usage event to obtain the control parameter distribution status of the Internet of Things node device, and the control parameter distribution status includes the numerical range of the control parameter of the Internet of Things node device and the distribution proportion of the control parameter of the Internet of Things node device corresponding to each numerical value in the numerical range.

[0076] Specifically, the usage event is a control event in which the IoT node device is directly or indirectly operated, driven, adjusted, etc. The control parameter is a parameter carried by the control instruction corresponding to the usage event, or a target parameter of the control instruction corresponding to the usage event.

[0077] For example, the control events of turning on, off, or adjusting the temperature of an air conditioner are all usage events, and the 26 degrees in the "adjusting the temperature of the air conditioner to 26 degrees" is the corresponding control parameter. In this embodiment, the control parameter is the target parameter of the control instruction corresponding to the usage event. In another embodiment, the current temperature of the air conditioner is 25 degrees, and the parameter carried by the control instruction is +1, so that the current temperature of the air conditioner is adjusted from 25 degrees to 26 degrees, and the parameter carried by the control instruction + 1 is used as the control parameter. Of course, the parameter carried by the control instruction corresponding to the usage event and the target parameter can also be used as one of the control parameters.

[0078] In some embodiments of the present invention, the numerical range of the control parameter is a numerical range consisting of historical numerical values ​​of the control parameter in historical usage events. In other embodiments of the present invention, the numerical range of the control parameter may also be a configurable range of the control parameter, which is typically limited by the hardware parameters of the IoT node device itself, or manually configured by design or manufacturing personnel.

[0079] The distribution ratio of each value of the control parameter in the numerical range is the proportion of the number of times the historical value of the control parameter is configured as each value in the numerical range in historical usage events, which is obtained by statistics of the control parameters of historical usage events.

[0080] Furthermore, before the step of receiving the usage data reported by the IoT node device, the method further includes:

[0081] Configure the facial image of the guarded object, the guarded object is the elderly who need to be monitored and guarded;

[0082] After the step of receiving the usage data reported by the IoT node device, the method further includes:

[0083] Calling video surveillance data to perform face recognition on surveillance images;

[0084] Determine, based on the face recognition result, whether the usage data reported by the IoT node device is the usage data generated when the protected object uses the IoT node device;

[0085] When the usage data reported by the IoT node device is not the usage data generated when the guarded object uses the IoT node device, the step of parsing the usage data according to the event type of the usage event to obtain the control parameter distribution state of the IoT node device is no longer performed, and the process returns to the step of receiving the usage data reported by the IoT node device;

[0086] When the usage data reported by the IoT node device is the usage data generated when the guarded object uses the IoT node device, a step is performed to parse the usage data according to the event type of the usage event to obtain the control parameter distribution state of the IoT node device.

[0087] Furthermore, after the step of receiving the usage data reported by the IoT node device, the method further includes:

[0088] parsing the usage data for usage events, event types, control parameters, and values ​​of the control parameters;

[0089] The usage event, event type, control parameter and the value of the control parameter are associated with the guard object and stored in a database.

[0090] Furthermore, the usage data includes an operation instruction executed by the IoT node device, and the step of identifying the event type of the usage event corresponding to the usage data specifically includes:

[0091] Determine an IoT node device that reports the usage data;

[0092] Extracting the operation instructions executed by the IoT node device from the usage data;

[0093] Determine whether the operation instruction causes the Internet of Things node device to enter a continuous working state from a non-working state, or causes the Internet of Things node device to enter another continuous working state from a continuous working state;

[0094] When the operation instruction causes the Internet of Things node device to enter a continuous working state from a non-working state, or causes the Internet of Things node device to enter another continuous working state from a continuous working state, determining that the usage event is a continuous event;

[0095] Otherwise, the usage event is determined to be a transient event.

[0096] Specifically, before the step of receiving usage data reported by the IoT node device, each time the IoT node device executes an operation instruction, it generates a data packet of usage data corresponding to the operation instruction. The usage data includes the operation instruction executed by the IoT node device and the control parameters corresponding to the operation instruction.

[0097] When the IoT node device is a device that mainly uses electricity to perform work tasks, the IoT node device being in a non-working state generally refers to the IoT node device being in a power-off state, or in a dormant state. For example, the light is in a light-off state, etc. Therefore, when the IoT node device enters a continuous working state from a non-working state, it is usually because the user turns on the power of the IoT node device or performs a wake-up operation, causing the IoT node device to enter a working state from a power-off state or a dormant state. For gas-using equipment such as gas stoves or gas water heaters, being in a non-working state means that it is in a state where gas is not supplied. For water-using or water treatment equipment such as water purifiers, being in a non-working state means that it is in a state where water is not supplied.

[0098] The continuous working state refers to the state in which the IoT node device continuously performs a specific action after being supplied with power, gas, or water. For example, if a light is powered and continuously in a working state of providing lighting, the corresponding continuous event is a lighting event. If a water dispenser is powered and continuously in a working state of cooling or heating drinking water, the corresponding continuous event is a heating event or a cooling event. If a gas stove is powered and continuously in a burning state, the corresponding continuous event is a heating event.

[0099] The IoT node device enters another continuous working state from one continuous working state, which means that under the operation of the user, a specific operation instruction is used to switch the continuous work performed by the IoT node device to another continuous work, such as switching the air conditioner from cooling mode to dehumidification mode.

[0100] The instantaneous event is an event that will not cause the Internet of Things node device to enter a continuous working state from a non-working state, and will not cause the Internet of Things node to enter another continuous working state from one continuous working state, including adjustments to the working parameters of the Internet of Things node, such as temperature adjustment events for air conditioners, water dispensers, etc., and shutdown events that shut down the power supply, water supply, and gas supply of the Internet of Things node device, such as turning off the TV, turning off the power of the air conditioner, etc.

[0101] Furthermore, the step of parsing the usage data according to the type of the usage event to obtain the control parameter distribution state of the Internet of Things node device specifically includes:

[0102] determining the control parameter of the usage event as a target control parameter;

[0103] Reading historical usage data of the guarded object corresponding to the usage event from a database;

[0104] determining a historical minimum value and a historical maximum value of the target control parameter from the historical usage data;

[0105] Determining a numerical range between the historical minimum value and the historical maximum value as a numerical range of the target control parameter;

[0106] Counting the number of occurrences of each value of the target control parameter within the value range of the target control parameter;

[0107] The ratio of the number of occurrences to the total number of occurrences of the target control parameter within the numerical range is determined as the distribution proportion of the corresponding numerical value.

[0108] Specifically, a usage event may include several control parameters. Therefore, when the target usage event has multiple control parameters, there are also multiple target control parameters, and the above steps are performed for each target control parameter.

[0109] According to actual implementation needs, the historical usage data can be all historical usage data of the guarded object or part of the historical usage data within a period of time, and can be adaptively adjusted according to the amount of usage data and changes in the usage habits of the guarded object.

[0110] In the step of determining the historical minimum value and the historical maximum value of the target control parameter from the historical usage data, the historical minimum value is the minimum value of the target control parameter among the values ​​that have appeared in the historical usage data. Similarly, the historical maximum value is the maximum value of the target control parameter among the values ​​that have appeared in the historical usage data.

[0111] Furthermore, after the step of parsing the usage data according to the type of the usage event to obtain the control parameter distribution state of the Internet of Things node device, the method further includes:

[0112] The control parameter distribution state of the Internet of Things node device and its corresponding usage event are associated with the guard object and stored in a database.

[0113] The control parameter distribution state of the Internet of Things node device can be stored in the form of key-value pairs of each discrete value of each control parameter within its numerical range and the corresponding distribution ratio, or the distribution ratio of each control parameter within its numerical range can be fitted into a curve function for storage.

[0114] Furthermore, the behavior habit model includes a plurality of usage event models connected in series on a time axis. The usage event model is composed of the numerical range and distribution ratio of the control parameters corresponding to each usage event. The steps of modeling the behavior habit model of the protected object based on the control parameter distribution state of the IoT node device specifically include:

[0115] Constructing the time axis of the behavioral habit model;

[0116] Counting the time range of occurrence of the usage event using the length of the time main axis as a period;

[0117] Constructing a usage event model based on the control parameter distribution state of the IoT node device;

[0118] The usage event model is associated with the time axis in chronological order using the occurrence time range as an association element.

[0119] Specifically, on the time axis, the usage event model can be viewed as a number of model units with a certain time range. Utilizing the time axis, the behavior habit model of the protected object can be visualized in a relatively intuitive manner within the time range of the time axis.

[0120] In some embodiments of the present invention, the time axis of the behavioral habit model has a preconfigured specific length. For example, the time axis can be one day, or it can be another time axis such as one week, one month, or one quarter. In other embodiments of the present invention, the time axis length can also be dynamically configured based on the behavioral habits of the protected object.

[0121] When the usage event is an instantaneous event, the time range of the occurrence of the usage event is the time range within which the historical occurrence time of the instantaneous event falls. When the usage event is a continuous event, the time range of the occurrence of the usage event is the superposition range of the time range covered by the continuous working state entered by the IoT node device due to the usage event in history. Taking the length of the time axis as one day as an example, when the guarded object usually watches TV from eight to ten in the morning, the time range of the usage event of watching TV is determined to be from eight to ten. Of course, in actual implementation, it can be a time range with more precise start and end time points obtained based on actual usage record statistics.

[0122] Furthermore, the steps of constructing the time axis of the behavioral habit model specifically include:

[0123] Read the historical usage data of the guarded object from the database;

[0124] Performing a periodic feature analysis on the historical usage data in units of usage events to determine a usage feature period of the usage data of the guarded object on various IoT node devices, wherein the usage feature period is a minimum period with continuity;

[0125] The length of the time axis of the behavior habit model is determined according to the usage characteristic period.

[0126] In the step of reading the historical usage data of the guarded object from the database, the historical usage data is the historical usage data corresponding to all usage events of the guarded object for all IoT node devices.

[0127] In most cases, a daily cycle aligns with people's habit of staying out at night and staying in bed during the day. However, many elderly people have multiple residences, so the protected person may live in different places at intervals. The time axis generated with a daily cycle will cause the constructed behavior habit model to lack some periodic features. In the technical solution of the above-mentioned embodiment, by performing a periodic feature analysis on the historical usage data based on usage events to determine a minimum continuous period as the usage feature period, the behavior habit model can maximize the reflection of the daily behavior habits of the protected person.

[0128] Furthermore, the steps of performing periodic feature analysis on the historical usage data in units of usage events to determine the usage feature period of the usage data of the guarded object for various IoT node devices specifically include:

[0129] Constructing an occurrence time series of each usage event, wherein the occurrence time series is a discrete time series;

[0130] Extract the characteristic period T corresponding to the usage event based on the discrete time series i , where i is 1 to n event A positive integer between event The number of usage events of the guarded object;

[0131] Calculate the characteristic period T of all usage events of the guard object i The least common multiple of :

[0132] T LCM = lcm(T i );

[0133] The least common multiple T LCM The characteristic usage period is determined.

[0134] Specifically, in the calculation formula of the least common multiple, lcm() is a function representation of the least common multiple.

[0135] In the technical solution of the above embodiment, constructing the occurrence time sequence of each usage event specifically includes:

[0136] When the usage event is an instantaneous event, its occurrence time is directly extracted from the historical usage data to construct an occurrence time series of the usage event;

[0137] When the usage event is a continuous event, the starting time point of each usage event is used as its occurrence time to construct an occurrence time sequence of the usage event.

[0138] In the technical solution of the above embodiment, the characteristic period is the minimum period extracted from the discrete time series, and the characteristic period of the discrete time series can be extracted using algorithms such as autocorrelation function or Fourier transform.

[0139] It should be known that the number of usage events n in the guard object event In the counting, an operation instruction of the IoT device is used as the number of usage events n event The counting target is the number of usage events n corresponding to the same operation instruction of the same IoT device that is executed multiple times. event Are only counted once.

[0140] Furthermore, the step of constructing a usage event model based on the control parameter distribution state of the IoT node device specifically includes:

[0141] Determine the number of control parameters k for each usage event i ;

[0142] For the i-th usage event, construct a k i dimensional orthogonal space, the k i Each dimension in the dimensional orthogonal space corresponds to one of the control parameters of the i-th usage event;

[0143] Associate the numerical range and distribution weight of each control parameter to the k i The usage event model corresponding to the i-th usage event is obtained by the corresponding dimension in the dimensional orthogonal space.

[0144] In the technical solution of the above embodiment, k i is n event The number of control parameters of the i-th usage event in the k usage events. Each control parameter of the usage event is in the k i dimensional orthogonal space corresponds to one dimension, so the usage event model is in the k iA numerical model of the distribution weights in a dimensional orthogonal space that is confined to a specific range of values ​​in each dimension.

[0145] When it is necessary to visualize the details of the usage event model, the usage event model can be reduced in dimension by filtering. For example, fewer than four control parameters can be selected from the usage event model as target control parameters to form an orthogonal space with dimensions less than four, so that their distribution proportions can be combined and visualized. In some embodiments, multiple strongly correlated control parameter combinations can be preconfigured for user selection.

[0146] Furthermore, the step of determining whether there is an abnormal usage event that does not match the behavioral habit model specifically includes:

[0147] When a usage event of any IoT node device is detected, the occurrence time and control parameters of the usage event are extracted;

[0148] Matching the occurrence time and control parameters with the usage event model on the time axis of the behavior habit model;

[0149] It is determined whether the usage event is an abnormal event based on the matching result.

[0150] Furthermore, the step of matching the occurrence time and control parameters with the usage event model on the time axis of the behavior habit model specifically includes:

[0151] Define a global outlier value and configure a global outlier threshold;

[0152] determining a target usage event model corresponding to the usage event;

[0153] Matching the occurrence time of the usage event with the occurrence time range of the target usage event model;

[0154] When the occurrence time of the usage event does not fall within the occurrence time range of the target usage event model, the global anomaly value is accumulated by a first step length, which is a pre-configured anomaly value accumulation step length corresponding to the anomaly of the usage event occurrence time.

[0155] Furthermore, the step of matching the occurrence time and control parameters with the usage event model on the time axis of the behavior habit model further includes:

[0156] Configure a distribution weight threshold;

[0157] Calculating a target numerical range corresponding to a distribution weight of each control parameter in the usage event model being greater than the distribution weight threshold;

[0158] Matching the values ​​of the control parameters of the usage event with the target value ranges of the control parameters corresponding to the target usage event model;

[0159] When the value of any control parameter in the usage event does not fall within the target value range, the global abnormal value is accumulated by a second step length, which is a pre-configured abnormal value accumulation step length corresponding to the abnormal control parameter of the usage event.

[0160] The above embodiment adopts a solution in which all control parameters use the same second step size. In the technical solutions of other embodiments of the present invention, independent outlier accumulation step sizes can also be configured for some control parameters. For control parameters that are not configured with independent outlier accumulation step sizes, a unified second step size is used as their outlier accumulation step size.

[0161] Furthermore, the global outlier value is accumulated in each statistical period, with the length of the time axis being the statistical period. That is, the global outlier value is valid only within one statistical period. After each statistical period, the global outlier value is reset to zero and accumulated again in the next statistical period.

[0162] Furthermore, when the global abnormal value is greater than the global abnormal threshold, emergency treatment measures are executed.

[0163] The second aspect of the present invention proposes an Internet of Things-based elderly protection system, including an Internet of Things hub device and an Internet of Things node device communicatively connected to the networked hub device, and the Internet of Things hub device is configured to implement the Internet of Things-based elderly protection method described in any one of the first aspects of the present invention.

[0164] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0165] While embodiments of the present invention have been described above, these embodiments do not exhaustively describe all details and do not limit the invention to the specific embodiments described. Obviously, many modifications and variations are possible based on the above description. These embodiments are selected and described in detail in this specification in order to better explain the principles and practical applications of the present invention, thereby enabling those skilled in the art to better utilize the present invention and its modifications. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A method for protecting the elderly based on the Internet of Things, characterized in that: include: Generate a control parameter distribution state of the IoT node device based on usage data of the IoT node device, wherein the IoT node device is a household smart device that is communicatively connected to the IoT hub device; Modeling is performed based on the control parameter distribution state of the IoT node device to obtain a behavior habit model of the guarded object; Monitor the daily usage behavior of IoT node devices; Determining whether there are any abnormal usage events that do not match the behavioral habit model; When there is an abnormal usage event that does not match the behavioral habit model, call the video surveillance data to analyze the real-time status of the guarded object; Determine whether emergency measures need to be implemented based on the analysis results of the real-time status.

2. The method for protecting the elderly based on the Internet of Things according to claim 1 is characterized in that: The step of generating a control parameter distribution state of the IoT node device according to the usage data of the IoT node device specifically includes: Receive usage data reported by IoT node devices; Identifying an event type of a usage event corresponding to the usage data, where the event type of the usage event includes an instantaneous event and a continuous event; The usage data is parsed according to the event type of the usage event to obtain the control parameter distribution status of the Internet of Things node device, and the control parameter distribution status includes the numerical range of the control parameter of the Internet of Things node device and the distribution proportion of the control parameter of the Internet of Things node device corresponding to each numerical value in the numerical range.

3. The method for protecting the elderly based on the Internet of Things according to claim 2 is characterized in that: The usage data includes an operation instruction executed by the IoT node device, and the step of identifying the event type of the usage event corresponding to the usage data specifically includes: Determine an IoT node device that reports the usage data; Extracting the operation instructions executed by the IoT node device from the usage data; Determine whether the operation instruction causes the Internet of Things node device to enter a continuous working state from a non-working state, or causes the Internet of Things node device to enter another continuous working state from a continuous working state; When the operation instruction causes the Internet of Things node device to enter a continuous working state from a non-working state, or causes the Internet of Things node device to enter another continuous working state from a continuous working state, determining that the usage event is a continuous event; Otherwise, the usage event is determined to be a transient event.

4. The method for protecting the elderly based on the Internet of Things according to claim 2, characterized in that: The step of parsing the usage data according to the type of the usage event to obtain the control parameter distribution state of the Internet of Things node device specifically includes: determining the control parameter of the usage event as a target control parameter; Reading historical usage data of the guarded object corresponding to the usage event from a database; determining a historical minimum value and a historical maximum value of the target control parameter from the historical usage data; Determining a numerical range between the historical minimum value and the historical maximum value as a numerical range of the target control parameter; Counting the number of occurrences of each value of the target control parameter within the value range of the target control parameter; The ratio of the number of occurrences to the total number of occurrences of the target control parameter within the numerical range is determined as the distribution proportion of the corresponding numerical value.

5. The method for protecting the elderly based on the Internet of Things according to claim 1 is characterized in that: The behavior habit model includes a plurality of usage event models connected in series on a time axis. The usage event model is composed of the numerical range and distribution ratio of the control parameters corresponding to each usage event. The steps of modeling the behavior habit model of the protected object based on the control parameter distribution state of the IoT node device specifically include: Constructing the time axis of the behavioral habit model; Counting the time range of occurrence of the usage event using the length of the time main axis as a period; Constructing a usage event model based on the control parameter distribution state of the IoT node device; The usage event model is associated with the time axis in chronological order using the occurrence time range as an association element.

6. The method for protecting the elderly based on the Internet of Things according to claim 5, characterized in that: The steps of constructing the time axis of the behavioral habit model specifically include: Read the historical usage data of the guarded object from the database; Performing a periodic feature analysis on the historical usage data in units of usage events to determine a usage feature period of the usage data of the guarded object on various IoT node devices, wherein the usage feature period is a minimum period with continuity; The length of the time axis of the behavior habit model is determined according to the usage characteristic period.

7. The method for protecting the elderly based on the Internet of Things according to claim 6 is characterized in that: The steps of performing periodic feature analysis on the historical usage data in units of usage events to determine the usage feature period of the usage data of the guarded object for various IoT node devices specifically include: Constructing an occurrence time series of each usage event, wherein the occurrence time series is a discrete time series; Extract the characteristic period T corresponding to the usage event based on the discrete time series i , where i is 1 to n event A positive integer between event The number of usage events of the guarded object; Calculate the characteristic period T of all usage events of the guard object i The least common multiple of : T LCM =lcm(T i ); The least common multiple T LCM The characteristic usage period is determined.

8. The method for protecting the elderly based on the Internet of Things according to claim 5 is characterized in that: The steps of constructing a usage event model based on the control parameter distribution state of the IoT node device specifically include: Determine the number of control parameters k for each usage event i ; For the i-th usage event, construct a k i dimensional orthogonal space, the k i Each dimension in the dimensional orthogonal space corresponds to one of the control parameters of the i-th usage event; Associate the numerical range and distribution weight of each control parameter to the k i The usage event model corresponding to the i-th usage event is obtained by the corresponding dimension in the dimensional orthogonal space.

9. The method for protecting the elderly based on the Internet of Things according to claim 8, characterized in that: The steps of determining whether there is an abnormal usage event that does not match the behavioral habit model specifically include: When a usage event of any IoT node device is detected, the occurrence time and control parameters of the usage event are extracted; Matching the occurrence time and control parameters with the usage event model on the time axis of the behavior habit model; It is determined whether the usage event is an abnormal event based on the matching result.

10. An elderly care system based on the Internet of Things, characterized in that: It includes an Internet of Things hub device and an Internet of Things node device communicatively connected to the networking hub device, and the Internet of Things hub device is configured to implement the Internet of Things-based elderly care method as described in any one of claims 1-9.

Images