Smart community data security storage method and system

Abstract

According to the smart community data security storage method and system, the abnormal database is arranged, so that abnormal data in monitoring data can be copied and stored, the abnormal database is equivalent to a special library focusing on abnormal conditions, and when related problems occur in a community, the abnormal data in the monitoring data can be conveniently stored. The abnormal data can be quickly positioned and called for targeted investigation and processing, the efficiency of dealing with abnormal events is effectively improved, the storage time of the abnormal database is longer than that of a normal database, and long-time storage is beneficial to quick calling of the abnormal data for screening and investigation when an accident occurs in a community, so that the safety of the community is improved. Compared with searching from a huge normal database, the method is more convenient and efficient.

Description

Technical Field

[0001] The present invention relates to the technical field of smart community data storage, and in particular to a smart community data security storage method and system. Background Art

[0002] A smart community refers to a model that uses various intelligent technologies and methods to integrate various existing service resources in the community to provide community residents with a variety of convenient services such as government affairs, business, entertainment, education, medical care, and life mutual assistance.

[0003] Smart communities include information management, intelligent facilities, personalized services, green environmental protection and security monitoring. In the process of realizing smart communities, a large amount of data foundation is needed. In order to facilitate the management within the community, more surveillance cameras will be installed inside the community. The surveillance cameras in the community can provide video data for the community, and can be analyzed by extracting surveillance data when abnormalities occur within the community. However, the existing surveillance data are stored together, and there is no classification of normal and abnormal surveillance data. When the surveillance camera detects abnormal data, it is still stored together with the overall surveillance data. This leads to the need to find abnormal data in the overall surveillance data when searching for abnormal data in the community, resulting in an increase in the workload of searching for surveillance. In addition, in the storage of existing surveillance data, the surveillance data will be deleted after being saved for a certain period of time. If the abnormal data is deleted together with the normal surveillance data, it will lead to the inability to check in the later retrieval.

[0004] To this end, we separated an anomaly database for storing abnormal data from the original monitoring data. By analyzing the monitoring data, we extracted the abnormal data from the monitoring data and copied and saved them in the anomaly database. The storage time of the anomaly database is longer than that of the original monitoring data to ensure the inspection of abnormal data after a long time, thereby facilitating the management within the community. Summary of the invention

[0005] In order to solve the above problems, the present invention proposes a smart community data security storage method and system, so as to more accurately solve the problem that the existing monitoring data mentioned above are all stored together, and the monitoring data are not classified as normal and abnormal. When the monitoring camera detects abnormal data, it is still stored together with the overall monitoring data. As a result, when searching for abnormal data in the community, it is necessary to search for abnormal data in the overall monitoring data, which increases the workload of searching for monitoring. In addition, in the existing storage of monitoring data, the monitoring data will be deleted after being saved for a certain period of time. If the abnormal data is deleted together with the normal monitoring data, it will make it impossible to check it in the later retrieval.

[0006] The present invention is achieved through the following technical solutions: The present invention proposes a method for securely storing data in a smart community, comprising the following steps: Step 1: Classify the database into long-term database, normal database and abnormal database; Step 2: Data collection: collect information about people in the community through manual and automatic collection, and save the data in a long-term database; Step 3: Tag surveillance cameras. Label the surveillance cameras in the community according to different situations. Step 4: Compare monitoring records. After the monitoring data is collected, it is entered into the normal database, and a comparative analysis is performed between the normal database and the long-term database.

[0007] Step 5: Sorting out abnormal data. The abnormal data generated in the comparison between the normal database and the long-term database are classified as abnormal data. The abnormal data include abnormal behavior data and abnormal face recognition data. The abnormal behavior data are divided into abnormal behavior trajectory data and abnormal behavior time data. There are two main types of facial recognition anomalies: one is that the surveillance data detected by the surveillance camera cannot match the facial information in the long-term database; the other is that the surveillance camera cannot collect facial information; Abnormal behavior trajectory: surveillance cameras throughout the community continuously capture the behavior trajectory of people. When the behavior trajectory of a person corresponding to the same facial data obviously does not conform to the normal traffic pattern in the community, it will be judged as abnormal behavior trajectory; Behavior time anomaly: Combine time information to set activity time. If surveillance cameras collect data outside the set time, these data are considered abnormal. Step 6: Processing of abnormal data. When abnormal data occurs, the property management will screen and filter the data, and the abnormal data will be stored for a long time.

[0008] In a possible implementation, the data collected in step 2 includes facial information, personnel data, residential information, family relationships, etc.

[0009] In one possible implementation, in step three, surveillance cameras within the community are labeled according to different situations, surveillance cameras within the same area are set to the same initial letter, and different digital marks are made according to the route, and the digital marks between adjacent surveillance cameras are arranged according to an arithmetic progression.

[0010] In a possible implementation, the retention time of the abnormal database is longer than the retention time of the normal database.

[0011] In a possible implementation, the inability of the surveillance camera in the face recognition anomaly to collect face information is determined by comparing face recognition anomaly data of adjacent surveillance cameras.

[0012] In one possible implementation, a smart community data security storage system includes a data acquisition module, a monitoring module, a data storage module and a data processing module.

[0013] In a possible implementation, the data acquisition module is divided into a manual acquisition module and an automatic acquisition module. The manual acquisition module is used to manually record data and then enter it into the system. The electronic entry is made with the help of a database management system client tool or a simple entry interface program, and is connected to a back-end long-term database to store data. The automatic acquisition module collects facial images through hardware equipment with human recognition function.

[0014] In a possible implementation, the monitoring module is composed of internal monitoring cameras in the community.

[0015] In a possible implementation, the data storage module includes a long-term database module, a normal database module and an abnormal database module.

[0016] In one possible implementation, the data processing module includes a monitoring record comparison module. After the monitoring data is entered into the normal database, the normal database is compared with the long-term database, abnormal data is identified and classified, and when abnormal data is detected, the property personnel are reminded through an automated notification system, and GIS is integrated to provide them with abnormal area map navigation function.

[0017] Beneficial effects of the present invention: The setting of the exception database can copy and store the abnormal data in the monitoring data. It is equivalent to a "special library" focusing on abnormal situations. When related problems arise in the community, it is convenient to quickly locate and call the abnormal data for targeted investigation and processing, effectively improving the efficiency of responding to abnormal events.

[0018] Through the face recognition function of the surveillance camera, the facial data in the monitoring data can be analyzed. Further, by labeling the surveillance camera, the travel routes of community residents can be analyzed through the facial data, which facilitates the recording of abnormal personnel data.

[0019] By adding a time threshold recording function to the surveillance camera, the personnel data outside the time threshold can be classified as abnormal data, and then the abnormal personnel in the abnormal time period can be recorded.

[0020] Data is collected manually and automatically. Manual collection makes it easier for people who are not familiar with electronic devices to register information and facilitates the collection of information of people within the community.

[0021] By setting up a long-term database, a normal database and an abnormal database, the monitoring data within the community and the basic information of the people within the community are classified and stored, and the abnormal data in the normal database is copied and stored in the abnormal database. The data within the community is carefully classified to facilitate data management.

[0022] Since the storage time of the abnormal database is longer than that of the normal database, long-term preservation helps to quickly call up abnormal data for screening and investigation when an accident occurs in the community. This is more convenient and efficient than searching from a huge normal database. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 A schematic diagram of a process of a smart community data security storage method proposed by the present invention; Figure 2 This is a system diagram of a smart community data security storage system proposed by the present invention. DETAILED DESCRIPTION

[0024] In order to more clearly and completely illustrate the technical solution of the present invention, the present invention is further described below in conjunction with the accompanying drawings.

[0025] Please refer to Figure 1 The present invention proposes a method for securely storing data in a smart community, comprising the following steps: Step 1: Classify the database into long-term database, normal database and abnormal database.

[0026] The long-term database is used to store basic information about people within the community.

[0027] The normal database is used to store all surveillance data within the community, covering all kinds of information such as images and sounds captured by each surveillance camera at different times, providing a comprehensive data basis for subsequent comparative analysis, traceability query and other operations; The abnormal database specifically copies and stores abnormal data that is screened out during the comparative analysis of the normal database. It is equivalent to a "special database" that focuses on abnormal situations, so that when relevant problems arise in the community, it can quickly locate and call abnormal data for targeted investigation and processing, effectively improving the efficiency of responding to abnormal events.

[0028] Step 2: Data collection: collect information about people in the community through manual and automatic collection, including personnel data, residence information, family relationships, etc. These data are saved in a long-term database.

[0029] Manual form filling and input collection can set up special registration points in community service centers, equipped with easy-to-operate forms and input devices to ensure the accuracy and completeness of information. At the same time, for some special cases, such as the elderly or residents who are not familiar with electronic devices, manual form filling and input collection provides a convenient way to collect their key data, ensuring that the data of every resident in the community can be fully collected without missing any important information points, laying a solid foundation for the community's precise management and considerate services.

[0030] Automatic collection mainly collects facial data through facial recognition equipment.

[0031] Step 3: Mark surveillance cameras. Label the surveillance cameras in the community according to different situations. Set the surveillance cameras in the same area to the same initial letter, and use different digital labels according to the route. The digital labels between adjacent surveillance cameras are arranged in an arithmetic progression. For example, the surveillance cameras in the same area are all set to the initial letter A, and the adjacent surveillance cameras are A01 and A02. This marking method enables the rapid location of specific surveillance cameras and their coverage areas in subsequent data viewing, anomaly tracking, and equipment maintenance, thereby improving overall management efficiency.

[0032] Step 4: Compare monitoring records. After the monitoring data is collected, it is entered into the normal database, and a comparative analysis is performed between the normal database and the long-term database.

[0033] Step 5: Sorting of abnormal data. The abnormal data generated in the comparison between the normal database and the long-term database are classified as abnormal data. The abnormal data are copied and stored in the abnormal database. The abnormal data include behavioral abnormal data and face recognition abnormal data. The behavioral abnormal data are divided into behavioral trajectory abnormal data and behavioral time abnormal data.

[0034] Facial recognition anomalies mainly manifest themselves in two situations. One is that the surveillance data detected by the surveillance camera cannot be matched with the facial information in the long-term database, that is, an unfamiliar face appears. The other is that the surveillance camera is unable to collect facial information due to various reasons, such as the face is blocked, the surveillance camera malfunctions, etc., resulting in failure to obtain facial information.

[0035] Abnormal behavior trajectory relies on the surveillance cameras throughout the community to continuously capture the behavior trajectory of people. When the behavior trajectory of the person corresponding to the same facial data obviously does not conform to the normal traffic pattern in the community, such as wandering around a certain door for a long time without any reason, or frequently and disorderly entering and exiting different unit buildings, it will be judged as abnormal behavior trajectory. Combined with the markings of the surveillance camera, the movement route of the same facial data can be collected. For example, A1, A2 and A3 all collect the data of a person, and through time judgment, A1, A2 and A3 can be arranged in order to judge the movement trajectory of the person, and compared with the pre-set abnormal judgment criteria, such as moving back and forth at two points, the data reflects that in a period of time, several surveillance cameras have collected the same facial data back and forth, so it can be judged as abnormal data.

[0036] Behavior time anomalies are combined with time information to set activity time. For example, the normal activity range is from 6:00 to 24:00. If surveillance cameras collect data outside the set time, these data will be judged as abnormal.

[0037] Step 6: Processing of abnormal data. When abnormal data is detected, the community contacts the property management to verify the data. For example, if the facial information does not belong to the community’s internal personnel, the property management will register the information to determine the purpose of the person’s entry and exit. When the behavior trajectory and time are abnormal, the property management can directly reach the abnormal area for investigation.

[0038] When the community cannot collect facial information for abnormal data, two situations may occur. One is that facial information cannot be collected due to deliberate obstruction, and the other is that the installation location of the surveillance camera cannot perform facial recognition or the surveillance camera is damaged or covered by debris and cannot work normally. By comparing the data of multiple adjacent surveillance cameras, the surveillance camera that detects the abnormality is compared with the data of other adjacent surveillance cameras. If the abnormal data detected by the surveillance camera is much larger than that of other adjacent surveillance cameras, it means that the surveillance camera is damaged. For example, A1 detects a large amount of abnormal facial recognition data. At this time, it will compare the data with the adjacent A2 and A3. If the data of the three are not much different, it indicates that the data abnormality is caused by deliberate obstruction. If the data of A1 is much higher than the abnormal data collected by A2 and A3, it means that the surveillance camera at A1 is installed in the wrong position or the surveillance camera is damaged. The property department can be notified to carry out repairs in time and clean up the abnormal data.

[0039] The database preservation time and related considerations include the following differences: the preservation time of abnormal databases is longer than that of normal databases. Normal databases will be cleaned up after being preserved for half a year, while the preservation time of abnormal databases can reach one year or more. This is mainly because the amount of data in abnormal databases is much smaller than that in normal databases, and it will not take up too much storage space, so it has the conditions for long-term storage. From a practical application perspective, preserving abnormal databases for a long time helps to quickly call abnormal data directly through the database when various accidents occur within the community, and quickly screen and investigate abnormal personnel. Compared with directly searching for relevant data from a huge normal database, the operation is more convenient and efficient.

[0040] refer to Figure 2 , Based on the above data storage method, the required system support system is proposed, as follows: It includes a data acquisition module, a monitoring module, a data storage module and a data processing module.

[0041] The data collection module includes a manual collection module and an automatic collection module. The manual collection module records manual data and enters the manually collected data into the system. The electronic entry part uses the client tools provided by the database management system (such as MySQL, Oracle, etc.) or develops a simple entry interface program to achieve connection with the back-end long-term database and store the manually collected data accurately.

[0042] The automatic acquisition module uses mature face recognition technology and corresponding hardware equipment, such as smart cameras based on deep learning face recognition algorithms (such as convolutional neural network CNN related algorithms). Its built-in image processing chip can pre-process the collected face images (including image grayscale, normalization, face detection and other operations), and then transmit the processed face data to the back-end system through network communication protocols (such as TCP / IP protocol) for storage and subsequent processing.

[0043] The monitoring module manages the surveillance cameras in the community by marking them, that is, setting the surveillance cameras in the same area to the same initial letter, and assigning different digital tags to the cameras in the form of arithmetic progression according to the route. The database table can be used to store the camera tag information, and each record contains the camera number (such as A01), the area where it is located, the coverage range coordinates and other fields. Develop the corresponding management interface for community managers to enter, modify and query the tag information, and use database operation statements (such as SQL statements) in the background to implement the addition, deletion, modification and query of these tag data.

[0044] The data storage module is divided into long-term database module, normal database module and abnormal database module.

[0045] The long-term database module is responsible for storing the data collected in the data collection module.

[0046] The normal database module is used to store the monitoring data collected by the monitoring module. It must have a large-capacity storage capacity to cope with the long-term and multi-regional monitoring data accumulation, and support efficient data writing and query operations to facilitate the subsequent comparative analysis, traceability query and other functions. Use a distributed file storage system (such as Hadoop Distributed File System-HDFS, suitable for massive data storage and management) or a traditional relational database (such as SQL Server, PostgreSQL, etc., which has advantages in data correlation query) to build a normal database, and choose an appropriate storage architecture and configuration based on the actual size of the community and the expected growth in data volume.

[0047] The abnormal database module is implemented using a relational database. Considering that its data volume is much smaller than that of a normal database, the hardware resource configuration can be relatively streamlined, but attention should be paid to the index design of the data in order to improve data retrieval efficiency, such as using appropriate B-tree indexes to optimize query performance.

[0048] The data processing module includes a monitoring record comparison module. After entering the collected monitoring data into the normal database, it conducts comparative analysis between the normal database and the long-term database. It can accurately identify abnormal data and classify and organize these abnormal data according to the rules.

[0049] Clarify the rules for comparing normal databases and long-term databases. For monitoring data, this may involve multiple aspects such as facial information matching and behavior trajectory analysis. For facial information, use the face recognition algorithm to compare the faces in the monitoring data with the facial information of people in the community stored in the long-term database. A feature vector-based comparison method can be used to calculate the distance between the feature vectors of two faces (such as the Euclidean distance). If the distance exceeds the set threshold, it is considered that the facial information does not match, and it may be an unfamiliar face, which is an abnormal situation.

[0050] In terms of behavioral trajectory, the time sequence and residence time of people appearing under different surveillance cameras are analyzed. For example, if a person moves between two surveillance cameras for too long or too short a time, which is inconsistent with the normal passage mode, it is judged as abnormal behavior trajectory. At the same time, the same face data frequently appears in different areas or in prohibited areas, which is also considered abnormal.

[0051] Ensure that a valid connection can be established between the abnormal database and the normal database. For relational databases, connection parameters can be configured through database connection strings (such as the connection string when using JDBC to connect to a MySQL database in Java), including database IP address, port number, user name, password, and other information.

[0052] Develop a data replication program or use the replication function of the database. For simple relational databases, such as MySQL, you can use SQL statements (such as INSERT INTO...SELECT... statements) to implement data replication. Assume that there is a table named "abnormal_monitoring_data" in the abnormal database and a table named "normal_monitoring_data" in the normal database, and the structures of the two tables are similar (both contain fields such as camera number, acquisition time, face data, and abnormal type). You can use SQL statements to replicate them.

[0053] After abnormal data is detected, an automated notification system is developed (such as a text message push platform, message push function in the community management app, etc.) to send abnormal situation reminders and related processing suggestions to property personnel. At the same time, the community's geographic information system is integrated to provide property personnel with map navigation functions to quickly locate abnormal areas.

[0054] Of course, the present invention may have many other implementations. Based on this implementation, other implementations obtained by ordinary technicians in this field without any creative work are all within the scope of protection of the present invention.

Claims

1. A method for securely storing data in a smart community, characterized by: The following steps are involved: Step 1: Classify the database into long-term database, normal database and abnormal database; Step 2: Data collection: collect information about people in the community through manual and automatic collection, and save the data in a long-term database; Step 3: Tag surveillance cameras. Label the surveillance cameras in the community according to different situations. Step 4: Comparison of monitoring records. After the monitoring data is collected, it is entered into the normal database, and a comparative analysis is performed between the normal database and the long-term database. Step 5: Sorting out abnormal data. The abnormal data generated in the comparison between the normal database and the long-term database are classified as abnormal data. The abnormal data include abnormal behavior data and abnormal face recognition data. The abnormal behavior data are divided into abnormal behavior trajectory data and abnormal behavior time data. There are two main types of facial recognition anomalies: one is that the surveillance data detected by the surveillance camera cannot match the facial information in the long-term database; the other is that the surveillance camera cannot collect facial information; Abnormal behavior trajectory: surveillance cameras throughout the community continuously capture the behavior trajectory of people. When the behavior trajectory of a person corresponding to the same facial data obviously does not conform to the normal traffic pattern in the community, it will be judged as abnormal behavior trajectory; Behavior time anomaly: Combine time information to set activity time. If surveillance cameras collect data outside the set time, these data are considered abnormal. Step 6: Processing of abnormal data. When abnormal data occurs, the property management will screen and filter the data, and the abnormal data will be stored for a long time.

2. The method for securely storing data in a smart community according to claim 1, characterized in that: The data collected in step 2 includes facial information, personnel data, residential information, family relationships, etc.

3. The method for securely storing data in a smart community according to claim 1, characterized in that: In the step three, the surveillance cameras within the community are labeled according to different situations, the surveillance cameras within the same area are set to the same initial letter, and different digital marks are made according to the route, and the digital marks between adjacent surveillance cameras are arranged according to an arithmetic progression.

4. The method for securely storing data in a smart community according to claim 1, characterized in that: The retention time of the abnormal database is longer than the retention time of the normal database.

5. The method for securely storing data in a smart community according to claim 1, characterized in that: The inability of the surveillance camera in the face recognition anomaly to collect face information is determined by comparing the face recognition anomaly data of adjacent surveillance cameras.

6. Smart community data security storage system, characterized by: The method for securely storing data in a smart community used to execute any one of items 1 to 5 comprises a data acquisition module, a monitoring module, a data storage module and a data processing module.

7. The smart community data security storage system according to claim 6 is characterized by: The data acquisition module is divided into a manual acquisition module and an automatic acquisition module. The manual acquisition module is used to manually record data and then enter it into the system. Electronic entry is made with the help of a database management system client tool or a simple entry interface program, and is connected to a back-end long-term database to store data. The automatic acquisition module collects facial images through hardware equipment with human recognition functions.

8. The smart community data security storage system according to claim 6 is characterized by: The monitoring module is composed of internal monitoring cameras in the community.

9. The smart community data security storage system according to claim 6, characterized in that: The data storage module includes a long-term database module, a normal database module and an abnormal database module.

Images