A smart building security monitoring system based on the Internet of Things

Abstract

The invention relates to the field of building security, and discloses a smart building security monitoring system based on the Internet of Things. The security monitoring system comprises an in-and-out channel monitoring module, a target area monitoring module, a personnel information module and an analysis control module. The in-and-out channel monitoring module is used to obtain real-time in-and-out channel image data; the target area monitoring module is used to obtain real-time image data in a target area; the personnel information allowed to enter the target area is stored and updated according to preset update rules through the personnel information module; the real-time in-and-out channel image data is analyzed through the analysis control module to obtain monitoring parameters of the target area monitoring module; the real-time image data and personnel information in the target area are analyzed, and it is judged whether there is a visitor who is not allowed to enter intruding according to the analysis result; it is avoided that when the flow of people is large, the visitor who is not allowed to enter cannot be discovered in time, thereby improving the safety and privacy of the smart building.

Description

Technical Field

[0001] The present invention relates to the field of building security, and in particular to a smart building security monitoring system based on the Internet of Things. Background Art

[0002] With the continuous development of information technology and the increasing requirements of people for building quality, smart buildings have come into being; smart buildings integrate core technologies such as the Internet of Things, cloud computing, artificial intelligence and 5G communication technology into all aspects of buildings to provide users with an efficient, comfortable and convenient humanized building environment.

[0003] Usually, users who choose smart buildings usually have higher requirements for the security and privacy of the buildings. At present, many buildings set up security personnel or access control at the entrance to identify visitors and block those who are not allowed to enter from the building entrance to ensure the safety and privacy of users.

[0004] However, when there is a large flow of people at the building entrance, it is difficult for access control and security personnel to identify every visitor, and they are unable to stop visitors who are not allowed to enter in time, which may threaten the safety of users or cause the privacy of users to be leaked. Summary of the invention

[0005] The purpose of the present invention is to provide a smart building security monitoring system based on the Internet of Things to solve the following technical problems:

[0006] How to avoid timely detection of visitors who are not allowed to enter when there is a large flow of people.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A smart building security monitoring system based on the Internet of Things, the security monitoring system comprising:

[0009] The entry and exit channel monitoring module is used to monitor the entry and exit channels of the target area and obtain real-time entry and exit channel image data;

[0010] The target area monitoring module includes a plurality of image acquisition units, which are used to monitor the target area according to the monitoring parameters of the target area monitoring module and obtain real-time image data in the target area;

[0011] A personnel information module, used to store and update the information of personnel allowed to enter the target area according to preset update rules;

[0012] The analysis and control module is connected to the entry and exit channel monitoring module, the target area monitoring module and the personnel information module. It obtains the monitoring parameters of the target area monitoring module by analyzing the real-time entry and exit channel image data; it then analyzes the real-time image data and personnel information in the target area, and determines whether there is an intruder who is not allowed to enter based on the analysis results.

[0013] As a further solution of the present invention: the analysis and control module includes an identification unit, an analysis unit and a control unit;

[0014] The recognition unit is a trained neural network model, which is used to recognize the real-time entry and exit channel image data and the real-time target area image data, obtain the number of people entering each entry and exit channel at the current moment and the number of people in the real-time target area image data; and add a marked contour line on the real-time target area image data, and the area surrounded by the marked contour line is the facial area of ​​the visitor on the real-time target area image data;

[0015] The analysis unit is connected to the recognition module, and is used to analyze the number of people entering each entry and exit channel at the current moment and the number of people in the real-time target area image data, and obtain the monitoring parameters of the target area monitoring module according to the analysis results; and compare the facial area of ​​the visitor with the personnel information to determine whether there is a visitor who is not allowed to enter; if it is determined that there is a visitor who is not allowed to enter, send a warning signal to the control unit; otherwise, no information is sent;

[0016] The control unit is connected to the analysis unit and the target area monitoring module, and is used to control the target area monitoring module according to the monitoring parameters; and to issue an early warning after receiving the early warning signal.

[0017] As a further solution of the present invention: the personnel information includes facial image information and facial image update time; and the monitoring parameters include frame rate.

[0018] As a further solution of the present invention: the process of obtaining the monitoring parameters of the target area monitoring module is as follows:

[0019] S1: Obtain the number of people entering each entry and exit channel at the current moment and the number of people in the real-time target area image data through the recognition unit;

[0020] S2: By analyzing the number of people entering each entrance and exit channel at the current moment and the number of people in the real-time target area image data, the estimated density index in the target area at the next moment is obtained;

[0021] S3: By analyzing the predicted density index, the frame rate of the image acquisition unit of the target area monitoring module at the next moment is obtained.

[0022] As a further solution of the present invention: the predicted density index in the target area at the next moment is:

[0023] ;

[0024] Calculate the estimated density index ;

[0025] in, is the first judgment function, when hour, ;when hour, ; is the number of entry and exit channels of the target area, ; is the number of image acquisition units, ; The number of people entering and exiting the channel is the real-time image data of the nth entry and exit channel; is the number of people in the real-time target area image data of the m-th image acquisition unit at the current moment; Preset the number of people in the target area; is the basic density index; is the first preset constant; is the total area of ​​the target region; is the target area captured by the mth image acquisition unit.

[0026] As a further solution of the present invention: the frame rate of the image acquisition unit of the target area monitoring module at the next moment is determined by: With preset threshold For comparison, use the formula:

[0027] ;

[0028] Determine the frame rate of the image acquisition unit of the target area monitoring module at the next moment ;

[0029] in, is the first-level frame rate, is the secondary frame rate, is the third-level frame rate, and .

[0030] As a further solution of the present invention: the preset update rule includes: by formula:

[0031] ;

[0032] Calculate update index ;

[0033] in, is the second judgment function, when hour, ;when hour, ; The feature similarity between the facial region of the visitor in the real-time target area and the matched facial image information; is the preset feature similarity; The completeness of the visitor's facial area within the real-time target area image data; To preset completeness; The clarity of the visitor's face area within the real-time target area image data; To preset clarity; The number of days from the facial image update time to the current time; is the unit-removing coefficient; is the second preset constant; is the third preset constant; is the first weight coefficient; is the second weight coefficient.

[0034] As a further solution of the present invention: the preset update rule also includes: updating the index With the default value Make a comparison;

[0035] when When the real-time target area image data is updated, the visitor's facial area is updated to the matching facial image information; otherwise, no update is performed.

[0036] Beneficial effects of the present invention:

[0037] (1) The present invention monitors the ingress and egress channels of the target area through an ingress and egress channel monitoring module and obtains real-time ingress and egress channel image data; monitors the target area according to the monitoring parameters of the target area monitoring module through the target area monitoring module and obtains real-time image data in the target area; stores and updates the information of personnel allowed to enter the target area through the personnel information module according to preset update rules; analyzes the real-time ingress and egress channel image data through the analysis control module to obtain the monitoring parameters of the target area monitoring module; then analyzes the real-time image data and personnel information in the target area and determines whether there is an intruder who is not allowed to enter according to the analysis result; thus, it is avoided that when the flow of people is large, it is impossible to timely discover the visitors who are not allowed to enter, thereby improving the security and privacy of the smart building;

[0038] (2) The present invention adjusts the frame rate according to the expected density index, thereby ensuring image clarity while saving energy and extending the life of the image acquisition unit. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The present invention will be further described below in conjunction with the accompanying drawings.

[0040] Figure 1 A system module framework diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] See also Figure 1 As shown, in one embodiment, a smart building security monitoring system based on the Internet of Things is provided, including:

[0043] The entry and exit channel monitoring module is used to monitor the entry and exit channels of the target area and obtain real-time entry and exit channel image data;

[0044] The target area monitoring module includes a plurality of image acquisition units, which are used to monitor the target area according to the monitoring parameters of the target area monitoring module and obtain real-time image data in the target area;

[0045] A personnel information module, used to store and update the information of personnel allowed to enter the target area according to preset update rules;

[0046] The analysis control module is connected with the access channel monitoring module, the target area monitoring module and the personnel information module, and obtains the monitoring parameters of the target area monitoring module by analyzing the real-time access channel image data; and then analyzes the real-time image data and personnel information in the target area, and judges whether there is a visitor who is not allowed to enter according to the analysis results;

[0047] Through the above technical scheme, this embodiment monitors the various entry and exit channels of the target area through the entry and exit channel monitoring module, and obtains real-time entry and exit channel image data; monitors the target area according to the monitoring parameters of the target area monitoring module through the target area monitoring module, and obtains real-time image data in the target area; stores the information of personnel allowed to enter the target area through the personnel information module and updates it according to the preset update rules; analyzes the real-time entry and exit channel image data through the analysis control module to obtain the monitoring parameters of the target area monitoring module; then analyzes the real-time image data and personnel information in the target area, and determines whether there is an intrusion of visitors who are not allowed to enter based on the analysis results; avoids the inability to timely discover visitors who are not allowed to enter when the flow of people is large, and improves the safety and privacy of smart buildings.

[0048] As an embodiment of the present invention, the analysis and control module includes an identification unit, an analysis unit and a control unit;

[0049] The recognition unit is a trained neural network model, which is used to recognize the real-time entry and exit channel image data and the real-time target area image data, obtain the number of people entering each entry and exit channel at the current moment and the number of people in the real-time target area image data; and add a marked contour line on the real-time target area image data, and the area surrounded by the marked contour line is the facial area of ​​the visitor on the real-time target area image data;

[0050] The analysis unit is connected to the recognition module, and is used to analyze the number of people entering each entry and exit channel at the current moment and the number of people in the real-time target area image data, and obtain the monitoring parameters of the target area monitoring module according to the analysis results; and compare the facial area of ​​the visitor with the personnel information to determine whether there is a visitor who is not allowed to enter; if it is determined that there is a visitor who is not allowed to enter, send a warning signal to the control unit; otherwise, no information is sent;

[0051] Through the above technical scheme, this embodiment identifies the real-time entry and exit channel image data and the real-time target area image data through the recognition unit to obtain the number of people entering each entry and exit channel at the current moment and the number of people in the real-time target area image data; and recognizes the faces of the people in the real-time entry and exit channel image data to identify whether the facial skin exposure of the visitors entering the entry and exit channels is higher than the preset skin exposure. If the facial skin exposure of the visitors entering the entry and exit channels is lower than the preset skin exposure, an early warning signal is sent to the control unit through the analysis unit. If the facial skin exposure of the visitors entering the entry and exit channels is higher than the preset skin exposure, the analysis unit is used to analyze the number of people entering each entry and exit channel at the current moment and the number of people in the real-time target area image data, and obtains the information based on the analysis results. Monitoring parameters of the target area monitoring module; the control unit controls the target area monitoring module according to the monitoring parameters to improve the picture quality of the real-time target area image data; the recognition module adds a marking contour line to the real-time target area image data. Due to the improvement of the picture quality, the accuracy of the marking can be improved. The area surrounded by the marking contour line is the facial area of ​​the visitor on the real-time target area image data. Due to the improvement of the accuracy of the marking, the facial area is more complete and clear. The analysis module compares the facial area of ​​the visitor with the personnel information to determine whether there is a visitor who is not allowed to enter, which can improve the accuracy of the judgment; if it is determined that there is a visitor who is not allowed to enter, a warning signal is sent to the control unit; the control unit issues a warning after receiving the warning signal;

[0052] It should be noted that the neural network model used by the recognition unit may be a convolutional neural network (CNN), and the specific training process is prior art and will not be described in detail here.

[0053] As an implementation manner of the present invention, the personnel information includes facial image information and facial image update time; and the monitoring parameters include frame rate.

[0054] As an implementation mode of the present invention, the process of obtaining the monitoring parameters of the target area monitoring module is as follows:

[0055] S1: Obtain the number of people entering each entry and exit channel at the current moment and the number of people in the real-time target area image data through the recognition unit;

[0056] S2: By analyzing the number of people entering each entrance and exit channel at the current moment and the number of people in the real-time target area image data, the estimated density index in the target area at the next moment is obtained;

[0057] S3: by analyzing the predicted density index, obtaining the frame rate of the image acquisition unit of the target area monitoring module at the next moment;

[0058] Through the above technical scheme, this embodiment first obtains the number of people entering each entry and exit channel at the current moment and the number of people in the real-time target area image data through the identification unit; then, by analyzing the number of people entering each entry and exit channel at the current moment and the number of people in the real-time target area image data, the expected density index in the target area at the next moment is obtained; finally, by analyzing the expected density index, the frame rate of the image acquisition unit of the target area monitoring module at the next moment is obtained.

[0059] As an implementation mode of the present invention, the predicted density index in the target area at the next moment is:

[0060] ;

[0061] Calculate the estimated density index ;

[0062] in, is the first judgment function, when hour, ;when hour, ; is the number of entry and exit channels of the target area, ; is the number of image acquisition units, ; The number of people entering and exiting the channel is the real-time image data of the nth entry and exit channel; is the number of people in the real-time target area image data of the m-th image acquisition unit at the current moment; Preset the number of people in the target area; is the basic density index; is the first preset constant; is the total area of ​​the target region; is the target area captured by the mth image acquisition unit;

[0063] Through the above technical solution, this embodiment is the total number of people entering the N entrance and exit channels at the current moment; the density in the target area is the total number of people in the image data of the M image acquisition units at the current moment; is the ratio of the total area of ​​the target area to the area of ​​the target area captured by M image acquisition units, through Calculate the estimated total number of people in the entire target area; is the total number of people entering the N entry and exit channels at the current moment and the estimated total number of people in the entire target area The sum of The larger it is, the more people are likely to be in the target area at the next moment; is the total number of people entering the N entry and exit channels at the current moment and the estimated total number of people in the entire target area The sum of the number of people in the target area The difference between When the number of people in the target area at the next moment may exceed the preset number of people in the target area ; N entrance and exit channels, the total number of people entering the entrance and exit channels at the current moment and the estimated total number of people in the entire target area The sum of the number of people in the target area The difference The larger the value, the more crowded the target area may be, so the density index is expected to be The bigger;

[0064] It should be noted that the number of people in the target area is preset. , basic density index and the first preset constant is the preset value, obtained based on experience, the total area of ​​the target area and the target area captured by the mth image acquisition unit It is obtained through measurement and will not be described in detail here.

[0065] As an embodiment of the present invention, the frame rate of the image acquisition unit of the target area monitoring module at the next moment is determined by: With preset threshold For comparison, use the formula:

[0066] ;

[0067] Determine the frame rate of the image acquisition unit of the target area monitoring module at the next moment ;

[0068] in, is the first-level frame rate, is the secondary frame rate, is the third-level frame rate, and ;

[0069] Through the above technical solution, this embodiment will estimate the density index With preset threshold For comparison, When the number of people at the next moment is large and the density is high, the frame rate of the image acquisition unit of the target area monitoring module at the next moment is increased to the first frame rate. , can improve the smoothness of the picture, make each frame of the picture clear, improve the picture quality, facilitate face recognition, and improve the accuracy of face recognition; when When the number of people at the next moment is large and the density is high, the frame rate of the image acquisition unit of the target area monitoring module at the next moment is adjusted to the secondary frame rate ,when When , it means that the number of people at the next moment is small and the density is low, and the frame rate of the image acquisition unit of the target area monitoring module at the next moment is adjusted to the third frame rate , according to the estimated density index Adjust frame rate , while ensuring picture clarity, it can also save energy and extend the life of the image acquisition unit;

[0070] It should be noted that the preset threshold , first-level frame rate , Secondary frame rate , three-level frame rate It is a preset value obtained based on experience and will not be described in detail here.

[0071] As an implementation mode of the present invention, the preset update rule includes: by formula:

[0072] ;

[0073] Calculate update index ;

[0074] in, is the second judgment function, when hour, ;when hour, ; The feature similarity between the facial region of the visitor in the real-time target area and the matched facial image information; is the preset feature similarity; The completeness of the visitor's facial area within the real-time target area image data; To preset completeness; The clarity of the visitor's face area within the real-time target area image data; To preset clarity; The number of days from the facial image update time to the current time; is the unit-removing coefficient; is the second preset constant; is the third preset constant; is the first weight coefficient; is the second weight coefficient;

[0075] Update the index With the default value Make a comparison;

[0076] when When the real-time target area image data is updated, the visitor's facial area is updated to the matching facial image information; otherwise, no update is performed.

[0077] Through the above technical solution, this embodiment uses the formula Calculate update index Only when and and hour, ;otherwise, ; Number of days from facial image update time to current time The longer it is, the greater the probability that the person's appearance corresponding to the facial image information may change, so it needs to be updated more frequently. The larger the size, the more complete the facial area of ​​the visitor is within the real-time target area image data. The higher the value, the more complete the facial features obtained, and the better the facial image information is. The larger the size, the clearer the facial area of ​​the visitor in the real-time target area image data. The higher the value, the clearer the facial features are, and the better the facial image information is. The larger the index is, the more it will be updated. With the default value Make a comparison;

[0078] when When the real-time target area image data is updated, the visitor's facial area is updated to the matching facial image information; otherwise, no update is performed.

[0079] It should be noted that the preset feature similarity , preset completeness , preset definition , remove unit coefficient , the second preset constant , the third preset constant , the first weight coefficient , the second weight coefficient and preset values The default value is the number of days from the facial image update time to the current time. This is prior art, obtained based on experience, and will not be described in detail here.

[0080] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A smart building security monitoring system based on the Internet of Things, characterized in that: The security monitoring system comprises: The entry and exit channel monitoring module is used to monitor the entry and exit channels of the target area and obtain real-time entry and exit channel image data; The target area monitoring module includes a plurality of image acquisition units, which are used to monitor the target area according to the monitoring parameters of the target area monitoring module and obtain real-time image data in the target area; A personnel information module, used to store and update the information of personnel allowed to enter the target area according to preset update rules; The analysis control module is connected with the access channel monitoring module, the target area monitoring module and the personnel information module, and obtains the monitoring parameters of the target area monitoring module by analyzing the real-time access channel image data; and then analyzes the real-time image data and personnel information in the target area, and judges whether there is a visitor who is not allowed to enter according to the analysis results; The preset update rule includes: by formula: Calculate the update index A; Wherein, g(X) is the second judgment function, when X>0, g(X)=1; when X≤0, g(X)=0; S x is the feature similarity between the visitor's facial area and the matched facial image information in the real-time target area; S0 is the preset feature similarity; Y x is the completeness of the visitor's face area in the real-time target area image data; Y0 is the preset completeness; Z x is the definition of the facial area of ​​the visitor in the real-time target area image data; Z0 is the preset definition; Δt is the number of days from the facial image update time to the current time; ε is the de-unit coefficient; C2 is the second preset constant; C3 is the third preset constant; γ1 is the first weight coefficient; γ2 is the second weight coefficient.

2. According to the Internet of Things-based smart building security monitoring system of claim 1, it is characterized in that: The analysis and control module includes an identification unit, an analysis unit and a control unit; The recognition unit is a trained neural network model, which is used to recognize the real-time entry and exit channel image data and the real-time target area image data, and obtain the number of people entering each entry and exit channel and the number of people in the real-time target area image data at the current moment; and adding a marking contour line to the real-time target area image data, wherein the area enclosed by the marking contour line is the facial area of ​​the visitor on the real-time target area image data; The analysis unit is connected to the recognition module, and is used to analyze the number of people entering each entry and exit channel at the current moment and the number of people in the real-time target area image data, and obtain the monitoring parameters of the target area monitoring module according to the analysis results; and compare the facial area of ​​the visitor with the personnel information to determine whether there is a visitor who is not allowed to enter; if it is determined that there is a visitor who is not allowed to enter, send a warning signal to the control unit; otherwise, no information is sent; The control unit is connected to the analysis unit and the target area monitoring module, and is used to control the target area monitoring module according to the monitoring parameters; And issue an early warning after receiving the early warning signal.

3. The smart building security monitoring system based on the Internet of Things according to claim 2 is characterized in that: The personnel information includes facial image information and facial image update time; the monitoring parameters include frame rate.

4. The smart building security monitoring system based on the Internet of Things according to claim 3 is characterized in that: The process of obtaining the monitoring parameters of the target area monitoring module is as follows: S1: Obtain the number of people entering each entry and exit channel at the current moment and the number of people in the real-time target area image data through the recognition unit; S2: By analyzing the number of people entering each entrance and exit channel at the current moment and the number of people in the real-time target area image data, the estimated density index in the target area at the next moment is obtained; S3: By analyzing the predicted density index, the frame rate of the image acquisition unit of the target area monitoring module at the next moment is obtained.

5. According to claim 4, a smart building security monitoring system based on the Internet of Things is characterized in that: The predicted density index at the next moment in the target area: Calculate the expected density index Q; Wherein, f(X) is the first judgment function, when X>0, f(X)=X; when X≤0, f(X)=0; N is the number of inlet and outlet channels of the target area, n∈N; M is the number of image acquisition units, m∈M; I n is the number of people entering and exiting the channel in real-time image data of the nth entry and exit channel; T m is the number of people in the real-time target area image data of the mth image acquisition unit at the current moment; P0 is the preset number of people in the target area; Q0 is the basic density index; C1 is the first preset constant; S a is the total area of ​​the target area; S m is the target area captured by the mth image acquisition unit.

6. According to the Internet of Things-based smart building security monitoring system of claim 5, it is characterized in that: The frame rate of the image acquisition unit of the target area monitoring module at the next moment is determined by comparing the expected density index Q with the preset threshold [R1, R2], and then using the formula: Determine the frame rate FPS of the image acquisition unit of the target area monitoring module at the next moment n ; Among them, FPS1 is the primary frame rate, FPS2 is the secondary frame rate, FPS3 is the tertiary frame rate, and FPS1>FPS2>FPS3.

7. The smart building security monitoring system based on the Internet of Things according to claim 1 is characterized in that: The preset update rule also includes: comparing the update index A with the preset value R3; When A>R3, the visitor's facial area in the real-time target area image data is updated to the matching facial image information; otherwise, no update is performed.

Images