Intelligent security monitoring method and system
By monitoring air flow and signal processing in real time, the VMD algorithm is used to decompose and reconstruct signals, and combining wind speed fluctuation index and abnormal coefficients, the problems of large calculation resources and high misjudgment rate of traditional security monitoring systems are solved, achieving more efficient and accurate Anomaly detection.
Patent Information
- Application Number
- CN202410143990.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-02-01
- Publication Date
- 2024-12-06
- Estimated Expiration
- 2044-02-01
AI Technical Summary
Traditional security monitoring systems take up too much computing resources when processing large amounts of multi-source data, resulting in inefficiency and are prone to unnecessary alarms due to misjudgment of environmental noise and vibration, which affects the accuracy and reliability of the system.
By monitoring the air flow in real time, collecting sound signals and vibration signal data, decomposing and reconstructing signals using VMD algorithm, combining wind speed fluctuation index and abnormal coefficients, unnecessary data processing is reduced, signal processing efficiency and accuracy are improved, and errors are reduced judgment rate.
It realizes more accurate judgment of abnormal flow patterns, reduces unnecessary data processing and false alarms, improves the accuracy of early warning and data utilization efficiency, and reduces analysis costs.
Smart Images

Figure CN118015808B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of security technology, and in particular to an intelligent security monitoring method and system. Background Art
[0002] With the development of society and the advancement of science and technology, people's demand for security precautions is getting higher and higher. The development background of the intelligent security monitoring system is gradually formed with the improvement of people's security needs and the continuous advancement of technology. On the basis of the traditional security monitoring system, the intelligent security monitoring system adopts artificial intelligence, big data, cloud computing and other advanced technologies to achieve more intelligent and efficient monitoring and management.
[0003] In a Chinese invention application with application publication number CN116761049B, a method and system for intelligent home security monitoring are disclosed, and the method includes the following steps: using a preset remote monitoring camera to shoot a home environment at a preset time interval to obtain a first transmission image; performing key frame extraction processing on the first transmission image to obtain a first key frame set; performing frame image overlap processing on the first key frame set based on the processing module in the remote monitoring camera to generate an environmental blurred image; performing image cutting processing on the environmental blurred image to generate an environmental cut image; performing image preprocessing on the environmental cut image to generate a standard environmental cut image; performing color depth detection processing on the standard environmental cut image to generate an environmental color depth image.
[0004] In the Chinese invention application with application publication number CN110012271A, an intelligent monitoring security system and its use method are disclosed, including a monitoring image acquisition module, a data processing system, a networked big database and a security center alarm monitoring system, the output end of the monitoring image acquisition module is electrically connected to the input end of the data processing system, and the networked big database and the data processing system realize a two-way wireless connection.
[0005] In view of the above invention, the prior art has the following deficiencies:
[0006] 1. Traditional analysis methods usually directly process and analyze a large amount of multi-source data. Due to the huge amount of data, existing technologies require powerful computing resources (such as high-performance computers or servers) to complete the analysis tasks, which leads to excessive occupation of computing resources, increases computing costs, and may lead to low analysis efficiency;
[0007] 2. In security monitoring or similar scenarios, if the sound and vibration originate from outside the security area, they may be misjudged as abnormal conditions, which may lead to unnecessary alarms or false alarms, affecting the accuracy and reliability of the system. Sometimes normal sounds and vibrations in the environment (such as wind, animals, traffic, etc.) may be mistaken for abnormal signals, leading to misjudgment, which not only reduces the reliability of the system, but may also trigger unnecessary countermeasures. Summary of the invention
[0008] 1. Technical issues to be solved
[0009] In view of the deficiencies in the prior art, the present invention provides an intelligent security monitoring method and system, which monitors air flow in real time, collects sound signal and vibration signal data, analyzes the discreteness of wind speed data through wind speed at different positions and wind speed changes at a single point, obtains indoor wind speed fluctuation index and single-point wind speed fluctuation index, and further associates to form an air flow anomaly coefficient. When the air flow anomaly coefficient is greater than or equal to the wind speed fluctuation threshold, the sound signal and vibration signal at the single point position are respectively decomposed into a number of IMF components by VMD algorithm, and reconstructed, and the reconstructed sound signal and vibration signal are fused into a new signal, and compared with the normal feature signal in the normal feature library, and the correlation coefficient between the fused new signal and all normal feature signals is calculated, and the maximum value of all correlation coefficients is used as the sound-vibration anomaly coefficient, and the sound-vibration anomaly threshold is pre-set, and the sound-vibration anomaly threshold is compared with the sound-vibration anomaly coefficient, and corresponding measures are taken according to the comparison result, thereby solving the technical problems mentioned in the background technology.
[0010] (II) Technical solution
[0011] To achieve the above objectives, the present invention is implemented through the following technical solutions: an intelligent security monitoring method, comprising the following steps:
[0012] Monitor air flow in real time and collect sound and vibration signal data;
[0013] By analyzing the wind speed at different locations and the wind speed changes at a single point, the discrete degree of wind speed data is analyzed to obtain the indoor wind speed fluctuation index and the single-point wind speed fluctuation index, which are further correlated to form the air flow anomaly coefficient;
[0014] When the air flow anomaly coefficient is greater than or equal to the wind speed fluctuation threshold, the sound signal and vibration signal at a single point are decomposed into several IMF components and reconstructed using the VMD algorithm;
[0015] The reconstructed sound signal and vibration signal are fused into a new signal, and compared with the normal feature signals in the normal feature library, and the correlation coefficient between the fused new signal and all normal feature signals is calculated, and the maximum value of all correlation coefficients is taken as the sound vibration abnormality coefficient;
[0016] The abnormal sound and vibration threshold is set in advance, and the abnormal sound and vibration threshold is compared with the abnormal sound and vibration coefficient, and corresponding measures are taken according to the comparison result.
[0017] Furthermore, air flow sensors are installed at different locations in the security area to monitor the air flow in real time; sensors are arranged around the air flow sensors to collect sound signals, vibration signals and image data, and the sensors include sound sensors, vibration sensors and cameras.
[0018] Further, the position of the air flow sensor is marked as i, i = 1, 2, ..., n, and n is a positive integer;
[0019] Obtain real-time air flow data at different locations, analyze the discrete degree of indoor wind speed data through wind speed at different locations, evaluate the fluctuation of indoor air flow, and obtain the indoor wind speed fluctuation index. The calculation formula is as follows:
[0020]
[0021] Among them, Isf represents the indoor wind speed fluctuation index, v i Indicates the wind speed at different locations. represents the average wind speed at different locations, and n represents the number of air flow sensors.
[0022] Furthermore, the air flow change data at different times of a single point location is obtained. Through the wind speed change at a single point location, the discrete degree of the wind speed data at the single point location is analyzed, the fluctuation of the air flow at the single point location is evaluated, and the single point wind speed fluctuation index is obtained. The calculation formula is as follows:
[0023]
[0024] Among them, Psf i represents the wind speed fluctuation index at position i, T represents the number at different times, and t represents the mark at different times. represents the wind speed at position i at time t, represents the average wind speed at location i.
[0025] Furthermore, the indoor wind speed fluctuation index and the single-point wind speed fluctuation index are obtained, and further associated to form the air flow anomaly coefficient. The calculation formula is as follows:
[0026]
[0027] Among them, Afc represents the air flow anomaly coefficient, Isf represents the indoor wind speed fluctuation index, and Psf i represents the wind speed fluctuation index at position i, n represents the number of air flow sensors, k1 and k2 represent weight coefficients, 0 < k1 < 1, 0 < k2 < 1, and k1 + k2 = 1.
[0028] Furthermore, a wind speed fluctuation threshold is preset. When the air flow anomaly coefficient is less than the wind speed fluctuation threshold, no processing is performed. When the air flow anomaly coefficient is greater than or equal to the wind speed fluctuation threshold, the single-point position where the maximum value of the single-point wind speed fluctuation index is located is obtained and marked as an abnormal point;
[0029] The sound signal data and vibration signal data of the abnormal point are obtained, and the sound signal and vibration signal data are respectively decomposed into several IMF components through the VMD algorithm;
[0030] The correlation coefficient between each IMF component and the original signal is calculated, and the IMF components with a correlation coefficient lower than the preset correlation threshold are removed;
[0031] The remaining IMF components are reconstructed to obtain the reconstructed sound and vibration signals.
[0032] Furthermore, the reconstructed sound signal and vibration signal are obtained, and the reconstructed sound signal and vibration signal are normalized to map the data to the range of [0, 1];
[0033] The reconstructed sound signal and vibration signal are fused into a new signal, and the corresponding calculation formula of the new signal is as follows:
[0034] V(t′) = X(t′) + Y(t′)
[0035] Among them, V(t') represents the fused new signal, X(t') represents the reconstructed sound signal, Y(t') represents the reconstructed vibration signal, and t' represents the time point.
[0036] Furthermore, the fused new signal is compared with the normal feature signal in the normal feature library, the correlation coefficient between the fused new signal and the normal feature signal is calculated, and the maximum value among all the correlation coefficients is used as the acoustic-vibration anomaly coefficient. The calculation formula of the correlation coefficient is as follows:
[0037]
[0038] Among them, γ represents the correlation coefficient, t' represents the time point, m represents the number of time points, represents the average value of the fused new signal V(t'), represents the average value of the normal feature signal C(t').
[0039] Furthermore, a sound vibration abnormality threshold is preset, and by comparing the sound vibration abnormality threshold with the sound vibration abnormality coefficient, corresponding measures are taken according to the comparison result, specifically:
[0040] When the sound vibration abnormality coefficient is greater than or equal to the sound vibration abnormality threshold, no processing is performed; when the sound vibration abnormality coefficient is less than the sound vibration abnormality threshold, an alarm is automatically issued and the collected data is uploaded to the storage end.
[0041] An intelligent security monitoring system includes a data collection module, an air flow detection module, an acoustic vibration signal processing module, an acoustic vibration signal fusion module and an early warning module; wherein,
[0042] Data collection module: install air flow sensors at different locations in the security area to monitor air flow in real time, and arrange sensors around the air flow sensors to collect sound and vibration signal data;
[0043] The air flow detection module analyzes the discrete degree of wind speed data through the wind speed at different positions and the wind speed changes at a single point, obtains the indoor wind speed fluctuation index and the single-point wind speed fluctuation index, and further correlates the indoor wind speed fluctuation index and the single-point wind speed fluctuation index to form an air flow anomaly coefficient;
[0044] The acoustic and vibration signal processing module obtains the single point position where the single point wind speed fluctuation index is the maximum when the air flow anomaly coefficient is greater than or equal to the wind speed fluctuation threshold, decomposes the sound signal and vibration signal of the single point position into several IMF components respectively through the VMD algorithm, calculates the correlation coefficient between each IMF component and the original signal, removes the IMF components with correlation coefficients lower than the preset correlation threshold, reconstructs the remaining IMF components, and obtains the reconstructed sound and vibration signals;
[0045] The acoustic-vibration signal fusion module normalizes the reconstructed sound signal and vibration signal and fuses them into a new signal. The fused new signal is compared with the normal feature signal in the normal feature library, and the correlation coefficient between the fused new signal and all normal feature signals is calculated. The maximum value of all correlation coefficients is taken as the acoustic-vibration abnormality coefficient.
[0046] The early warning module pre-sets the sound and vibration abnormality threshold, compares the sound and vibration abnormality threshold with the sound and vibration abnormality coefficient, and takes corresponding measures based on the comparison results.
[0047] (III) Beneficial effects
[0048] The present invention provides an intelligent security monitoring method and system, which has the following beneficial effects:
[0049] (1) By comprehensively considering the wind speed fluctuations of the whole room and at specific locations, the abnormal air flow conditions can be evaluated more comprehensively. The indoor wind speed fluctuation index reflects the fluctuation of the overall indoor air flow, while the single-point wind speed fluctuation index focuses on the wind speed fluctuation at a specific location. Combining the two can more accurately determine the abnormal flow pattern.
[0050] (2) By calculating the correlation coefficient between each IMF component and the original signal and eliminating the IMF components with correlation coefficients lower than the preset threshold, the dimension of the feature space is reduced, making subsequent pattern recognition or classification tasks more efficient. After processing and reconstruction by the VMD algorithm, the noise and irrelevant components in the signal are suppressed, and the data quality is improved.
[0051] (3) The reconstructed sound signal and vibration signal are fused into a new signal, and the characteristics of the two signals are comprehensively utilized to extract more comprehensive and accurate information, thereby improving the efficiency and accuracy of signal processing and further enhancing the signal's feature expression ability and recognition performance.
[0052] (4) By first determining whether there is abnormal air flow and then analyzing the sound and vibration signals, unnecessary data processing can be reduced, which helps to improve data utilization efficiency, reduce analysis costs, avoid abnormal warnings due to factors outside the security area, narrow the monitoring range, improve the accuracy of warnings, and reduce misjudgment of abnormal situations. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] Figure 1 This is a schematic diagram of the steps of the intelligent security monitoring method of the present invention;
[0054] Figure 2 This is a flow chart of the intelligent security monitoring method of the present invention;
[0055] Figure 3 It is a schematic diagram of the structure of the intelligent security monitoring system of the present invention. DETAILED DESCRIPTION
[0056] 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.
[0057] See also Figure 1 and Figure 2 The present invention provides an intelligent security monitoring method, comprising the following steps:
[0058] Step 1: Install air flow sensors at different locations in the security area to monitor air flow in real time, and arrange sensors around the air flow sensors to collect sound and vibration signal data;
[0059] The step 1 includes the following contents:
[0060] Step 101: Install air flow sensors at different locations in the security area to monitor air flow in real time;
[0061] Step 102: Arrange sensors around the air flow sensor to collect data such as sound signals and vibration signals, wherein the sensors include sound sensors, vibration sensors, and cameras;
[0062] It should be noted that installing air flow sensors at key locations in the security area can monitor information such as air flow speed and direction. This information can help determine whether there are abnormal situations, such as the activities of suspicious people, the use of vents, etc. Arranging other sensors around the air flow sensors can further confirm abnormal situations, so as to collect data in the security area in an all-round way;
[0063] Step 103: preprocess the collected data, including data cleaning, noise removal, missing value filling and other operations, and store the preprocessed data in a database or storage device.
[0064] It should be noted that preprocessing the collected data is a necessary step, including data cleaning, noise removal, and missing value filling. These operations can improve the accuracy and reliability of the data and provide a better foundation for subsequent data analysis.
[0065] When using, combine the contents of step 101 to step 103:
[0066] By installing air flow sensors, sound sensors and vibration sensors at different locations in the security area, comprehensive monitoring of the security area is achieved, the accuracy and reliability of security monitoring are improved, and it is helpful to detect abnormal situations in time and take corresponding measures.
[0067] Step 2: Analyze the discrete degree of wind speed data through the wind speed at different locations and the wind speed changes at a single point, obtain the indoor wind speed fluctuation index and the single-point wind speed fluctuation index, and further correlate the indoor wind speed fluctuation index and the single-point wind speed fluctuation index to form the air flow anomaly coefficient;
[0068] The step 2 includes the following contents:
[0069] Step 201: mark the location of the air flow sensor as i, where i=1, 2, ..., n, and n is a positive integer;
[0070] Step 202: Obtain the real-time air flow data at different positions, analyze the dispersion degree of the indoor wind speed data based on the wind speeds at different positions, evaluate the fluctuation condition of the indoor air flow, and obtain the indoor wind speed fluctuation index. The calculation formula is as follows:
[0071]
[0072] where Isf represents the indoor wind speed fluctuation index, v i represents the wind speed at different positions, represents the average wind speed at different positions, and n represents the number of air flow sensors;
[0073] Step 203: Obtain the air flow change data at a single point position at different times, analyze the dispersion degree of the wind speed data at the single point position based on the wind speed change at the single point position, evaluate the fluctuation of the air flow at the single point position, and obtain the single-point wind speed fluctuation index. The calculation formula is as follows:
[0074]
[0075] where Psf i represents the wind speed fluctuation index at position i, T represents the number of different times, t represents the mark of different times, represents the wind speed at position i at time t, represents the average wind speed at position i;
[0076] Specifically, when the wind speed fluctuation at different positions in the room is greater, it indicates that there may be abnormal flow conditions in the room. Therefore, it is necessary to further analyze the wind speed fluctuation at a single point position. When the wind speed fluctuation at a single point position is greater, it indicates that the possibility of abnormal air flow in the room is greater, and vice versa;
[0077] Step 204: Obtain the indoor wind speed fluctuation index and the single-point wind speed fluctuation index, and further associate them to form an air flow anomaly coefficient. The calculation formula is as follows:
[0078]
[0079] where Afc represents the air flow anomaly coefficient, Isf represents the indoor wind speed fluctuation index, Psf i represents the wind speed fluctuation index at position i, n represents the number of air flow sensors, k1 and k2 represent weight coefficients, 0 < k1 < 1, 0 < k2 < 1, and k1 + k2 = 1.
[0080] It should be noted that the setting of the weight coefficient needs to be adjusted according to the actual scenario. There is no universal weight setting that can be applied to all situations. This is because the air flow pattern and abnormality standard in different scenarios may be different. Therefore, in order to obtain the best detection effect, the weight coefficient needs to be adjusted according to the specific application scenario.
[0081] When used, combine the contents of step 201 to step 204:
[0082] By comprehensively considering the wind speed fluctuations of the whole room and at specific locations, the abnormal air flow conditions can be evaluated more comprehensively. The indoor wind speed fluctuation index reflects the fluctuation of the overall indoor air flow, while the single-point wind speed fluctuation index focuses on the wind speed fluctuation at a specific location. Combining the two can more accurately judge abnormal flow patterns.
[0083] Step 3: When the air flow anomaly coefficient is greater than or equal to the wind speed fluctuation threshold, obtain the single point position where the single point wind speed fluctuation index is the maximum, and decompose the sound signal and vibration signal of the single point position into several IMF components respectively by using the VMD algorithm, calculate the correlation coefficient between each IMF component and the original signal, and remove the IMF components with correlation coefficients lower than the preset correlation threshold, reconstruct the remaining IMF components, and obtain the reconstructed sound and vibration signals;
[0084] The step three includes the following contents:
[0085] Step 301: pre-set a wind speed fluctuation threshold. When the air flow anomaly coefficient is less than the wind speed fluctuation threshold, no processing is performed. When the air flow anomaly coefficient is greater than or equal to the wind speed fluctuation threshold, obtain the single point position where the maximum value of the single point wind speed fluctuation index is located and mark it as an abnormal point.
[0086] Step 302: Acquire the sound signal data and vibration signal data of the abnormal point, and decompose the sound signal and vibration signal data into a plurality of IMF components respectively by using the VMD algorithm;
[0087] It should be noted that the VMD algorithm is a new signal processing method that can decompose complex signals into several intrinsic mode functions (IMFs). Each IMF component contains different frequency and time information of the signal. The basic principle of the VMD algorithm is to continuously update the frequency and time information of each IMF component through iteration, so that each IMF component can adaptively approximate the original signal.
[0088] Step 303: Calculate the correlation coefficient between each IMF component and the original signal, and remove the IMF component with a correlation coefficient lower than a preset correlation threshold. The calculation formula of the correlation coefficient is as follows:
[0089]
[0090] Among them, Cor(A,B) represents the correlation coefficient, A j and B j Represent the j-th sample value of the IMF component and the original signal respectively, and Represent the mean of the IMF component and the original signal respectively;
[0091] Step 304: reconstruct the remaining IMF components to obtain reconstructed sound and vibration signals.
[0092] It should be noted that the correlation coefficient between each IMF component and the original signal is calculated. The correlation coefficient is used to measure the similarity between the IMF component and the original signal. According to the preset correlation threshold, those IMF components with correlation coefficients lower than the threshold are eliminated. These eliminated IMF components may contain more noise or unimportant information.
[0093] When used, combine the contents of step 301 to step 304:
[0094] By calculating the correlation coefficient between each IMF component and the original signal and removing the IMF components with correlation coefficients lower than the preset threshold, the dimension of the feature space is reduced, making subsequent pattern recognition or classification tasks more efficient. After processing and reconstruction by the VMD algorithm, the noise and irrelevant components in the signal are suppressed and the data quality is improved.
[0095] Step 4: Normalize the reconstructed sound signal and vibration signal and fuse them into a new signal. Compare the fused new signal with the normal feature signal in the normal feature library, calculate the correlation coefficient between the fused new signal and all normal feature signals, and take the maximum value of all correlation coefficients as the sound and vibration abnormality coefficient.
[0096] The step 4 includes the following contents:
[0097] Step 401: Obtain the reconstructed sound signal and vibration signal, and normalize the reconstructed sound signal and vibration signal to map the data to the range of [0,1]. The calculation formula is as follows:
[0098]
[0099] Among them, X(t') represents the reconstructed sound signal, X(t') min Represents the minimum value of the reconstructed sound signal, X(t') max represents the maximum value of the reconstructed sound signal, Y(t') represents the reconstructed vibration signal, and Y(t') min Represents the minimum value of the reconstructed vibration signal, Y(t') maxrepresents the maximum value of the reconstructed vibration signal, and t' represents the time point;
[0100] It should be noted that the purpose of normalizing the reconstructed sound and vibration signals is to normalize the data of different dimensions and magnitudes into a unified standard range, eliminate the differences between the data, and enable the data to be uniformly processed and compared;
[0101] Step 402: The reconstructed sound signal and vibration signal are merged into a new signal. The calculation formula corresponding to the new signal is as follows:
[0102] V(t′)=X(t′)+Y(t′)
[0103] Among them, V(t') represents the new signal after fusion;
[0104] Step 403: Compare the fused new signal with the normal feature signal in the normal feature library, and calculate the correlation coefficient between the fused new signal and the normal feature signal. The calculation formula is as follows:
[0105]
[0106] Among them, γ represents the correlation coefficient, m represents the number of time points, represents the average value of the new signal V(t') after fusion, Represents the average value of the normal characteristic signal C(t');
[0107] Step 404: Calculate the correlation coefficient between the fused new signal and all normal feature signals, and take the maximum value of all correlation coefficients as the acoustic vibration abnormality coefficient.
[0108] It should be noted that by comparing the fused new signal with the normal feature signal in the normal feature library, the abnormal signal can be detected and the abnormal features can be extracted. The correlation coefficient can be calculated to measure the similarity between the fused new signal and the normal feature signal. By finding the maximum value in the correlation coefficient, it can be determined whether there is an abnormality.
[0109] When used, combine the contents of step 401 to step 404:
[0110] The reconstructed sound signal and vibration signal are fused into a new signal, and the characteristics of the two signals are comprehensively utilized to extract more comprehensive and accurate information, thereby improving the efficiency and accuracy of signal processing and further enhancing the signal's feature expression ability and recognition performance.
[0111] Step 5: Pre-set the abnormal sound and vibration threshold, compare the abnormal sound and vibration threshold with the abnormal sound and vibration coefficient, and take corresponding measures according to the comparison results, specifically:
[0112] When the sound vibration abnormality coefficient is greater than or equal to the sound vibration abnormality threshold, it means that there is no abnormal situation at present and no action is taken; when the sound vibration abnormality coefficient is less than the sound vibration abnormality threshold, it means that there is an abnormal situation at present and an alarm is automatically issued, including SMS alarm, phone alarm and APP push, etc., and the collected data is uploaded.
[0113] By first determining whether there is abnormal air flow and then analyzing the sound and vibration signals, unnecessary data processing can be reduced, which helps to improve data utilization efficiency, reduce analysis costs, avoid abnormal warnings due to factors outside the security area, narrow the monitoring scope, improve the accuracy of warnings, and reduce misjudgment of abnormal situations.
[0114] See also Figure 3 The present invention also provides an intelligent security monitoring system, including a data collection module, an air flow detection module, an acoustic vibration signal processing module, an acoustic vibration signal fusion module and an early warning module; wherein,
[0115] Data collection module: install air flow sensors at different locations in the security area to monitor air flow in real time, and arrange sensors around the air flow sensors to collect sound and vibration signal data;
[0116] The air flow detection module analyzes the discrete degree of wind speed data through the wind speed at different positions and the wind speed changes at a single point, obtains the indoor wind speed fluctuation index and the single-point wind speed fluctuation index, and further correlates the indoor wind speed fluctuation index and the single-point wind speed fluctuation index to form an air flow anomaly coefficient;
[0117] The acoustic and vibration signal processing module obtains the single point position where the single point wind speed fluctuation index is the maximum when the air flow anomaly coefficient is greater than or equal to the wind speed fluctuation threshold, decomposes the sound signal and vibration signal of the single point position into several IMF components respectively through the VMD algorithm, calculates the correlation coefficient between each IMF component and the original signal, removes the IMF components with correlation coefficients lower than the preset correlation threshold, reconstructs the remaining IMF components, and obtains the reconstructed sound and vibration signals;
[0118] The acoustic-vibration signal fusion module normalizes the reconstructed sound signal and vibration signal and fuses them into a new signal. The fused new signal is compared with the normal feature signal in the normal feature library, and the correlation coefficient between the fused new signal and all normal feature signals is calculated. The maximum value of all correlation coefficients is taken as the acoustic-vibration abnormality coefficient.
[0119] The early warning module pre-sets the sound and vibration abnormality threshold, compares the sound and vibration abnormality threshold with the sound and vibration abnormality coefficient, and takes corresponding measures based on the comparison results.
[0120] In the application, the several formulas involved are all calculated by removing dimensions and taking their numerical values. The formula is a formula of the most recent real situation obtained by collecting a large amount of data and performing software simulation. The coefficients in the formula are set by technicians in this field according to actual conditions.
[0121] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. A person of ordinary skill in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented in combination with electronic hardware, computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution.
[0122] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0123] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.
Claims
1. An intelligent security monitoring method, characterized in that: The following steps are involved: Monitor air flow in real time and collect sound and vibration signal data; By analyzing the wind speed at different locations and the wind speed changes at a single point, the discrete degree of wind speed data is analyzed to obtain the indoor wind speed fluctuation index and the single-point wind speed fluctuation index, which are further correlated to form the air flow anomaly coefficient; When the air flow anomaly coefficient is greater than or equal to the wind speed fluctuation threshold, the sound signal and vibration signal at a single point are decomposed into several IMF components and reconstructed using the VMD algorithm; The reconstructed sound signal and vibration signal are fused into a new signal, and compared with the normal feature signals in the normal feature library, and the correlation coefficient between the fused new signal and all normal feature signals is calculated, and the maximum value of all correlation coefficients is taken as the sound vibration abnormality coefficient; Pre-set the abnormal sound and vibration threshold, compare the abnormal sound and vibration threshold with the abnormal sound and vibration coefficient, and take corresponding measures according to the comparison results; The position of the air flow sensor is marked as i, i = 1, 2, ..., n, and n is a positive integer; Obtain real-time air flow data at different locations, analyze the discrete degree of indoor wind speed data through wind speed at different locations, evaluate the fluctuation of indoor air flow, and obtain the indoor wind speed fluctuation index. The calculation formula is as follows: Among them, Isf represents the indoor wind speed fluctuation index, v i Indicates the wind speed at different locations. represents the average wind speed at different locations, and n represents the number of air flow sensors; The air flow change data at different times of a single point location are obtained. Through the wind speed change at a single point location, the discrete degree of the wind speed data at the single point location is analyzed, the fluctuation of the air flow at the single point location is evaluated, and the single point wind speed fluctuation index is obtained. The calculation formula is as follows: Among them, Psf i represents the wind speed fluctuation index at position i, T represents the number at different times, and t represents the mark at different times. represents the wind speed at position i at time t, represents the average wind speed at location i; The indoor wind speed fluctuation index and the single-point wind speed fluctuation index are obtained, and further correlated to form the air flow anomaly coefficient. The calculation formula is as follows: Among them, Afc represents the air flow anomaly coefficient, Isf represents the indoor wind speed fluctuation index, and Psf i represents the wind speed fluctuation index at position i, n represents the number of air flow sensors, k1 and k2 represent weight coefficients, where 0 < k1 < 1, 0 < k2 < 1, and k1 + k2 = 1; The wind speed fluctuation threshold is pre-set. When the air flow anomaly coefficient is less than the wind speed fluctuation threshold, no processing is performed. When the air flow anomaly coefficient is greater than or equal to the wind speed fluctuation threshold, the single point position where the maximum value of the single-point wind speed fluctuation index is located is obtained and marked as an abnormal point. Acquire the sound signal data and vibration signal data of the abnormal point, and decompose the sound signal and vibration signal data into several IMF components respectively through the VMD algorithm; calculate the correlation coefficient between each IMF component and the original signal, and remove the IMF components with correlation coefficients lower than the preset correlation threshold; reconstruct the remaining IMF components to obtain the reconstructed sound and vibration signals; The reconstructed sound signal and vibration signal are obtained, and the reconstructed sound signal and vibration signal are normalized to map the data to the range of [0,1]; the reconstructed sound signal and vibration signal are fused into a new signal, and the calculation formula corresponding to the new signal is as follows: V(t′)=X(t′)+Y(t′) Among them, V(t') represents the new signal after fusion, X(t') represents the reconstructed sound signal, Y(t') represents the reconstructed vibration signal, and t' represents the time point; The fused new signal is compared with the normal feature signal in the normal feature library, and the correlation coefficient between the fused new signal and the normal feature signal is calculated. The maximum value of all correlation coefficients is taken as the acoustic vibration abnormality coefficient. The calculation formula of the correlation coefficient is as follows: Among them, γ represents the correlation coefficient, t' represents the time point, and m represents the number of time points. Represents the average value of the new signal V(t' after fusion, Represents the average value of the normal characteristic signal C(t').
2. The intelligent security monitoring method according to claim 1, characterized in that: Air flow sensors are installed at different locations in the security area to monitor air flow conditions in real time. Sensors are arranged around the air flow sensors to collect sound signals, vibration signals and image data. The sensors include sound sensors, vibration sensors and cameras.
3. The intelligent security monitoring method according to claim 1, characterized in that: Preset the abnormal sound and vibration threshold, compare the abnormal sound and vibration threshold with the abnormal sound and vibration coefficient, and take corresponding measures according to the comparison results, specifically: When the sound vibration abnormality coefficient is greater than or equal to the sound vibration abnormality threshold, no processing is performed; when the sound vibration abnormality coefficient is less than the sound vibration abnormality threshold, an alarm is automatically issued and the collected data is uploaded to the storage end.
4. An intelligent security monitoring system, used to implement the method described in any one of claims 1 to 3, characterized in that: It includes a data collection module, an air flow detection module, an acoustic vibration signal processing module, an acoustic vibration signal fusion module and an early warning module; among which, Data collection module: install air flow sensors at different locations in the security area to monitor air flow in real time, and arrange sensors around the air flow sensors to collect sound and vibration signal data; The air flow detection module analyzes the discrete degree of wind speed data through the wind speed at different positions and the wind speed changes at a single point, obtains the indoor wind speed fluctuation index and the single-point wind speed fluctuation index, and further correlates the indoor wind speed fluctuation index and the single-point wind speed fluctuation index to form an air flow anomaly coefficient; The acoustic and vibration signal processing module obtains the single point position where the single point wind speed fluctuation index is the maximum when the air flow anomaly coefficient is greater than or equal to the wind speed fluctuation threshold, decomposes the sound signal and vibration signal of the single point position into several IMF components respectively through the VMD algorithm, calculates the correlation coefficient between each IMF component and the original signal, removes the IMF components with correlation coefficients lower than the preset correlation threshold, reconstructs the remaining IMF components, and obtains the reconstructed sound and vibration signals; The acoustic-vibration signal fusion module normalizes the reconstructed sound signal and vibration signal and fuses them into a new signal. The fused new signal is compared with the normal feature signal in the normal feature library, and the correlation coefficient between the fused new signal and all normal feature signals is calculated. The maximum value of all correlation coefficients is taken as the acoustic-vibration abnormality coefficient. The early warning module pre-sets the sound and vibration abnormality threshold, compares the sound and vibration abnormality threshold with the sound and vibration abnormality coefficient, and takes corresponding measures based on the comparison results.
Citation Information
Patent Citations
Intelligent monitoring security and protection system and use method thereof
CN110012271A
A Home Intelligent Security Monitoring Method and System
CN116761049B
Civil air defense alarm intelligent operation and maintenance management method and system
CN116246442A
Electricity utilization inspection system based on Internet of Things platform
CN116365716A
Autonomous navigation robot environment perception control method and system
CN116721512A
