Surveillance video anomaly alarm platform and alarm system based on big data analysis

Abstract

This application discloses a surveillance video anomaly alarm platform and alarm system based on big data analysis. A surveillance video anomaly alarm platform and alarm system based on big data analysis has the beneficial effect of receiving the device number of a data acquisition device, establishing a calculation data set based on the device number of the acquisition device, analyzing the data in the calculation data set via the cloud, performing descriptive statistical analysis on the data features in the calculation data set, clustering and associating the data in the calculation data set based on the descriptive statistical analysis, calling the time series of the calculation data set, performing spatial data analysis on the clustered and associated data, and providing feedback on the anomaly classification in the spatial data analysis. This reduces the computing power occupied by the cloud server. This solves the drawback of traditional surveillance video anomaly analysis methods that occupy a large amount of cloud server computing power, and enables the cloud server to search for abnormal surveillance video devices, and the surveillance video devices themselves to issue alarms for surveillance video anomalies.

Description

Technical Field

[0001] The present application relates to the field of data analysis, and more specifically, to a surveillance video anomaly alarm platform and alarm system based on big data analysis. Background Art

[0002] By analyzing and mining video surveillance data, we can monitor and provide early warnings for security issues. For example, by analyzing abnormal behavior in video data, we can promptly identify and address security threats. Monitoring and early warning can also help identify dangerous incidents. Video surveillance data can provide important insights for urban management and planning. By analyzing and mining video surveillance data, we can understand important information such as population mobility within a city, providing a scientific basis for urban management and planning.

[0003] Traditional surveillance video anomaly analysis methods often rely on cloud data analysis, which consumes a lot of cloud server computing power. The current industry challenge is to identify surveillance video anomalies based on big data analysis while using less cloud server computing power. To address this drawback of traditional surveillance video anomaly analysis methods that consume a lot of cloud server computing power, it is necessary to propose a surveillance video anomaly alarm platform and alarm system based on big data analysis. Summary of the Invention

[0004] The content of this application is used to briefly introduce concepts that will be described in detail in the detailed description section below. The content of this application is not intended to identify key features or essential features of the technical solution for which protection is sought, nor is it intended to limit the scope of the technical solution for which protection is sought.

[0005] To address the technical issues mentioned in the background technology section above, some embodiments of the present application provide a method for anomaly alarming in surveillance videos based on big data analysis, including:

[0006] The device number of the receiving data acquisition device;

[0007] Based on the device number of the acquisition device, a calculation data set is established;

[0008] Perform descriptive statistical analysis on the data features in the calculated data set;

[0009] Based on descriptive statistical analysis, the data of the calculation data set are clustered and associated;

[0010] Call the time series of the calculation data set and perform spatial data analysis on the cluster-related data;

[0011] Anomaly classification in feedback space data analysis;

[0012] Select an anomaly classification in spatial data analysis;

[0013] Determine whether the abnormal classification associated with the cluster is a surveillance video device;

[0014] If the abnormal category associated with the cluster is not a surveillance video device, then return to select an abnormal category in the spatial data analysis until all abnormal categories are selected;

[0015] If the abnormal category associated with the cluster is a surveillance video device, the abnormal information of the surveillance video is obtained based on the device-side analysis algorithm, and an abnormal category in the spatial data analysis is selected until all abnormal categories are selected.

[0016] Furthermore, based on the device network access information, the system type of the device is determined;

[0017] If the system type of the device is the first system type, obtaining the serial number of the device using the service code;

[0018] If the system type of the device is the second system type, obtaining the device identifier using the universal identification code in the device network access information;

[0019] If the system type of the device is the third system type, the disk call function is used to obtain the serial number of the storage volume of the device.

[0020] Furthermore, based on the system type of the device, a framework for receiving the computational data set is established;

[0021] Incorporating the copula of the first system type into the receiving framework of the computational dataset;

[0022] Incorporating the copula of the second system type into the receiving framework of the computational dataset;

[0023] Incorporating the third system type's link function into the receiving framework of the computational dataset;

[0024] Based on the receiving framework of the calculation data set, the serial number of the device obtained in the first system type, the identifier of the device obtained in the second system type, and the serial number of the storage volume of the device obtained in the third system type are included in the receiving framework.

[0025] Further, analyzing the serial numbers of the devices obtained in the first system type based on the connection function of the first system type of the receiving frame of the calculation data set;

[0026] Incorporating analysis results of serial numbers of devices obtained in the first system type into a calculation data set;

[0027] analyzing identifiers of devices obtained in the second system type based on a connection function of the second system type for the received frame of the computing data set;

[0028] Incorporating the analysis results of the identifiers of the devices obtained in the second system type into the calculation data set;

[0029] analyzing the serial number of the storage volume of the device obtained in the third system type based on a connection function of the third system type of the receiving frame of the calculation data set;

[0030] The analysis results of the serial numbers of the storage volumes of the devices obtained in the third system type are incorporated into the calculation data set.

[0031] Further, receiving a database query statement;

[0032] Establish a database query statement channel between the calculation data set and the database;

[0033] Selecting a characteristic data of the calculation data set; the characteristic data includes a serial number of a device obtained in the first system type, an identifier of a device obtained in the second system type, and a serial number of a storage volume of a device obtained in the third system type;

[0034] Based on the database query statement channel, call the corresponding data in the database;

[0035] Determine the central position measure, dispersion measure and data distribution shape of the corresponding data of the called database;

[0036] Return one feature data of the selected calculation data set until all feature data are selected.

[0037] Furthermore, based on the nonlinear function, the central position measurement, dispersion degree measurement and data distribution shape of a data in the database are clustered and associated;

[0038] Obtaining at least one time series of the data in the database;

[0039] If the number of the time sequence of the data in the database is 1, the time sequence of the data in the database is assigned the sequence number 1;

[0040] If the number of time series of the data in the database is greater than 1, all time series of the data in the database are assigned sequence numbers in order of time complexity.

[0041] Furthermore, a histogram function for spatial data analysis is established;

[0042] Assign the first vertical dimension of the square function to the time complexity parameter;

[0043] Assign the second vertical dimension of the square function to the labeled parameter of the data;

[0044] Assign the third vertical dimension of the square function to the quantitative parameter of the time series;

[0045] Call to calculate the time series of the dataset;

[0046] Associating all data of the computational dataset with the data of the database;

[0047] The associated data are incorporated into the spatial data analysis histogram to form a spatial data analysis histogram about the time series of the data.

[0048] Furthermore, the surface formed by the square function of spatial data analysis is differentiated;

[0049] Obtaining differential characteristic points parallel to the first perpendicular dimensional differential line;

[0050] Assign the data corresponding to the differential feature points to the abnormal data;

[0051] Find the device corresponding to the differential feature point;

[0052] The device number of the device corresponding to the differential feature point is included in the abnormal classification.

[0053] Furthermore, the video content of the device classified as abnormal is digitized;

[0054] identifying real-time digital video corresponding to digital video content;

[0055] Detecting the real-time digital video stream, and detecting video behavior data based on the picture content and the sound content;

[0056] Identify abnormal events corresponding to video traffic based on preset abnormal trigger events.

[0057] Some embodiments of the present application further provide a surveillance video anomaly alarm system based on big data analysis, including:

[0058] A server, configured to execute the monitoring video abnormality alarm method based on big data analysis;

[0059] The data acquisition devices are all connected to the server for communication.

[0060] In summary:

[0061] The system receives the device number of the data collection device, creates a computational dataset based on the device number, analyzes the data in the computational dataset using the cloud, performs descriptive statistical analysis on the data features in the computational dataset, clusters and associates the data based on the descriptive statistical analysis, calls the time series of the computational dataset, performs spatial data analysis on the clustered data, and provides feedback on anomaly classifications in the spatial data analysis. This reduces the computing power used by the cloud server.

[0062] Select an anomaly category from the spatial data analysis. Determine whether the clustered anomaly category is a surveillance video device. If the clustered anomaly category is not a surveillance video device, return to select an anomaly category from the spatial data analysis until all anomaly categories have been selected. If the clustered anomaly category is a surveillance video device, obtain information about surveillance video anomalies based on the device-side analysis algorithm and return to select an anomaly category from the spatial data analysis until all anomaly categories have been selected. This solves the problem of traditional surveillance video anomaly analysis methods occupying a large amount of cloud server computing power. It enables cloud servers to search for abnormal surveillance video devices, and the surveillance video devices themselves to issue alerts regarding surveillance video anomalies.

[0063] By analyzing and mining video surveillance data, we can monitor and provide early warnings for security issues. For example, by analyzing abnormal behavior in video data, we can promptly identify and address security threats. Monitoring and early warning can also help identify dangerous incidents. Video surveillance data can provide important insights for urban management and planning. By analyzing and mining video surveillance data, we can understand important information such as population mobility within a city, providing a scientific basis for urban management and planning. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The drawings constituting a part of this application are used to provide a further understanding of this application and make other features, purposes and advantages of this application more apparent. The drawings and descriptions of the exemplary embodiments of this application are used to explain this application and do not constitute an improper limitation on this application.

[0065] In addition, throughout the drawings, the same or similar reference numerals represent the same or similar elements. It should be understood that the drawings are schematic and that the elements and components are not necessarily drawn to scale.

[0066] In the attached figure:

[0067] Figure 1 This example provides a method flow chart of a surveillance video abnormality alarm method based on big data analysis.

[0068] Figure 2 This example provides a structural connection diagram of a surveillance video anomaly alarm system based on big data analysis.

[0069] Reference numerals:

[0070] 100-Server; 200-Data acquisition equipment. DETAILED DESCRIPTION

[0071] The following will describe embodiments of the present application in more detail with reference to the accompanying drawings. Although certain embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be construed as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are for illustrative purposes only and are not intended to limit the scope of protection of the present application.

[0072] It should also be noted that, for ease of description, only the parts related to the invention are shown in the drawings. In the absence of conflict, the embodiments and features in the embodiments of this application can be combined with each other.

[0073] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments.

[0074] Reference Figure 1 , a monitoring video abnormality alarm method based on big data analysis, comprising:

[0075] S100, receiving the device number of the data acquisition device.

[0076] Specifically, based on the device network access information, the system type of the device is determined. If the system type of the device is the first system type, the serial number of the device is obtained using the service code.

[0077] Using TelephonyManager to obtain the IMEI is only applicable to devices with phone functionality and requires permission. On Android 6.0 and above, you need to declare the permission in AndroidManifest.xml.

[0078] Use Build.SERIAL, which is applicable to Android 8.0 and earlier versions. You can use Build.SERIAL to obtain the device serial number, but this is an unstable unique identifier that depends on the manufacturer's implementation.

[0079] If the system type of the device is the second system type, the device identifier is obtained using the universal identification code in the device network access information.

[0080] Obtain IDFA, which is the advertising identifier assigned by Apple to the user's device and can be used in scenarios such as ad tracking.

[0081] Get UUID, a universally unique identifier, which can be generated and used by the application in the APP according to the UUID generation standard.

[0082] If the system type of the device is the third system type, the disk call function is used to obtain the serial number of the storage volume of the device.

[0083] Use the GetVolumeInformation function, which can obtain information about the specified disk volume, including the volume serial number, which can be used as a device identifier.

[0084] More specifically, based on the system type of the device, a framework for receiving the calculation dataset is established. A connection function for the first system type is incorporated into the framework for receiving the calculation dataset. A connection function for the second system type is incorporated into the framework for receiving the calculation dataset. A connection function for the third system type is incorporated into the framework for receiving the calculation dataset. Based on the framework for receiving the calculation dataset, the serial number of the device obtained from the first system type, the identifier of the device obtained from the second system type, and the serial number of the storage volume of the device obtained from the third system type are incorporated into the framework.

[0085] getAndroidId(), this is the most commonly used method, returns the device's ANDROID ID, which is usually unique, but some devices may change after a reset. GetDeviceSerial(): Gets the device's serial number, applicable to Android 8.0 and below. After Android 9.0, Google does not recommend using the serial number as a unique identifier. GetId(): Gets the device's IMEI number through the phone manager, requires permission, and is applicable to devices with phone functions. Permission Management: The code includes a dynamic permission request section to ensure that there is no crash when obtaining the IMEI on Android 6.0 and above.

[0086] S200: Establish a calculation data set based on the device number of the acquisition device.

[0087] Specifically, based on the connection function of the first system type in the receiving frame of the calculation data set, the serial number of the device obtained in the first system type is analyzed.

[0088] The analysis results of the serial numbers of the devices obtained in the first system type are incorporated into the calculation data set.

[0089] Based on calculating a connection function of the second system type for the received frame of the data set, identifiers of the devices obtained in the second system type are analyzed.

[0090] The analysis results of the identifiers of the devices obtained in the second system type are incorporated into the calculation data set.

[0091] Based on calculating a connection function of the third system type of the receiving frame of the data set, the serial number of the storage volume of the device obtained in the third system type is analyzed.

[0092] The analysis results of the serial numbers of the storage volumes of the devices obtained in the third system type are incorporated into the calculation data set.

[0093] S300, performing descriptive statistical analysis on the data features in the calculation data set.

[0094] Specifically, a database query statement is received. A database query statement channel is established between a computational dataset and a database. Feature data of the computational dataset is selected. The feature data includes a serial number of a device obtained in a first system type, an identifier of a device obtained in a second system type, and a serial number of a storage volume of a device obtained in a third system type. Based on the database query statement channel, corresponding data of the database is called. A central position measure, a dispersion measure, and a data distribution shape of the corresponding data of the called database are determined. A feature data of the selected computational dataset is returned until all feature data are selected.

[0095] By calculating the data features in the data set and performing descriptive statistical analysis, we can find the overall trend of change caused by some fundamental factors over a long period of time. We can then use these data features to track abnormal data.

[0096] The descriptive statistical analysis results of these data characteristics include cyclic changes, irregular changes, etc.

[0097] S400, based on descriptive statistical analysis, cluster association is performed on the data of the calculation data set.

[0098] Specifically, based on a nonlinear function, the central position measure, dispersion measure, and data distribution shape of a data item in the database are clustered and associated. At least one time series of the data item in the database is obtained. If the number of time series of the data item in the database is 1, the time series of the data item in the database is assigned a sequence number of 1. If the number of time series of the data item in the database is greater than 1, all time series of the data item in the database are assigned sequence numbers in order of time complexity.

[0099] S500, calling the time series of the calculation data set and performing spatial data analysis on the cluster-related data.

[0100] Specifically, a spatial data analysis histogram function is established. The first vertical dimension of the histogram function is assigned a time complexity parameter. The second vertical dimension of the histogram function is assigned a data label parameter. The third vertical dimension of the histogram function is assigned a time series quantity parameter. The time series of the computational dataset is called. All data in the computational dataset is associated with the data in the database. The associated data is incorporated into the spatial data analysis histogram function to form a spatial data analysis histogram function for the data's time series.

[0101] After S500 , the surface formed by the spatial data analysis histogram function is differentiated. Differential feature points parallel to the first perpendicular differential line are obtained. The data corresponding to the differential feature points are assigned to abnormal data. The device corresponding to the differential feature points is searched. The device number of the device corresponding to the differential feature point is included in the abnormal classification.

[0102] S600, anomaly classification in feedback space data analysis.

[0103] S700: Select an anomaly classification in spatial data analysis.

[0104] S800: Determine whether the abnormal classification associated with the cluster is a surveillance video device.

[0105] S910: If the abnormal classification associated with the cluster is not a surveillance video device, return to selecting an abnormal classification in the spatial data analysis until all abnormal classifications are selected.

[0106] S920: If the abnormal classification associated with the cluster is a surveillance video device, obtain the abnormal information of the surveillance video based on the device-side analysis algorithm, and return to select an abnormal classification in the spatial data analysis until all abnormal classifications are selected.

[0107] Specifically, the device video content of the abnormal classification is digitized. The real-time digital video corresponding to the digital video content is identified. The real-time digital video stream is detected, and video behavior data is detected based on the image content and the sound content. The abnormal event corresponding to the video flow is identified based on a preset abnormal trigger event.

[0108] This embodiment relates to a monitoring video abnormality alarm method based on big data analysis.

[0109] The system receives the device number of the data collection device, creates a computational dataset based on the device number, analyzes the data in the computational dataset using the cloud, performs descriptive statistical analysis on the data features in the computational dataset, clusters and associates the data based on the descriptive statistical analysis, calls the time series of the computational dataset, performs spatial data analysis on the clustered data, and provides feedback on anomaly classifications in the spatial data analysis. This reduces the computing power used by the cloud server.

[0110] Select an anomaly category from the spatial data analysis. Determine whether the clustered anomaly category is a surveillance video device. If the clustered anomaly category is not a surveillance video device, return to select an anomaly category from the spatial data analysis until all anomaly categories have been selected. If the clustered anomaly category is a surveillance video device, obtain information about surveillance video anomalies based on the device-side analysis algorithm and return to select an anomaly category from the spatial data analysis until all anomaly categories have been selected. This solves the problem of traditional surveillance video anomaly analysis methods occupying a large amount of cloud server computing power. It enables cloud servers to search for abnormal surveillance video devices, and the surveillance video devices themselves to issue alerts regarding surveillance video anomalies.

[0111] By analyzing and mining video surveillance data, we can monitor and provide early warnings for security issues. For example, by analyzing abnormal behavior in video data, we can promptly identify and address security threats. Monitoring and early warning can also help identify dangerous incidents. Video surveillance data can provide important insights for urban management and planning. By analyzing and mining video surveillance data, we can understand important information such as population mobility within a city, providing a scientific basis for urban management and planning.

[0112] Reference Figure 2 , a surveillance video abnormality alarm system based on big data analysis, including:

[0113] The server 100 is used to execute the monitoring video abnormality alarm method based on big data analysis.

[0114] The data collection devices 200 are all in communication connection with the server 100 .

[0115] This embodiment relates to a surveillance video anomaly alarm system based on big data analysis. Server 100 monitors and provides early warnings for security issues by analyzing and mining video surveillance data. Data acquisition device 200 monitors the video device itself and generates alerts for surveillance video anomalies. For example, by analyzing abnormal behavior in video data, security threats can be promptly identified and addressed. Monitoring and early warning can identify dangerous events. Video surveillance data can provide important insights for urban management and planning. By analyzing and mining video surveillance data, important information such as population mobility within a city can be understood, providing a scientific basis for urban management and planning.

[0116] The above description is only an illustration of some preferred embodiments of the present application and the technical principles used. Those skilled in the art should understand that the scope of the invention involved in the embodiments of the present application is not limited to the technical solutions formed by the specific combination of the above-mentioned technical features, but should also cover other technical solutions formed by any combination of the above-mentioned technical features or their equivalent features without departing from the above-mentioned inventive concept. For example, the above-mentioned features and the technical features with similar functions disclosed in the embodiments of the present application (but not limited to) are replaced with each other to form a technical solution.

Claims

1. A surveillance video abnormality alarm method based on big data analysis, comprising: The device number of the receiving data acquisition device; Based on the device number of the acquisition device, a calculation data set is established; Perform descriptive statistical analysis on the data features in the calculated data set; Based on descriptive statistical analysis, the data of the calculation data set are clustered and associated; Call the time series of the calculation data set and perform spatial data analysis on the cluster-related data; Establish a spatial data analysis square function; assign the first vertical dimension of the square function to the time complexity parameter; assign the second vertical dimension of the square function to the data label parameter; assign the third vertical dimension of the square function to the time series quantity parameter; call the time series of the calculation data set; associate all data of the calculation data set with the data in the database; The associated data is incorporated into the spatial data analysis histogram to form a spatial data analysis histogram about the time series of the data; Anomaly classification in feedback space data analysis; Select an anomaly classification in spatial data analysis; Determine whether the abnormal classification associated with the cluster is a surveillance video device; If the abnormal category associated with the cluster is not a surveillance video device, then return to select an abnormal category in the spatial data analysis until all abnormal categories are selected; If the abnormal category associated with the cluster is a surveillance video device, the abnormal information of the surveillance video is obtained based on the device-side analysis algorithm, and an abnormal category in the spatial data analysis is selected until all abnormal categories are selected.

2. The method for anomaly alarm of surveillance video based on big data analysis according to claim 1 is characterized in that: The device number of the receiving data acquisition device includes: Determine the device's system type based on the device's network access information; If the system type of the device is the first system type, obtaining the serial number of the device using the service code; If the system type of the device is the second system type, obtaining the device identifier using the universal identification code in the device network access information; If the system type of the device is the third system type, the disk call function is used to obtain the serial number of the storage volume of the device.

3. The method for monitoring video abnormality alarm based on big data analysis according to claim 2 is characterized in that: The device number of the receiving data acquisition device also includes: Establish a framework for receiving computational data sets based on the device's system type; Incorporating the copula of the first system type into the receiving framework of the computational dataset; Incorporating the copula of the second system type into the receiving framework of the computational dataset; Incorporating the third system type's link function into the receiving framework of the computational dataset; Based on the receiving framework of the calculation data set, the serial number of the device obtained in the first system type, the identifier of the device obtained in the second system type, and the serial number of the storage volume of the device obtained in the third system type are included in the receiving framework.

4. The method for anomaly alarm of surveillance video based on big data analysis according to claim 3 is characterized in that: The step of establishing a calculation data set based on the device number of the acquisition device includes: analyzing serial numbers of devices obtained in the first system type based on a connection function of the first system type of the received frame of the computing data set; Incorporating analysis results of serial numbers of devices obtained in the first system type into a calculation data set; analyzing identifiers of devices obtained in the second system type based on a connection function of the second system type for the received frame of the computing data set; Incorporating the analysis results of the identifiers of the devices obtained in the second system type into the calculation data set; analyzing the serial number of the storage volume of the device obtained in the third system type based on a connection function of the third system type of the receiving frame of the calculation data set; The analysis results of the serial numbers of the storage volumes of the devices obtained in the third system type are incorporated into the calculation data set.

5. The method for anomaly alarm of surveillance video based on big data analysis according to claim 4 is characterized in that: The descriptive statistical analysis of the data features in the calculation data set includes: Receive database query statements; Establish a database query statement channel between the calculation data set and the database; Selecting a characteristic data of the calculation data set; the characteristic data includes a serial number of a device obtained in the first system type, an identifier of a device obtained in the second system type, and a serial number of a storage volume of a device obtained in the third system type; Based on the database query statement channel, call the corresponding data in the database; Determine the central position measure, dispersion measure and data distribution shape of the corresponding data of the called database; Return one feature data of the selected calculation data set until all feature data are selected.

6. The method for anomaly alarm of surveillance video based on big data analysis according to claim 5, characterized in that: The clustering and association of the data in the calculation data set based on descriptive statistical analysis includes: Based on nonlinear functions, the central position measurement, dispersion measurement and data distribution shape of a data in the database are clustered and associated; Obtaining at least one time series of the data in the database; If the number of the time sequence of the data in the database is 1, the time sequence of the data in the database is assigned the sequence number 1; If the number of time series of the data in the database is greater than 1, all time series of the data in the database are assigned sequence numbers in order of time complexity.

7. The method for anomaly alarm of surveillance video based on big data analysis according to claim 6, characterized in that: Prior to the anomaly classification in the feedback space data analysis, the following steps are included: Differentiate the surface formed by the histogram function for spatial data analysis; Obtaining differential characteristic points parallel to the first perpendicular dimensional differential line; Assign the data corresponding to the differential feature points to the abnormal data; Find the device corresponding to the differential feature point; The device number of the device corresponding to the differential feature point is included in the abnormal classification.

8. The method for anomaly alarm of surveillance video based on big data analysis according to claim 7, characterized in that: If the cluster-associated abnormal classification is a surveillance video device, then based on the device-side analysis algorithm, the monitoring video abnormality information is obtained, and an abnormal classification in the spatial data analysis is selected until all abnormal classifications are selected, including: Digital processing of device video content for abnormal classification; identifying real-time digital video corresponding to digital video content; Detect real-time digital video streams and detect video behavior data based on picture and sound content; Identify abnormal events corresponding to video traffic based on preset abnormal trigger events.

9. A surveillance video anomaly alarm system based on big data analysis, comprising: A server, configured to execute the monitoring video abnormality alarm method based on big data analysis according to any one of claims 1 to 8; The data acquisition devices are all connected to the server for communication.

Images