An intelligent door lock monitoring and management system and method based on cloud computing

Abstract

The present invention discloses an intelligent door lock monitoring and management system and method based on cloud computing, including an alarm information acquisition module, a first image data acquisition module, a change trend analysis module, a subjectivity model establishment module, and an alarm signal regulation module; the alarm information acquisition module is used to acquire alarm signal transmission information; the first image data acquisition module is used to acquire image data with a human figure within a first time threshold before the alarm signal occurs; the change trend analysis module is used to analyze the change trend of the rotation angle each time the intelligent door lock is used, and analyze the relationship between the actual stay duration and the average stay duration when there is no change trend; the subjectivity model establishment module is used to analyze and establish a subjectivity model when there is a change trend; the alarm signal regulation module adjusts whether to output an alarm signal based on the results of the change trend analysis module and the subjectivity model establishment module.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent door locks, and particularly to an intelligent door lock monitoring and management system and method based on cloud computing. Background Technique

[0002] At present, the installation of intelligent door locks has basically been implemented in every household. The security problem of intelligent door locks has become the primary concern of each user. Therefore, an alarm function has emerged in the gradually upgraded functions of intelligent door locks. This alarm function can transmit an alarm sound to remind the household to close the door in time when the door is not closed tightly; however, other problems have also arisen based on this situation. For example, when the household makes a door-closing action but subjectively does not want the door to be closed tightly, such as when the household goes downstairs to sign for a courier and leaves the door ajar in daily life, it can enable the user to quickly enter the house when they have no free hand to open the door. But at this time, due to the alarm function of the intelligent door lock, the alarm sound will keep ringing, causing noise interference to the surrounding households. At the same time, when evaluating the service life of the intelligent door lock, it is impossible to accurately obtain the objective uniqueness of the number of alarm occurrences of the intelligent door lock. Summary of the Invention

[0003] The purpose of the present invention is to provide an intelligent door lock monitoring and management system and method based on cloud computing to solve the problems raised in the above background technique.

[0004] To solve the above technical problems, the present invention provides the following technical solution: An intelligent door lock monitoring and management method based on cloud computing, including the following steps:

[0005] Obtain the alarm signal transmission information of the intelligent door lock within the monitoring period. The alarm signal transmission information includes the image data in the intelligent image acquisition device connected to the intelligent door lock and the conventional data of the intelligent door lock. The conventional data includes the usage time of the intelligent door lock and the rotation angle of the installation carrier of the intelligent door lock;

[0006] Obtain the image data with a human figure within the first time threshold before the alarm signal occurs, denoted as the first image data; extract the rotation angle of the installation carrier of the intelligent door lock when the first image data is generated. The rotation angle is the included angle formed by the real-time position of the installation carrier of the intelligent door lock after the household opens the door lock and the position of the installation carrier in the closed state; obtain the change trend of the rotation angle each time the intelligent door lock is used. The change trend is the angle change trend after the rotation angle reaches the maximum value of this rotation.

[0007] When it is monitored that there is no change trend within the first time threshold, obtain the actual residence duration t1 of the rotation angle, and the average residence duration t0 of the intelligent door lock within the monitoring period. The average residence duration is the average residence duration when the intelligent door lock has no angle change after reaching the maximum rotation angle each time; when t1 > t0, an alarm signal is output after t0; when t1 ≤ t0, the alarm signal is turned off; because when the angle of the door does not change after being opened to the maximum, it indicates that the household has a certain purpose in maintaining the angle of the door. Analyze the average residence duration of the household in the historical data, and there may be a possibility of forgetting to close the door when it exceeds, so an alarm is given to remind the user to close the door. At the same time, analyze the data of the monitoring period. Even if there is a misjudgment, the data of the previous time will be recorded during the next door opening analysis, so that the average residence duration of the household when opening the door will be more accurate as the data increases;

[0008] When it is monitored that there is a change trend within the first time threshold, analyze the first image data and the conventional data of the intelligent door lock, and establish a subjective degree model for the household to close the door lock; the existence of a change trend indicates that the household's action of closing the door causes a change in the rotation angle of the door, but the result is that the door is not closed and an alarm signal is output. Therefore, it is necessary to analyze whether the household's action of not closing the door is due to subjective factors that the household wants to leave a gap or due to objective factors of the door lock itself that causes it not to be closed.

[0009] Based on the subjective degree model of the household to close the door lock, regulate whether to output the alarm signal. Regulating according to the model is to reduce the mistransmission rate of the alarm signal, avoid noise interference caused by the alarm sounding in the case of unnecessary reminders to the household, and determine the objective uniqueness of the origin of the alarm transmission, which is convenient for merchants to repair the objective faults existing in the intelligent door lock.

[0010] Furthermore, analyze the first image data and the conventional data of the intelligent door lock, and establish a subjective degree model for the household to close the door lock, including the following steps:

[0011] Obtain the opening time T0 of the intelligent door lock, the time T1 when the intelligent door lock transmits the alarm signal, the time T2 when the household closes the door lock after the alarm signal sounds, and the time T3 when it is opened again within the second time threshold after closing; use the formula:

[0012]

[0013] Calculate the preliminary response degree x of the household, where T10 represents the average alarm signal transmission interval duration of the household in the monitoring period, T21 represents the average reaction closing duration of the household after the alarm signal is transmitted in the monitoring period, and T32 represents the average interval duration between two adjacent door openings of the household in the monitoring period; if there is no re-opening within the second time threshold after closing, then let T3 - T2 = 0;

[0014] The preliminary response degree reflects the response actions of the household after hearing the alarm sound. Since when the household fails to close the door due to door lock problems, they should quickly close the door after hearing the alarm reminder and the probability of opening the door again within a short time is relatively small. However, if the user deliberately does not close the door tightly because they want to leave a crack, they may not choose to close the door immediately after hearing the alarm sound. If they choose to close the door immediately, the possibility of closing the door again is greater because leaving a crack is a subjective intention, indicating that the household has activities in the open door state. Therefore, a preliminary judgment can be made on the differences in the alarm of the door lock by the user from the difference relationship at different times;

[0015] Set the preliminary response degree threshold x0, extract the first image data corresponding to x≥x0, denoted as the target image data, and extract the first image data corresponding to x<x0, denoted as the comparison image data;

[0016] Further construct a subjective degree model for the household to close the door lock based on the target image data and the comparison image data.

[0017] Furthermore, constructing a subjective degree model for the household to close the door lock based on the target image data and the comparison image data includes the following specific steps:

[0018] Locate the images from the start of the household's door lock closing action to the completion of the door lock closing action in the target image data and the comparison image data respectively. This process is the stage process, and obtain the durations h1 and h2 experienced in the stage process, as well as the actual moving distances g1 and g2 of the intelligent door lock installation carrier during this process. The actual moving distance takes any point on the intelligent door lock as the target point, and the straight-line distance that the target point moves from the start of the household's door lock closing action to the completion of the door lock closing action by the intelligent door lock installation carrier;

[0019] Calculate the subjective action rate of the household V = G / H, V = {v1, v2}, G = {g1, g2}, H = {h1, h2}, where v1 represents the subjective action rate of the household in the target image data, and v2 represents the subjective action rate of the household in the comparison image data;

[0020] Obtain the average subjective action rate v0 of the household during the monitoring period, and calculate the subjective action index y = |v10 - v20| / v0, where v10 represents the average subjective action rate of the household in the target image data, and v20 represents the average subjective action rate of the household in the comparison image data;

[0021] Set the subjective action index threshold y0. If y≥y0, establish a first-level subjective degree model z1, z1 = [v1min, v1max], where v1min represents the minimum value of the subjective action rate of the household in the target image data, and v1max represents the maximum value of the subjective action rate of the household in the target image data; if y<y0, establish a second-level subjective degree model z2.

[0022] Further, establishing the secondary subjective degree model z2 includes the following specific steps:

[0023] Establish the angular relationship of the human image in the image data. The angular relationship is the angle size between line segment one formed by the connection line between the hand and the elbow with the hand, elbow, and shoulder as fixed points, and line segment two formed by the connection line between the elbow and the shoulder;

[0024] Extract the angular relationship of the target image data as the target angle, and the angular relationship of the comparison image data as the comparison angle;

[0025] Equalize the stage processes of the target image data and the comparison image data based on the straight-line distance, obtain the target image data corresponding to the equal division point I as the equal division target image data, and the comparison image data corresponding to the equal division point I as the equal division comparison image data. Calculate the difference si of the target angles in n adjacent equal division target image data and the difference ci of the comparison angles in n adjacent equal division comparison image data, i = {1, 2,..., n - 1}, where i represents the number of target angle differences;

[0026] Fit the difference si of the target angles and the equal division point I to obtain data set one (si, I(i + 1)), where I(i + 1) represents the (i + 1)-th equal division point. Obtain data set one under different equal division points and fit to generate curve one of the difference between the equal division point and the target angle; Fit the difference ci of the comparison angles and the equal division point I to obtain data set two (ci, I(i + 1)), and obtain data set two under different equal division points and fit to generate curve two of the difference between the equal division point and the comparison angle;

[0027] Compare the similarity of curve one and curve two. When the similarity is less than the similarity threshold, establish the secondary subjective degree model z2 as the set of all target angles that satisfy the deviation threshold of curve one. The deviation threshold is a value between 0 and the difference between the maximum and minimum values of the corresponding differences of curve one.

[0028] Further, based on the subjective degree model of the household closing the door lock, regulate whether to output the alarm signal, including the following steps:

[0029] Set the priority of the primary subjective degree model to be higher than that of the secondary subjective degree model. When the real-time monitored data of the household meets the primary subjective degree model, that is, when the real-time subjective action rate of the household belongs to z1, turn off the alarm signal; Analyze the primary subjective degree model first because the data obtained from the analysis rate is relatively simple and convenient for quick analysis;

[0030] When the real-time subjective action rate of the household does not belong to z1, calculate the real-time subjective action index y1 by combining historical data. If y1 still satisfies y1≥y0, update the primary subjective degree model z1 and turn off the alarm signal; if y1<y0, then judge the secondary subjective degree model.

[0031] If there is a secondary subjective degree model and the angle relationship of the real-time monitored household satisfies the secondary subjective degree model, turn off the alarm signal; if the angle relationship of the real-time monitored household does not satisfy the secondary subjective degree model or there is no secondary subjective degree model, then continue to transmit the alarm signal.

[0032] An intelligent door lock monitoring and management system based on cloud computing, including an alarm information acquisition module, a first image data acquisition module, a change trend analysis module, a subjective degree model establishment module, and an alarm signal regulation module;

[0033] The alarm information acquisition module is used to acquire alarm signal transmission information;

[0034] The first image data acquisition module is used to acquire image data with a human figure within the first time threshold before the alarm signal occurs;

[0035] The change trend analysis module is used to analyze the change trend of the rotation angle each time the intelligent door lock is used, and analyze the relationship between the actual stay duration and the average stay duration when there is no change trend, and transmit a regulation signal to the alarm signal regulation module;

[0036] The subjective degree model establishment module is used to analyze the first image data and the conventional data of the intelligent door lock and establish a subjective degree model when there is a change trend, and transmit a regulation signal to the alarm signal regulation module;

[0037] The alarm signal regulation module adjusts whether to output the alarm signal based on the results of the change trend analysis module and the subjective degree model establishment module.

[0038] Further, the subjective degree model establishment module includes an image data discrimination unit, a primary subjective degree model establishment unit, and a secondary subjective degree model establishment unit;

[0039] The image data discrimination unit is used to analyze the preliminary response degree of the household to close the door lock and divide the image data into target image data and comparison image data;

[0040] The primary subjective degree model establishment unit is used to analyze the subjective action rate of the household and establish a primary subjective degree model;

[0041] The secondary subjective degree model establishment unit is used to analyze the angle relationship in the household image data and establish a secondary subjective degree model.

[0042] Further, the alarm signal control module includes a priority setting unit, a model updating unit, and a signal output unit;

[0043] The priority setting unit is used to set the priority of the first-level subjective degree model higher than that of the second-level subjective degree model;

[0044] The model updating unit is used to update the first-level subjective degree model during the analysis of the first-level subjective degree model;

[0045] The signal output unit is used to output an alarm signal for the situation that meets the conditions during the judgment process.

[0046] Compared with the prior art, the beneficial effects achieved by the present invention are as follows: Based on the responses of residents after the intelligent door lock makes an alarm sound, the present invention classifies residents according to different subjective reactions, and then further analyzes the behavior actions of residents towards the intelligent door lock in different situations to establish a subjective degree model for double-layer judgment; enabling the intelligent image acquisition device connected to the intelligent door lock to match with the subjective degree model after capturing the corresponding information, and quickly making a response analysis result to control whether the alarm information is transmitted. The present invention reduces the mistransmission rate of alarm signals, avoids noise interference caused by unnecessary alarm sounds to remind residents, and determines the objective uniqueness of the origin of alarm transmission, facilitating merchants to repair the objective existing fault problems of the intelligent door lock. Description of the Drawings

[0047] The drawings are used to provide a further understanding of the present invention and constitute a part of the specification. They are used together with the embodiments of the present invention to explain the present invention, but do not constitute a limitation to the present invention. In the drawings:

[0048] Figure 1 is a schematic structural diagram of an intelligent door lock monitoring and management system based on cloud computing according to the present invention. Detailed Embodiments

[0049] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0050] Please refer to Figure 1 , the present invention provides a technical solution: An intelligent door lock monitoring and management method based on cloud computing, including the following steps:

[0051] Obtain the alarm signal transmission information of the intelligent door lock during the monitoring period. The alarm signal transmission information includes the image data in the intelligent image acquisition device connected to the intelligent door lock and the conventional data of the intelligent door lock. The conventional data includes the usage time of the intelligent door lock and the rotation angle of the installation carrier of the intelligent door lock.

[0052] Obtain the image data with a human figure within the first time threshold before the alarm signal occurs, which is recorded as the first image data. Extract the rotation angle of the installation carrier of the intelligent door lock when the first image data is generated. The rotation angle is the included angle formed by the real-time position of the installation carrier of the intelligent door lock after the household opens the door and the position of the installation carrier in the closed state. Obtain the change trend of the rotation angle each time the intelligent door lock is used. The change trend is the angle change trend after the rotation angle reaches the maximum value of this rotation. The presence of a human figure indicates that the present invention analyzes the situation where the household forgets to close the door tightly after going out. The first time threshold is the time period from when the household closes the door to when they leave before the alarm signal occurs, and it is set and adjusted according to the duration of the household closing the door.

[0053] When monitoring that there is no change trend within the first time threshold, obtain the actual stay duration t1 of the rotation angle, and the average stay duration t0 of the intelligent door lock during the monitoring period. The average stay duration is the average stay duration when the intelligent door lock has no angle change after reaching the maximum value of the rotation angle each time. When t1 > t0, an alarm signal is output after t0. When t1 ≤ t0, the alarm signal is turned off. Because when the angle of the door does not change after being opened to the maximum, it indicates that the household has a certain purpose in maintaining the angle of the door. Analyze the average stay duration of the household in the historical data. When it exceeds, there may be a possibility of forgetting to close the door, and an alarm is given to remind the user to close the door. At the same time, analyze the data of the monitoring period. Even if there is a misjudgment, the data of the previous time will be recorded during the next door opening analysis, so that the average stay duration of the household when opening the door will be more accurate as the data increases.

[0054] When monitoring that there is a change trend within the first time threshold, analyze the first image data and the conventional data of the intelligent door lock, and establish a subjective degree model for the household to close the door lock. The presence of a change trend indicates that the household's action of closing the door causes a change in the rotation angle of the door, but the result is that the door is not closed and an alarm signal is output. Therefore, it is necessary to analyze whether the household's action of not closing the door is due to subjective factors that the household wants to leave a gap or due to objective factors such as problems with the door lock itself.

[0055] Based on the subjective degree model of the household closing the door lock, regulate whether to output the alarm signal. Regulating according to the model is to reduce the false transmission rate of the alarm signal, avoid noise interference caused by unnecessary alarm sounds when reminding the household, and determine the objective uniqueness of the origin of the alarm transmission, which is convenient for merchants to repair the objectively existing fault problems of the intelligent door lock.

[0056] Analyzing the first image data and the conventional data of the smart door lock to establish a subjective degree model of the resident closing the door lock includes the following steps:

[0057] Get the time T0 when the smart door lock is opened, the time T1 when the smart door lock transmits the alarm signal, the time T2 when the resident closes the door lock after the alarm signal sounds, and the time T3 when the door lock is opened again within the second time threshold after closing; use the formula:

[0058]

[0059] Calculate the initial responsiveness x of the resident, where T10 represents the average alarm signal transmission interval of the resident during the monitoring period, T21 represents the average reaction closing time of the resident after the alarm signal is transmitted during the monitoring period, and T32 represents the average interval between two consecutive door openings of the resident during the monitoring period; if there is no re-opening within the second time threshold after closing, set T3-T2=0; all the above time data belong to the usage time of the smart door lock in the conventional data of the smart door lock; the second time threshold can refer to the resident opening the door of the smart door lock again within five minutes after hearing the alarm and choosing to close the door, indicating that the resident closes the door to stop the alarm, but in fact the resident has the purpose of opening and closing the door, such as moving things and wanting to keep the door open;

[0060] The initial response reflects the reaction of the residents after hearing the alarm. If the residents do not close the door due to door lock problems, they should close the door quickly after hearing the alarm and the probability of opening the door again in a short period of time is low. If the user deliberately does not close the door tightly because he wants to leave a gap, he may not choose to close the door immediately after hearing the alarm. If he chooses to close the door immediately, he is more likely to close it again. Because leaving a gap is a subjective intention, it means that the residents have kept the door open. Therefore, the difference relationship at different times can make a preliminary judgment on the difference between the user's door lock alarm.

[0061] Set a preliminary response threshold x0, extract the first image data corresponding to when x≥x0, record it as target image data, extract the first image data corresponding to when x<x0, record it as comparison image data;

[0062] A subjective degree model of residents closing door locks is further constructed based on the target image data and the comparison image data.

[0063] Based on the target image data and the comparison image data, a subjective degree model of the resident closing the door lock is further constructed, including the following specific steps:

[0064] Locate the images of the process from when the household starts to close the door lock to when the household finishes closing the door lock in the target image data and the comparison image data respectively as a stage process, and obtain the durations h1 and h2 experienced by the stage process, as well as the actual moving distances g1 and g2 of the intelligent door lock installation carrier during this process. The actual moving distance takes any point on the intelligent door lock as the target point, and it is the straight-line distance that the target point moves when the intelligent door lock installation carrier moves from when the household starts to close the door lock to when the household finishes closing the door lock.

[0065] Calculate the subjective action rate V of the household, V = G / H, V = {v1, v2}, G = {g1, g2}, H = {h1, h2}, where v1 represents the subjective action rate of the household in the target image data, and v2 represents the subjective action rate of the household in the comparison image data.

[0066] Obtain the average subjective action rate v0 of the household during the monitoring period, and calculate the subjective action index y = |v10 - v20| / v0, where v10 represents the average subjective action rate of the household in the target image data, and v20 represents the average subjective action rate of the household in the comparison image data.

[0067] Set the subjective action index threshold y0. If y ≥ y0, establish a first-level subjective degree model z1, z1 = [v1min, v1max], where v1min represents the minimum value of the subjective action rate of the household in the target image data, and v1max represents the maximum value of the subjective action rate of the household in the target image data. If y < y0, establish a second-level subjective degree model z2. Analyzing the subjective action rate is because when the household has a subjective intention, there is a certain control over the door, and the time when the hand leaves the door will be later than the time when the hand leaves the door in the case of closing the door casually without awareness, which has a certain degree of discrimination.

[0068] Establishing the second-level subjective degree model z2 includes the following specific steps:

[0069] Establish the angular relationship of the human image in the image data. The angular relationship is the angle size between line segment one formed by the connection between the hand and the elbow and line segment two formed by the connection between the elbow and the shoulder, with the hand, elbow, and shoulder as fixed points.

[0070] Extract the angular relationship of the target image data as the target angle, and the angular relationship of the comparison image data as the comparison angle.

[0071] Equalize the stage processes of the target image data and the comparison image data based on the straight-line distance. Obtain the target image data corresponding to the equal division point I as the equal division target image data, and the comparison image data corresponding to the equal division point I as the equal division comparison image data. Calculate the difference si of the target angle in n adjacent equal division target image data and the difference ci of the comparison angle in n adjacent equal division comparison image data, where i = {1, 2,..., n - 1}, and i represents the number of target angle differences.

[0072] Fit the difference si of the target angle and the equal division point I to obtain the data set one (si, I(i + 1)), where I(i + 1) represents the (i + 1)-th equal division point. Obtain the data sets one under different equal division points and fit them to generate the curve one of the difference between the equal division point and the target angle. Since si represents adjacent values, the first difference should correspond to the difference between the data of the second equal division point and the first equal division point. Fit the difference ci of the comparison angle and the equal division point I to obtain the data set two (ci, I(i + 1)), obtain the data sets two under different equal division points and fit them to generate the curve two of the difference between the equal division point and the comparison angle.

[0073] Compare the similarity between curve one and curve two. When the similarity is less than the similarity threshold, establish the secondary subjective degree model z2 as the set of all target angles that satisfy the deviation threshold of curve one. The deviation threshold is a value between 0 and the difference between the maximum and minimum values of the corresponding differences of curve one.

[0074] For example, if the maximum value of curve one is (12°, equal division point 3) and the minimum value is (9°, equal division point 5), then the deviation threshold is [0, 12 - 9] = [0, 3]. Therefore, in the fitting curve formed by the angular relationship of real-time monitoring of the household, the angular relationships corresponding to the fitting curves whose angular deviation thresholds satisfy [0, 3] can all illustrate the subjective intention of the household.

[0075] Because the fitting curve is composed of the angular relationships corresponding to positions, representing a series of behavioral actions of the household closing the door. When the deviation threshold is smaller, it means that the actions of the household are smoother and more uniform. In combination with real life, when the door is not closed during the door closing process, the household will control the strength and speed subjectively when leaving a gap in the door. Therefore, the actions reflected by the curves that satisfy the deviation threshold can all reflect the subjective control of the household.

[0076] Based on the subjective degree model of the household closing the door lock, regulate whether to output the alarm signal, including the following steps:

[0077] Set the priority of the primary subjective degree model to be higher than that of the secondary subjective degree model. When the real-time monitored data of the household satisfies the primary subjective degree model, that is, when the real-time subjective action rate of the household belongs to z1, turn off the alarm signal. Analyze the primary subjective degree model first because the data obtained by analyzing the rate is relatively simple and convenient for quick analysis.

[0078] When the real-time subjective action rate of the household does not belong to z1, calculate the real-time subjective action index y1 by combining historical data. If y1 still satisfies y1≥y0, update the primary subjective degree model z1 and turn off the alarm signal; if y1<y0, then judge the secondary subjective degree model.

[0079] If there is a secondary subjective degree model and the real-time monitored angle relationship of the household satisfies the secondary subjective degree model, turn off the alarm signal; if the real-time monitored angle relationship of the household does not satisfy the secondary subjective degree model or there is no secondary subjective degree model, then continue to transmit the alarm signal.

[0080] An intelligent door lock monitoring and management system based on cloud computing, including an alarm information acquisition module, a first image data acquisition module, a change trend analysis module, a subjective degree model establishment module, and an alarm signal regulation module;

[0081] The alarm information acquisition module is used to acquire alarm signal transmission information;

[0082] The first image data acquisition module is used to acquire image data with a human figure within the first time threshold before the alarm signal occurs;

[0083] The change trend analysis module is used to analyze the change trend of the rotation angle each time the intelligent door lock is used, and analyze the relationship between the actual stay duration and the average stay duration when there is no change trend, and transmit a regulation signal to the alarm signal regulation module;

[0084] The subjective degree model establishment module is used to analyze the first image data and the conventional data of the intelligent door lock and establish a subjective degree model when there is a change trend, and transmit a regulation signal to the alarm signal regulation module;

[0085] The alarm signal regulation module adjusts whether to output the alarm signal based on the results of the change trend analysis module and the subjective degree model establishment module.

[0086] The subjective degree model establishment module includes an image data discrimination unit, a primary subjective degree model establishment unit, and a secondary subjective degree model establishment unit;

[0087] The image data discrimination unit is used to analyze the preliminary response degree of the household to close the door lock and divide the image data into target image data and comparison image data;

[0088] The primary subjective degree model establishment unit is used to analyze the subjective action rate of the household and establish a primary subjective degree model;

[0089] The secondary subjective degree model establishment unit is used to analyze the angle relationship in the household image data and establish a secondary subjective degree model.

[0090] The alarm signal regulation module includes a priority setting unit, a model updating unit, and a signal output unit;

[0091] The priority setting unit is used to set the priority of the first-level subjective model to be higher than that of the second-level subjective model;

[0092] The model updating unit is used to update the first-level subjective model during the analysis of the first-level subjective model;

[0093] The signal output unit is used to output an alarm signal for the situation that meets the conditions during the judgment process.

[0094] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device.

[0095] Finally, it should be noted that the above are only preferred embodiments of the present invention and are not used to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An intelligent door lock monitoring and management method based on cloud computing, characterized in that, Including the following steps: Obtain the alarm signal transmission information of the intelligent door lock within the monitoring period. The alarm signal transmission information includes the image data in the intelligent image acquisition device connected to the intelligent door lock and the conventional data of the intelligent door lock. The conventional data includes the usage time of the intelligent door lock and the rotation angle of the installation carrier of the intelligent door lock; Obtain the image data with a human figure within the first time threshold before the alarm signal occurs, denoted as the first image data; extract the rotation angle of the installation carrier of the intelligent door lock when the first image data is generated; obtain the change trend of the rotation angle each time the intelligent door lock is used. The change trend is the angle change trend after the rotation angle reaches the maximum value of this rotation; When it is monitored that there is no change trend within the first time threshold, obtain the actual stay duration t1 of the rotation angle and the average stay duration t0 of the intelligent door lock within the monitoring period. The average stay duration is the average stay duration when the intelligent door lock has no angle change after reaching the maximum value of the rotation angle each time; When t1 > t0, output an alarm signal after t0; when t1 ≤ t0, turn off the alarm signal; When it is monitored that there is a change trend within the first time threshold, analyze the first image data and the conventional data of the intelligent door lock, and establish a subjective degree model for the household to close the door lock; Based on the subjective degree model of the household to close the door lock, regulate whether to output the alarm signal.

2. The intelligent door lock monitoring and management method based on cloud computing according to claim 1, wherein: The analysis of the first image data and the conventional data of the intelligent door lock to establish a subjective degree model for the household to close the door lock includes the following steps: Obtain the opening time T0 of the intelligent door lock, the time T1 when the intelligent door lock transmits the alarm signal, the time T2 when the household closes the door lock after the alarm signal sounds, and the time T3 when it is opened again within the second time threshold after closing; use the formula: Calculate the preliminary response degree x of the household, where T10 represents the average alarm signal transmission interval duration of the household in the monitoring period, T21 represents the average reaction door closing duration of the household after the alarm signal is transmitted in the monitoring period, and T32 represents the average interval duration between two adjacent door openings of the household in the monitoring period; if there is no re-opening within the second time threshold after closing, then let T3 - T2 = 0; Set a preliminary response threshold x0, extract the first image data corresponding to x ≥ x0, denoted as the target image data, and extract the first image data corresponding to x < x0, denoted as the comparison image data; Further construct a subjective degree model for the household to close the door lock based on the target image data and the comparison image data.

3. The intelligent door lock monitoring and management method based on cloud computing according to claim 2, characterized in that: The further construction of a subjective degree model for the household to close the door lock based on the target image data and the comparison image data includes the following specific steps: Locate the images from the image where the household starts the door closing action to the image where the household completes the door closing action in the target image data and the comparison image data respectively as a stage process, and obtain the duration h1 and h2 experienced in the stage process, as well as the actual moving distance g1 and g2 of the installation carrier of the intelligent door lock during this process. The actual moving distance is the straight-line distance that a target point on the intelligent door lock moves from the time when the household starts the door closing action to the time when the household completes the door closing action; Calculate the subjective action rate of the household V = G / H, V = {v1, v2}, G = {g1, g2}, H = {h1, h2}, where v1 represents the subjective action rate of the household in the target image data, and v2 represents the subjective action rate of the household in the comparison image data; Obtain the average subjective action rate v0 of the household during the monitoring period, and calculate the subjective action index y of the household as y = |v10 - v20| / v0, where v10 represents the average subjective action rate of the household in the target image data, and v20 represents the average subjective action rate of the household in the comparison image data; Set the subjective action index threshold y0. If y ≥ y0, establish a first-level subjective degree model z1, z1 = [v1min, v1max], where v1min represents the minimum subjective action rate of the household in the target image data, and v1max represents the maximum subjective action rate of the household in the target image data; if y < y0, establish a second-level subjective degree model z2.

4. The intelligent door lock monitoring and management method based on cloud computing according to claim 3, characterized in that: The establishment of the second-level subjective degree model z2 includes the following specific steps: Establish the angular relationship of the human image in the image data. The angular relationship is the angle size between line segment one formed by the connection between the hand and the elbow and line segment two formed by the connection between the elbow and the shoulder, with the hand, elbow, and shoulder as fixed points. Extract the angular relationship of the target image data as the target angle, and the angular relationship of the comparison image data as the comparison angle; Equalize the stage processes of the target image data and the comparison image data based on the straight-line distance, obtain the target image data corresponding to the equal division point I as the equal division target image data, and the comparison image data corresponding to the equal division point I as the equal division comparison image data. Calculate the difference si of the target angles in n adjacent equal division target image data and the difference ci of the comparison angles in n adjacent equal division comparison image data, i = {1, 2,..., n - 1}, and i represents the number of target angle differences. Fit the difference si of the target angles and the equal division point I to obtain data set one (si, I(i + 1)), where I(i + 1) represents the (i + 1)-th equal division point. Obtain data set one under different equal division points and fit it to generate curve one of the difference between the equal division point and the target angle; fit the difference ci of the comparison angles and the equal division point I to obtain data set two (ci, I(i + 1)), obtain data set two under different equal division points and fit it to generate curve two of the difference between the equal division point and the comparison angle; Compare the similarity of curve one and curve two. When the similarity is less than the similarity threshold, establish the second-level subjective degree model z2 as the set of all target angles that satisfy the deviation threshold of curve one. The deviation threshold is a value between 0 and the difference between the maximum and minimum values of the corresponding differences of curve one.

5. The intelligent door lock monitoring and management method based on cloud computing according to claim 4, characterized in that: Based on the subjective degree model of the household closing the door lock, regulate whether to output the alarm signal, including the following steps: Set the priority of the first-level subjective degree model to be higher than that of the second-level subjective degree model. When the real-time monitored data of the household satisfies the first-level subjective degree model, that is, when the real-time subjective action rate of the household belongs to z1, turn off the alarm signal; When the real-time subjective action rate of the household does not belong to z1, calculate the real-time subjective action index y1 in combination with historical data. If y1 still satisfies y1 ≥ y0, update the first-level subjective degree model z1 and turn off the alarm signal; if y1 < y0, then judge the second-level subjective degree model; When there is a secondary subjective degree model and the angle relationship of the monitored household satisfies the secondary subjective degree model in real time, the alarm signal is turned off; when the angle relationship of the monitored household does not satisfy the secondary subjective degree model or there is no secondary subjective degree model in real time, the alarm signal is continuously transmitted.

6. A cloud computing-based intelligent door lock monitoring and management system applying the cloud computing-based intelligent door lock monitoring and management method according to any one of claims 1-5, characterized in that, It includes an alarm information acquisition module, a first image data acquisition module, a change trend analysis module, a subjective degree model establishment module, and an alarm signal regulation module; The alarm information acquisition module is used to acquire alarm signal transmission information; The first image data acquisition module is used to acquire image data with a human figure within the first time threshold before the alarm signal occurs; The change trend analysis module is used to analyze the change trend of the rotation angle each time the intelligent door lock is used, and analyze the relationship between the actual stay duration and the average stay duration when there is no change trend, and transmit a regulation signal to the alarm signal regulation module; The subjective degree model establishment module is used to analyze the first image data and the conventional data of the intelligent door lock and establish a subjective degree model when there is a change trend, and transmit a regulation signal to the alarm signal regulation module; The alarm signal regulation module adjusts whether to output the alarm signal based on the results of the change trend analysis module and the subjective degree model establishment module.

7. An intelligent door lock monitoring and management system based on cloud computing according to claim 6, characterized in that: The subjective degree model establishment module includes an image data discrimination unit, a primary subjective degree model establishment unit, and a secondary subjective degree model establishment unit; The image data discrimination unit is used to analyze the initial response degree of the household to close the door lock and divide the image data into target image data and comparison image data; The primary subjective degree model establishment unit is used to analyze the subjective action rate of the household and establish a primary subjective degree model; The secondary subjective degree model establishment unit is used to analyze the angle relationship in the household image data and establish a secondary subjective degree model.

8. An intelligent door lock monitoring and management system based on cloud computing according to claim 7, characterized in that: The alarm signal regulation module includes a priority setting unit, a model update unit, and a signal output unit; The priority setting unit is used to set the priority of the primary subjective degree model higher than that of the secondary subjective degree model; The model update unit is used to update the primary subjective degree model in the primary subjective degree model analysis; The signal output unit is used to output the alarm signal for the situations that meet the conditions during the judgment process.

Images