Smart home monitoring and early warning method and system based on Internet of Things sensing equipment

Abstract

The invention relates to the technical field of home environment detection, and particularly discloses a smart home monitoring and early warning method and system based on Internet of Things sensing equipment, and the method comprises the steps: carrying out the statistics of the operation data of each piece of home equipment, and obtaining an operation matrix; determining a color value parameter of each equipment contour in real time according to the operation matrix, and updating the home map at the corresponding moment; determining an abnormal value at each position according to the home map at each moment, and determining an image acquisition frequency at each position according to the abnormal value at each position; acquiring images at each position based on the image acquisition frequency, inputting the trained risk identification model, and outputting a risk identification result; according to the invention, periodic analysis is carried out on the working parameters of various home devices, whether the home devices are abnormal or not is judged, a selective image acquisition and detection process is carried out on the whole home environment according to the judgment result, in the process, the number of images needing to be acquired is sharply reduced, and the power consumption is greatly reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of home environment detection, and in particular to a smart home monitoring and early warning method and system based on Internet of Things sensor equipment. Background Art

[0002] The smart home monitoring and early warning system is a system that uses smart devices and sensors to monitor the home environment in real time and issue early warnings when abnormal situations occur.

[0003] Today's monitoring and early warning systems are often used in scenarios where the elderly live alone. In the current social context, many elderly people do not choose to live with their children, and many elderly people do not have children. As they age, their physical condition becomes less and less ideal, and it is easy for them to have some unexpected situations, such as falling, which require timely reporting; the best way is to install a large number of cameras in the elderly's homes to monitor and warn them in real time, but this method consumes a lot of power. When the elderly want to save electricity, they are likely to turn off the cameras in the hope of saving electricity, which is obviously inappropriate. Therefore, how to reduce the energy consumption of the monitoring and early warning system is the technical problem that the technical solution of the present invention wants to solve. Summary of the invention

[0004] The purpose of the present invention is to provide a smart home monitoring and early warning method and system based on Internet of Things sensor devices to solve the problems raised in the above background technology.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A smart home monitoring and early warning method based on Internet of Things sensor equipment, the method comprising:

[0007] Get the device location of home devices in real time, create device outlines in the floor plan based on the device location, and obtain a home map with time;

[0008] Obtain the sensor device type of the home appliance, use the sensor device type as a column and the time as a row to establish a basic matrix, and count the operation data of each home appliance based on the basic matrix to obtain an operation matrix;

[0009] Determine the color value parameters of each device outline in real time according to the operation matrix, and update the home map at the corresponding moment;

[0010] Determine the abnormal value at each location according to the home map at each time, and determine the image acquisition frequency at each location according to the abnormal value at each location; wherein the equipment used for acquiring images includes but is not limited to fixed cameras and mobile cameras;

[0011] Obtain images at each position based on the image acquisition frequency, input them into the trained risk recognition model, and output the risk recognition results.

[0012] As a further solution of the present invention, the step of obtaining the device position of the home appliance in real time, creating a device contour in the floor plan based on the device position, and obtaining a home map containing time includes:

[0013] Obtain the device position containing time according to the locator built in the home appliance; the device position is the relative position based on the home origin; the home origin is the bottom endpoint of the center line of the door;

[0014] Read the floor plan, obtain the top view contour of the home appliance, and scale the top view contour of the home appliance according to the scale of the floor plan to obtain the device contour;

[0015] For any time, read the device positions at the nearest moment of all home appliances corresponding to this time, and insert the device contour at the device position in the floor plan to obtain the home map at this time;

[0016] Among them, for the process of obtaining the device position, the acquisition frequency of the device position is adjusted in real time according to the change situation of the device position.

[0017] As a further solution of the present invention, the step of obtaining the type of the sensing device of the home appliance, taking the type of the sensing device as the column and the time as the row, establishing a basic matrix, and statistically analyzing the operation data of each home appliance based on the basic matrix to obtain an operation matrix includes:

[0018] Obtain the sensing devices installed in each home appliance, query the types of the sensing devices, and arrange the types in a preset order as the column labels;

[0019] Determine the acquisition moments according to a preset time period, and arrange the acquisition moments in time order as the row labels;

[0020] Statistically analyze the column labels and row labels to obtain a basic matrix;

[0021] Receive the operation data containing time uploaded by each sensing device in the home appliance, and determine the row and column positions according to the sensing device and time;

[0022] Insert the operation data into the corresponding row and column positions in the basic matrix to obtain an operation matrix.

[0023] As a further solution of the present invention, the step of determining the color value parameters of each device contour in real time according to the operation matrix and updating the home map at the corresponding moment includes:

[0024] For any home map, a sub-matrix is intercepted from the running matrix according to a preset time period; the number of columns of the sub-matrix is the same as that of the running matrix, and the number of rows of the sub-matrix is less than that of the running matrix;

[0025] Perform periodic analysis on each column of data in the sub-matrix, and determine the color value parameters of each device contour according to the results of the periodic analysis;

[0026] Insert the color value parameters into the device contours in the home map to obtain the updated home map.

[0027] As a further solution of the present invention, the step of performing periodic analysis on each column of data in the sub-matrix and determining the color value parameters of each device contour according to the results of the periodic analysis includes:

[0028] Perform Fourier transform on each column of data to obtain frequency domain components;

[0029] Compare the frequency domain components with preset standard components to determine the difference degree of each column of data;

[0030] Statistically analyze the difference degrees of all columns of data, and determine the color value parameters of each device contour according to the difference degrees.

[0031] As a further solution of the present invention, the method of determining the outliers at each position according to the home map at each moment and determining the image acquisition frequency at each position includes:

[0032] For the home map at any moment, spread the color value parameters of the device contours of the home map to obtain the color value parameters at each position in the home map;

[0033] Determine the outliers according to the color value parameters;

[0034] Determine the image acquisition frequency at each position according to the outliers;

[0035] Perform image acquisition on each position based on the image acquisition frequency;

[0036] Among them, the fixed camera includes a camera installed on a fixed device, and the mobile camera includes a camera installed on a mobile device.

[0037] The technical solution of the present invention also provides an intelligent home monitoring and warning system based on Internet of Things sensing devices, and the system includes:

[0038] A home map acquisition module, configured to obtain the device positions of home devices in real time, and create device contours in the house type map based on the device positions to obtain a home map containing time;

[0039] The operation matrix generation module is used to obtain the types of sensing devices of home appliances, take the types of sensing devices as columns and time as rows to establish a basic matrix, and based on the basic matrix, count the operation data of each home appliance to obtain an operation matrix;

[0040] The home map update module is used to determine the color value parameters of each device contour in real time according to the operation matrix and update the home map at the corresponding moment;

[0041] The acquisition frequency determination module is used to determine the outliers at each position according to the home map at each moment, and determine the image acquisition frequency at each position according to the outliers at each position; the devices for acquiring images include but are not limited to fixed cameras and mobile cameras;

[0042] The risk identification module is used to obtain the images at each position based on the image acquisition frequency, input them into the trained risk identification model, and output the risk identification result.

[0043] As a further solution of the present invention, the home map acquisition module includes:

[0044] The device position acquisition unit is used to obtain the device position containing time according to the locator built in the home appliance; the device position is the relative position based on the home origin; the home origin is the bottom endpoint of the center line of the door;

[0045] The contour scaling unit is used to read the house type plan, obtain the top view contour of the home appliance, and scale the top view contour of the home appliance according to the scale of the house type plan to obtain the device contour;

[0046] The contour insertion unit is used to, for any time, read the device positions at the nearest moment corresponding to all home appliances at this time, and insert the device contour at the device position in the house type plan to obtain the home map at this time;

[0047] Among them, for the acquisition process of the device position, the acquisition frequency of the device position is adjusted in real time according to the change situation of the device position.

[0048] As a further solution of the present invention, the operation matrix generation module includes:

[0049] The column label generation unit is used to obtain the sensing devices installed in each home appliance, query the types of the sensing devices, and arrange the types in a preset order as column labels;

[0050] The row label generation unit is used to determine the acquisition moments according to a preset time period, and arrange the acquisition moments in time order as row labels;

[0051] The template generation unit is used to count the column labels and row labels to obtain a basic matrix;

[0052] A position determination unit, configured to receive the operation data containing time uploaded by each sensing device in the home appliance, and determine the row and column positions according to the sensing device and the time;

[0053] A data insertion unit, configured to insert the operation data into the corresponding row and column positions in the basic matrix to obtain an operation matrix.

[0054] As a further solution of the present invention, the home map update module includes:

[0055] A matrix intercepting unit, configured to intercept a sub-matrix from the operation matrix according to a preset time period for any home map; the number of columns of the sub-matrix is the same as the number of columns of the operation matrix, and the number of rows of the sub-matrix is less than the number of rows of the operation matrix;

[0056] A period analysis unit, configured to perform period analysis on each column of data in the sub-matrix, and determine the color value parameters of each device contour according to the period analysis result;

[0057] A map output unit, configured to insert the color value parameters into the device contour in the home map as the updated home map.

[0058] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention performs period analysis on the working parameters of various home appliances, determines whether there is an abnormality in the home appliances, and then performs a selective image acquisition and detection process on the entire home environment according to the determination result. In this process, the image acquisition link with extremely high power consumption is optimized, and the number of images to be acquired is sharply reduced, greatly reducing the power consumption. Brief Description of the Drawings

[0059] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention.

[0060] Figure 1 It is a flowchart of a smart home monitoring and early warning method based on Internet of Things sensing devices.

[0061] Figure 2 It is a block diagram of the composition structure of a smart home monitoring and early warning system based on Internet of Things sensing devices. Detailed Embodiments

[0062] In order to make the technical problems, technical solutions and beneficial effects to be solved by the present invention clearer, the following further details the present invention in conjunction with the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0063] Figure 1It is a flowchart of a smart home monitoring and warning method based on Internet of Things sensing devices. In an embodiment of the present invention, a smart home monitoring and warning method based on Internet of Things sensing devices includes the following steps:

[0064] Step S100: Real-time obtain the device location of home appliances, create a device contour in the floor plan based on the device location, and obtain a home map containing time.

[0065] The home appliances mentioned in this application refer to home appliances with built-in information collection modules and transmission modules, including televisions, refrigerators, air conditioners, etc. containing various sensors; the home scene is a residential space where home appliances are installed inside the residential space. Obtain the floor plan of the residential space, real-time obtain the device location of home appliances, and the device location can be fixed or mobile. For example, a television is generally fixed, and a sweeping robot is generally mobile. Each device has a certain volume and a definite contour. After real-time obtaining the device location of home appliances, read the device contour (known data) of the home appliances, and insert the device contour into the floor plan to obtain a home map. It should be noted that the process of obtaining the device contour and inserting the device contour into the floor plan needs to consider time, that is, when real-time obtaining the device location, record the time of the device location, and when creating the home map, insert the device locations at the same time into the same home map. Thus, it can be obtained that the home map also contains a time tag.

[0066] In an example of the technical solution of the present invention, when creating a home map, some time points will be determined first. For example, a home map is created every 30s. This also means that the device locations of all home appliances are obtained every 30s, and the device contour is inserted into the floor plan according to the obtained device location, so that a home map can be obtained every 30s.

[0067] Step S200: Obtain the type of sensing device of the home appliance, use the type of sensing device as columns and time as rows to establish a basic matrix, and statistically analyze the operation data of each home appliance based on the basic matrix to obtain an operation matrix.

[0068] Some sensing devices are pre-installed in each home appliance, and these sensing devices are used to obtain the operation parameters during the operation of the home appliance. The types of sensing devices are not unique. Some are used to obtain temperature, and some are used to obtain signals. Obtain the type of sensing device of the sensing device in the home appliance, use the type of sensing device as columns and time as rows to establish a basic matrix, and statistically analyze the operation data of each home appliance based on the basic matrix to obtain an operation matrix.

[0069] It is worth mentioning that in the above process, it is default that there is only one sensing device of each type. In fact, there may be multiple sensing devices of the same type. For example, multiple signal collectors are set in a TV set. At this time, each sensing device is regarded as a type of sensing device. That is to say, the type of sensing device corresponds one by one to the device number, and each device corresponds to a sensing device type.

[0070] Step S300: Determine the color value parameters of each device contour in real time according to the operation matrix, and update the home map at the corresponding time;

[0071] Each operation matrix represents the data generated by each home device during operation. By analyzing the operation matrix, it can be determined whether the working state of each home device is normal. In an example of the technical solution of the present invention, the parameter for evaluating whether it is normal is set as the color value parameter. For example, when the color value parameter adopts the gray value, the more normal the operation process of the home device is, the larger the gray value is, and the closer the corresponding device contour is to white and the brighter it is; Based on this, for the home map at each moment, determine the color value parameters of each device contour in the home map according to the obtained operation matrix, and then update the home map; After the above processing, the home map obtained in step S300 includes multiple device contours containing color value parameters.

[0072] Step S400: Determine the abnormal values at each position according to the home maps at each moment, and determine the image acquisition frequency at each position according to the abnormal values at each position; The devices used for image acquisition include but are not limited to fixed cameras and mobile cameras;

[0073] Analyze the home maps at each moment, and the analysis process can adopt a conventional image recognition scheme, which can quickly determine the abnormal degree at each position in the home map, which is represented by the parameter of the abnormal value. Determine the image acquisition frequency at each position according to the abnormal values at each position; In the technical solution of the present invention, a home camera is default equipped. The home camera can be a fixed camera, such as a camera installed on the ceiling and a camera installed on fixed devices (such as refrigerators and TVs), etc. The home camera can also be a mobile camera, such as a camera installed on a sweeping robot.

[0074] Step S500: Obtain the images at each position based on the image acquisition frequency, input them into the trained risk recognition model, and output the risk recognition result;

[0075] The image acquisition frequency at each position is different. The image acquisition frequency corresponding to the position with a higher outlier value is higher. Images at each position are obtained based on the image acquisition frequency. The more abnormal the position, the more images are obtained. The obtained images are input into a trained risk recognition model to output a risk recognition result. The ultimate goal of the technical solution of this application is actually the abnormal detection of the elderly. It is not actually complicated to determine whether there are risk behaviors of the elderly based on images. There are already related solutions in the prior art, and the detection targets are very limited, including whether there is a phenomenon of falling to the ground and sleeping for too long. To detect these targets, only a conventional positioning algorithm is needed to locate the elderly, and the movement speed of the elderly is calculated according to the positioning result. When the time when the movement speed is zero is greater than a preset time threshold, it is considered that a risk has occurred. Generally speaking, the risk recognition model is used to match a preset risk situation in the image and then output the matched risk situation.

[0076] Regarding step S100, the steps of obtaining the device position of the home appliance in real time and creating a device contour in the floor plan based on the device position to obtain a home map containing time include:

[0077] Obtain the device position containing time according to the locator built in the home appliance; the device position is the relative position based on the home origin; the home origin is the bottom endpoint of the center line of the door;

[0078] Read the floor plan, obtain the top view contour of the home appliance, and scale the top view contour of the home appliance according to the scale of the floor plan to obtain the device contour;

[0079] For any time, read the device positions at the nearest moment of all home appliances corresponding to this time, and insert the device contour at the device position in the floor plan to obtain the home map at this time.

[0080] In an example of the technical solution of the present invention, the device position containing time is obtained according to the locator built in the home appliance. The device position is generally the coordinate of the locator in the device, which is an absolute coordinate. To simplify the data, a home origin is preset in this application, and then the absolute coordinate is converted into a relative coordinate; the home origin generally uses the bottom endpoint of the center line of the door; then, read the floor plan, query the top view contour of the home appliance in the preset database, and scale the top view contour of the home appliance according to the scale of the floor plan to obtain the device contour; finally, for the time when the home map needs to be generated, read the device positions at the nearest moment of all home appliances corresponding to this time, and insert the device contour at the device position in the floor plan to obtain the home map at this time.

[0081] It should be noted that, as a preferred embodiment of the technical solution of the present invention, the process of obtaining the device position is optimized. Under normal circumstances, the acquisition frequencies of all device positions should be the same, but this is not the case in this application. For the process of obtaining the device position, the acquisition frequency of the device position is adjusted in real time according to the change situation of the device position. That is, for home devices with almost unchanged device positions, the acquisition frequency of their device positions will be very low, and the longer the immobile time, the lower the acquisition frequency. This process can greatly simplify the number of data uploads of the locator and reduce energy consumption.

[0082] Regarding step S200, the steps of obtaining the type of sensing device of the home device, using the type of sensing device as columns and time as rows to establish a basic matrix, and statistically analyzing the operation data of each home device based on the basic matrix to obtain an operation matrix include:

[0083] Obtain the sensing devices installed in each home device, query the types of the sensing devices, and arrange the types in a preset order as column labels;

[0084] Determine the acquisition moments according to a preset time period, and arrange the acquisition moments in chronological order as row labels;

[0085] Statistically analyze the column labels and row labels to obtain a basic matrix;

[0086] Receive the operation data containing time uploaded by each sensing device in the home device, and determine the row and column positions according to the sensing device and time;

[0087] Insert the operation data into the corresponding row and column positions in the basic matrix to obtain an operation matrix.

[0088] In an example of the technical solution of the present invention, the generation process of the operation matrix is described. Each home device needs to generate an operation matrix. Obtain the sensing devices installed in each home device, query the types of the sensing devices, and arrange the types in a preset order as column labels. Determine the acquisition moments according to a preset time period, and arrange the acquisition moments in chronological order as row labels. Statistically analyze the column labels and row labels to obtain a basic matrix; the rows of the obtained basic matrix represent time and the columns represent devices; when receiving the operation data containing time uploaded by each sensing device in the home device, determine the row and column positions according to the sensing device and time, and insert the operation data into the row and column positions to obtain an operation matrix; the total number of columns of the operation matrix obtained in this application is fixed because the number of sensing devices is fixed, and the total number of rows of the operation matrix is constantly increasing. As time goes by, the number of rows keeps increasing.

[0089] It is worth mentioning that when inserting the operation data into the row and column positions, the present application generally also introduces a process of dimensionless processing, that is, calculating the ratio of it to the maximum value of the historical operation data, so that all the values in the operation matrix are within the range of 0 to 1; if the maximum value of the historical operation data changes, all the data in the operation matrix need to be updated once.

[0090] Step S300, the step of determining the color value parameters of each device profile in real time according to the operation matrix and updating the home map at the corresponding moment includes:

[0091] For any home map, intercept a sub-matrix from the operation matrix according to a preset time period; the number of columns of the sub-matrix is the same as that of the operation matrix, and the number of rows of the sub-matrix is less than that of the operation matrix;

[0092] Perform periodic analysis on each column of data in the sub-matrix, and determine the color value parameters of each device profile according to the results of the periodic analysis;

[0093] Insert the color value parameters into the device profile in the home map as the updated home map.

[0094] In an example of the technical solution of the present invention, for any home map, read the time tag of the home map, locate the corresponding row in the operation matrix according to the time tag, and then use this row as the last row and read the row data forward; the number of rows of the row data read forward is determined by the preset time period, and each row of data corresponds to a time. Assuming that the time period is one day, then the row data within the previous day is read, and the row data read is a sub-matrix.

[0095] Perform periodic analysis on each column of data in the sub-matrix, and determine the color value parameters of each device profile according to the results of the periodic analysis. In practical applications, the habits of the users in the home scenario are stable. Reflected in the sensing devices of each home device, the data of the sensing devices have clear periodicity. Therefore, by performing periodic analysis on each column of data, it can be determined whether each device is in a normal state, and then the color value parameters are determined, and whether it is normal is represented by the color value parameters; finally, the color value parameters are inserted into the device profile in the home map to obtain the updated home map.

[0096] Further, the step of performing periodic analysis on each column of data in the sub-matrix and determining the color value parameters of each device profile according to the results of the periodic analysis includes:

[0097] Perform Fourier transform on each column of data to obtain frequency domain components;

[0098] Compare the frequency domain components with the preset standard components to determine the difference degree of each column of data;

[0099] Statistically analyze the difference degrees of all column data, and determine the color value parameters of each device contour according to the difference degrees.

[0100] The above content specifically describes the process of determining the color value parameters. In this application, a Fourier transform scheme is used to perform periodic analysis on each column of data. Fourier transform is performed on each column of data to obtain frequency domain components, and they are compared with preset standard components. The standard components are used to characterize the normal periodic characteristics of the sensing device. The process of determining them is as follows: Obtain the data of each sensing device in a day in daily life, perform Fourier transform on it to obtain a frequency domain graph, and select the most common frequency domain graph in the frequency domain graph as the standard component; The process of selecting the most common frequency domain graph can be to calculate the difference between each frequency domain graph and other frequency domain graphs, then calculate the average difference, and select the frequency domain graph with the smallest average difference as the most common frequency domain graph; It should be noted that the data for which Fourier transform is performed are all data within a preset time period, and their data volumes are the same, both being the number of rows.

[0101] Compare the frequency domain components with the preset standard components to determine the difference degrees of each column of data. Statistically analyze the difference degrees of all column data, calculate the average difference degree, and determine the color value parameters of each device contour according to the average difference degree. When the color value parameters use gray values, the gray value is inversely proportional to the average difference degree. That is, the larger the average difference degree, the more abnormal the working state of the device, and the smaller the gray value, the closer the corresponding visual state is to black.

[0102] Regarding step S400, the determining the abnormal values at each position according to the home map at each moment and determining the image acquisition frequency at each position according to the abnormal values at each position includes:

[0103] For the home map at any moment, spread the color value parameters of each device contour in the home map to obtain the color value parameters at each position in the home map;

[0104] Determine the abnormal values according to the color value parameters;

[0105] Determine the image acquisition frequency at each position according to the abnormal values;

[0106] Perform image acquisition on each position based on the image acquisition frequency;

[0107] Among them, the fixed camera includes a camera installed on a fixed device, and the mobile camera includes a camera installed on a mobile device.

[0108] In an example of the technical solution of the present invention, for the home map at any moment, the color value parameters of the device contours of the home map are diffused. The meaning of diffusion is that the farther a certain position is from the device contour, the smaller the influence of this position by the device contour. For any position, it is affected by multiple device contours at the same time. By synthesizing the influences of multiple devices, the outlier value of each position is obtained. When the color value parameter uses the gray value, the calculation process of the outlier value is as follows:

[0109] In the formula, Y(x,y) is the outlier value of the position (x,y), N is the total number of device contours, L i is the gray value of the i-th device contour, dis(c i ,(x,y)) is the distance between the center of the i-th device contour and the position (x,y); it should be noted that since 255 represents white and 0 represents black, the influence value of the black device contour on the surrounding is still 0. However, the above calculation process considers the influences of all device contours, and essentially calculates the average value of the influences of all device contours. In addition, there is a step of rounding up in the process of calculating the influence. When a certain device contour is not zero, the outlier values at other positions will not be zero.

[0110] In addition, the outlier value calculated above can also be used as a gray value to fill the home map.

[0111] Finally, the image acquisition frequency of each position is determined according to the outlier value, and image acquisition is performed on each position based on the image acquisition frequency. The image acquisition frequency is inversely proportional to the outlier value, that is, the smaller the outlier value, the higher the risk level, and the greater the image acquisition frequency; among them, the fixed camera includes the camera installed on the fixed device, and the mobile camera includes the camera installed on the mobile device. For the areas that can be monitored by certain fixed devices, the fixed devices are used, and for the areas that cannot be monitored, the mobile devices are used.

[0112] Figure 2 As shown in the block diagram of the composition structure of the intelligent home monitoring and warning system based on the Internet of Things sensing device, the technical solution of the present invention also provides an intelligent home monitoring and warning system based on the Internet of Things sensing device. The system 10 includes:

[0113] A home map acquisition module 11, which is used to obtain the device positions of home devices in real time, create device contours in the floor plan based on the device positions, and obtain a home map containing time;

[0114] An operation matrix generation module 12, which is used to obtain the type of sensing device of the home device, use the type of sensing device as the column and the time as the row to establish a basic matrix, and statistically obtain the operation data of each home device based on the basic matrix to obtain an operation matrix;

[0115] A home map update module 13, which is used to determine the color value parameters of each device contour in real time according to the operation matrix, and update the home map at the corresponding moment;

[0116] An acquisition frequency determination module 14, which is used to determine the outliers at each position according to the home map at each moment, and determine the image acquisition frequency at each position according to the outliers at each position; the devices used to acquire images include but are not limited to fixed cameras and mobile cameras;

[0117] A risk identification module 15, which is used to obtain the images at each position based on the image acquisition frequency, input the trained risk identification model, and output the risk identification result.

[0118] Furthermore, the home map acquisition module 11 includes:

[0119] A device position acquisition unit, which is used to acquire the device position containing time according to the locator built in the home device; the device position is the relative position based on the home origin; the home origin is the bottom endpoint of the center line of the door;

[0120] A contour scaling unit, which is used to read the house type drawing, obtain the top view contour of the home device, and scale the top view contour of the home device according to the scale of the house type drawing to obtain the device contour;

[0121] A contour insertion unit, which is used to, for any time, read the device positions of all home devices at the nearest moment corresponding to this time, and insert the device contour at the device position in the house type drawing to obtain the home map at this time;

[0122] Among them, for the acquisition process of the device position, the acquisition frequency of the device position is adjusted in real time according to the change situation of the device position.

[0123] Specifically, the operation matrix generation module 12 includes:

[0124] A column label generation unit, which is used to acquire the sensing devices installed in each home device, query the types of the sensing devices, and arrange the types in a preset order as the column labels;

[0125] A row label generation unit, which is used to determine the acquisition moments according to a preset time period, and arrange the acquisition moments in time sequence as the row labels;

[0126] A template generation unit, which is used to count the column labels and row labels to obtain the basic matrix;

[0127] A position determination unit, which is used to receive the operation data containing time uploaded by each sensing device in the home device, and determine the row and column positions according to the sensing device and time;

[0128] A data insertion unit for inserting operation data into corresponding row and column positions in a base matrix to obtain an operation matrix.

[0129] Furthermore, the home map update module 13 includes:

[0130] A matrix intercepting unit for intercepting a sub-matrix from the operation matrix according to a preset time period for any home map; the number of columns of the sub-matrix is the same as that of the operation matrix, and the number of rows of the sub-matrix is less than that of the operation matrix;

[0131] A period analysis unit for performing period analysis on each column of data in the sub-matrix and determining the color value parameters of each device profile according to the period analysis result;

[0132] A map output unit for inserting the color value parameters into the device profiles in the home map as the updated home map.

[0133] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention. Any equivalent structural or equivalent process transformation made by using the specification and drawings of the present invention, or directly or indirectly applied to other related technical fields, shall be equally included in the patent protection scope of the present invention.

Claims

1. A smart home monitoring and early warning method based on Internet of Things sensor equipment, characterized in that: The method comprises: Get the device location of home devices in real time, create device outlines in the floor plan based on the device location, and obtain a home map with time; Obtain the sensor device type of the home appliance, use the sensor device type as a column and the time as a row to establish a basic matrix, and count the operation data of each home appliance based on the basic matrix to obtain an operation matrix; Determine the color value parameters of each device outline in real time according to the operation matrix, and update the home map at the corresponding moment; Determine the abnormal value at each location according to the home map at each time, and determine the image acquisition frequency at each location according to the abnormal value at each location; wherein the equipment used for acquiring images includes but is not limited to fixed cameras and mobile cameras; Based on the image acquisition frequency, images at each location are acquired, the trained risk identification model is input, and the risk identification results are output.

2. The smart home monitoring and early warning method based on the Internet of Things sensor device according to claim 1 is characterized in that: The steps of acquiring the device positions of home devices in real time, creating device outlines in the floor plan based on the device positions, and obtaining a home map containing time include: Obtaining the device position including time according to the locator built into the home device; the device position is a relative position based on the home origin; the home origin is the bottom end point of the center line of the door; Read the floor plan to obtain the top view outline of the home equipment, and scale the top view outline of the home equipment according to the scale of the floor plan to obtain the equipment outline; At any time, read the device positions of all home devices corresponding to the time at the most recent moment, insert the device outlines into the device positions in the floor plan, and obtain the home map at the time; In the process of acquiring the device location, the frequency of acquiring the device location is adjusted in real time according to changes in the device location.

3. The smart home monitoring and early warning method based on the Internet of Things sensor device according to claim 1 is characterized in that: The steps of obtaining the sensor device type of the household device, taking the sensor device type as a column and the time as a row, establishing a basic matrix, and counting the operation data of each household device based on the basic matrix to obtain the operation matrix include: Obtain the sensor devices installed in each home device, query the types of the sensor devices, and arrange the types according to a preset order as column labels; Determine the collection time according to the preset time period, and arrange the collection time in chronological order as a row mark; Count the column and row labels to get the basic matrix; Receive the operation data including time uploaded by each sensor device in the home appliance, and determine the row and column position according to the sensor device and time; Insert the running data into the corresponding row and column positions in the basic matrix to obtain the running matrix.

4. The smart home monitoring and early warning method based on the Internet of Things sensor device according to claim 1 is characterized in that: The step of determining the color value parameters of each device outline in real time according to the operation matrix and updating the home map at the corresponding moment includes: For any home map, a sub-matrix is ​​obtained by cutting out the operation matrix according to a preset time period; the number of columns of the sub-matrix is ​​the same as the number of columns of the operation matrix, and the number of rows of the sub-matrix is ​​less than the number of rows of the operation matrix; Performing periodic analysis on each column of data in the submatrix, and determining the color value parameters of each device profile according to the periodic analysis results; The color value parameters are inserted into the device outline in the home map as an updated home map.

5. The smart home monitoring and early warning method based on the Internet of Things sensor device according to claim 4 is characterized in that: The step of performing periodic analysis on each column of data in the submatrix and determining the color value parameter of each device profile according to the periodic analysis result comprises: Perform Fourier transform on each column of data to obtain the frequency domain component; Compare the frequency domain components with the preset standard components to determine the difference of each column of data; The differences of all columns of data are counted, and the color value parameters of each device profile are determined according to the differences.

6. The smart home monitoring and early warning method based on the Internet of Things sensor device according to claim 1 is characterized in that: Determining the abnormal value at each position according to the home map at each time, and determining the image acquisition frequency at each position according to the abnormal value at each position includes: For the home map at any time, the color value parameters of the contours of each device in the home map are diffused to obtain the color value parameters of each position in the home map; Determining an outlier based on the color value parameter; Determine the image acquisition frequency at each location based on the outliers; Capturing images at various locations based on an image acquisition frequency; Among them, fixed cameras include cameras installed on fixed devices, and mobile cameras include cameras installed on mobile devices.

7. A smart home monitoring and early warning system based on Internet of Things sensor equipment, characterized in that: The system comprises: A home map acquisition module is used to obtain the device location of home devices in real time, create device outlines in the floor plan based on the device location, and obtain a home map with time; An operation matrix generation module is used to obtain the sensor device type of the household device, take the sensor device type as a column and the time as a row, establish a basic matrix, and count the operation data of each household device based on the basic matrix to obtain the operation matrix; A home map updating module, used to determine the color value parameters of each device outline in real time according to the operation matrix, and update the home map at the corresponding moment; An acquisition frequency determination module is used to determine the abnormal value at each location according to the home map at each time, and determine the image acquisition frequency at each location according to the abnormal value at each location; wherein the device used to acquire images includes but is not limited to a fixed camera and a mobile camera; The risk identification module is used to obtain images at various locations based on the image acquisition frequency, input the trained risk identification model, and output the risk identification results.

8. The smart home monitoring and early warning system based on the Internet of Things sensor device according to claim 7 is characterized in that: The home map acquisition module includes: A device position acquisition unit, used to acquire the device position including time according to a locator built into the home device; the device position is a relative position based on the home origin; the home origin is the bottom end point of the center line of the door; A contour scaling unit is used to read the floor plan, obtain the top view contour of the home appliance, and scale the top view contour of the home appliance according to the scale of the floor plan to obtain the device contour; The contour insertion unit is used to read the device positions of all household devices corresponding to any time at the latest moment, insert the device contours into the device positions in the floor plan, and obtain the home map at that time; In the process of acquiring the device location, the frequency of acquiring the device location is adjusted in real time according to changes in the device location.

9. The smart home monitoring and early warning system based on the Internet of Things sensor device according to claim 7 is characterized in that: The operation matrix generation module comprises: A column label generating unit is used to obtain the sensor devices installed in each household device, query the types of the sensor devices, and arrange the types according to a preset order as column labels; A line mark generating unit, used to determine the collection time according to a preset time period, and arrange the collection time in chronological order as a line mark; A template generation unit is used to count column labels and row labels to obtain a basic matrix; A position determination unit, used to receive the operation data including time uploaded by each sensor device in the home appliance, and determine the row and column position according to the sensor device and time; The data insertion unit is used to insert the operation data into the corresponding row and column positions in the basic matrix to obtain the operation matrix.

10. The smart home monitoring and early warning system based on the Internet of Things sensor device according to claim 7 is characterized in that: The home map updating module includes: A matrix interception unit is used to intercept a sub-matrix in the operation matrix according to a preset time period for any home map; the number of columns of the sub-matrix is ​​the same as the number of columns of the operation matrix, and the number of rows of the sub-matrix is ​​less than the number of rows of the operation matrix; A cycle analysis unit, used for performing a cycle analysis on each column of data in the submatrix, and determining a color value parameter of each device profile according to the cycle analysis result; The map output unit is used to insert the color value parameters into the device outline in the home map as an updated home map.

Images