Public safety video image intelligent application system and device and storage medium
By introducing data acquisition, processing, fusion and storage modules into the video surveillance system, the problem that existing systems cannot effectively analyze massive video data is solved, intelligent processing and deep integration of data are realized, and the efficiency of public safety management is improved and Accuracy.
Patent Information
- Application Number
- CN202411546170.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-01
- Publication Date
- 2024-11-29
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The existing video surveillance systems lack intelligent data processing and analysis capabilities, and are difficult to effectively respond to the real-time analysis needs of massive video data, and cannot automatically identify complex security threats.
Provide an intelligent application system for public safety video images, including data acquisition module, data processing module, data fusion module, storage classification module and data storage module. By receiving multi-source data, intelligent preliminary processing is carried out, and the fusion public security information network is associated with The business data in it is classified and stored according to the data characteristics to realize in-depth fusion and correlation analysis of data.
It significantly improves the efficiency of data processing and storage, realizes deep fusion and correlation analysis of multi-source data, can automatically identify potential connections between data, improves retrieval accuracy and collaborative combat capabilities, and enhances the intelligence of public safety management. and efficient.
Smart Images

Figure CN119046498A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the technical field of intelligent applications, and in particular to a public security video image intelligent application system, device and storage medium. Background Art
[0002] With the acceleration of urbanization and the complication of public security, public safety has become a focus of great concern to the government and all sectors of society. As an important means of maintaining public safety, video surveillance systems play an increasingly important role in preventing crimes, combating illegal activities, and responding to emergencies. However, in the face of the growing amount of video data and complex security situations, traditional video surveillance systems can no longer meet the current needs of public safety management.
[0003] At present, many cities have established large-scale video surveillance networks to achieve full coverage of key areas. These systems usually include front-end cameras, transmission networks, storage devices, and management platforms. Video data is transmitted to the backend through a dedicated network and stored in a large-capacity hard disk array after encoding and compression. The management platform provides basic functions such as real-time monitoring, video playback, and alarm linkage, and supports simple retrieval and analysis of video data.
[0004] However, existing video surveillance systems lack intelligent data processing and analysis capabilities, and are unable to effectively respond to the real-time analysis needs of massive video data and automatically identify complex security threats. This situation needs to be further improved. Summary of the invention
[0005] In order to solve the problem that the existing video surveillance system lacks intelligent data processing and analysis capabilities, the present application provides a public security video image intelligent application system, device and storage medium, which adopts the following technical solutions: In a first aspect, the present application provides a public safety video image intelligent application system, comprising: The data acquisition module is used to receive video image data and IoT sensing data from the public security video transmission network to obtain original data; A data processing module, used to determine a preliminary processing strategy according to the type of the original data, and process the original data according to the preliminary processing strategy to obtain processed data; A data fusion module, used to associate and fuse the processed data with existing business data in the public security information network to obtain fused data; A storage classification module, used to determine the storage category of the data according to the characteristics of the fused data, and obtain a classification result; The data storage module is used to store the fused data into an original library, a resource library, a subject library, a business library or a knowledge library based on the classification result.
[0006] By adopting the above technical solution, in large-scale urban security, videos, images and IoT data from different sources are often processed and stored in isolation, making it difficult to achieve effective association and value mining between data. This application first receives multi-source data from the public security video transmission network, then performs intelligent preliminary processing based on the data type, then associates and integrates the processed data with the existing business data in the public security information network, and finally classifies and stores the data based on data characteristics; it not only significantly improves the efficiency of data processing and storage, but also realizes deep integration and association analysis of multi-source data; it can automatically process different types of data, intelligently identify potential connections between data, and achieve efficient retrieval and utilization through classified storage.
[0007] Optionally, a data sharing module is further included, and the data sharing module works in coordination with the data fusion module and the data storage module, and specifically includes: Demand receiving unit, used to receive data demand strategies from superior departments; A data type determination unit, used to determine the data type that needs to be extracted from the data storage module and the data type that needs to be aggregated from the lower-level departments according to the data demand strategy, and obtain the aggregated data type; A data aggregation unit, for extracting relevant data from an original library, a resource library, a subject library, a business library or a knowledge library of the data storage module based on the aggregate data type, and aggregating designated data from subordinate departments to obtain aggregated data; A fusion analysis unit, used to perform fusion analysis on the aggregated data using the data fusion module to obtain a fusion analysis result; A data set generation unit, used to generate a new data set according to the fusion analysis result, and associate and fuse the new data set with the existing business data through the data fusion module; A delivery strategy formulation unit, used to formulate a data delivery strategy based on the new data set; The data distribution unit is used to distribute the new data set to relevant subordinate departments according to the data distribution strategy, and store it in the corresponding database through the storage classification module and the data storage module.
[0008] By adopting the above technical solution, in actual applications, such as cross-regional crime tracking, data sharing between superior and subordinate departments is often delayed and inaccurate, resulting in the inability to coordinate actions in a timely and effective manner. This application first receives the data demand strategy of the superior department, and determines the type of data that needs to be extracted and aggregated based on this. Then, relevant data is aggregated from multi-level databases and subordinate departments, and the data fusion module is used for in-depth analysis to generate new data sets and associate and merge them with existing business data. Finally, an accurate data distribution strategy is formulated to distribute the new data sets to relevant subordinate departments and store them. This not only greatly improves the efficiency and accuracy of data sharing, but also realizes the two-way flow of data, significantly enhances the cross-departmental and cross-regional collaborative combat capabilities, and provides more intelligent and efficient technical support for public safety management.
[0009] Optionally, a cloud-edge collaborative processing module is further included, and the cloud-edge collaborative processing module includes: An edge receiving unit, used to receive the real-time video image data and IoT sensing data from the data acquisition module at the edge node to obtain real-time data; An edge analysis unit, used to perform preliminary analysis on the real-time data according to a preset edge processing strategy to obtain preliminary analysis results; A data transmission unit, used for transmitting the preliminary analysis result to the cloud to obtain the analysis result received by the cloud; A cloud receiving unit, used to receive analysis results from multiple edge nodes in the cloud to obtain multi-node analysis results; A deep analysis unit, used to perform deep analysis and cross-region data fusion based on the multi-node analysis results to obtain a deep fusion result; A strategy updating unit, configured to determine a new edge processing strategy and parameters according to the deep fusion result, obtain a new edge strategy and store it in the data storage module; The policy issuing unit is used to extract the new edge policy from the data storage module and issue it to the corresponding edge node.
[0010] By adopting the above technical solutions, traditional public security video surveillance systems face network bandwidth pressure and response delay problems when processing large-scale real-time data. For example, in urban traffic monitoring, traditional systems often transmit all video data to the central server for processing, resulting in network congestion and analysis delays, and are unable to respond to emergencies in a timely manner. This application first receives and preliminarily analyzes real-time data at the edge node, and transmits the analysis results to the cloud. The cloud then aggregates the results of multiple edge nodes for in-depth analysis and cross-regional data fusion, and updates the edge processing strategy based on the fusion results and sends it to the corresponding nodes. This not only significantly reduces the amount of data transmission and reduces network bandwidth pressure, but also realizes the coordinated optimization of edge intelligence and cloud intelligence, greatly improving the real-time and intelligence level of the system.
[0011] Optionally, it further includes an intelligent retrieval module, the intelligent retrieval module including: A request receiving unit, used to receive a search request from a user and obtain the search request; A database determination unit, configured to determine the type of database to be queried according to the content of the retrieval request and the classification result of the storage classification module, and obtain the queried database type; A parallel search unit, used to perform a search in parallel in the original library, resource library, subject library, business library and knowledge library of the data storage module based on the query database type to obtain a search result; A result fusion unit, used to use the data fusion module to fuse the search results according to a preset fusion rule to obtain a fused search result; A relevance calculation unit, used to calculate the relevance score of each search item based on the fused search result and the associated information provided by the data fusion module to obtain a relevance score; A sorting unit, used to sort the search results according to the relevance scores to obtain a sorting result; The result returning unit is used to return the sorting result to the user.
[0012] By adopting the above technical solution, in the investigation of complex cases, traditional systems often require manual searches in multiple independent databases one by one, which is time-consuming and labor-intensive and easy to miss key information; this application first intelligently determines the query database type based on the user's request, then performs retrieval in multiple databases in parallel, uses the data fusion module to fuse the retrieval results, calculates the relevance score based on the associated information, and finally intelligently sorts the results; not only does it significantly improve the retrieval efficiency, but it also greatly improves the retrieval accuracy through result fusion and relevance calculation. The system can automatically associate multi-source data, discover potential information connections, provide users with more comprehensive and valuable retrieval results, and enhance the ability of public security personnel to quickly obtain and analyze key information.
[0013] Optionally, an intelligent analysis module is also included, and the intelligent analysis module includes: A data receiving unit, used for receiving video image data and obtaining data to be analyzed; An object detection unit, configured to perform object detection on the data to be analyzed using a pre-trained deep learning model to obtain an object detection result; A feature extraction unit, used to extract the features of the target object according to the object detection result to obtain the target feature; A feature comparison unit, used to compare the target feature with known features in the knowledge base to obtain a comparison result; An identity determination unit, configured to determine the identity of the target object based on the comparison result to obtain a target identity; A behavior analysis unit, configured to perform behavior analysis according to the target identity and obtain a behavior analysis result; An early warning judgment unit, used to judge whether it is necessary to trigger an event early warning based on the behavior analysis result, and obtain an early warning judgment result; The alarm generation unit is used to generate an alarm and push relevant information to the designated operator when an early warning is required.
[0014] By adopting the above technical solutions, in the security of large public places, traditional systems often rely on manual observation and find it difficult to detect and respond to potential threats in a timely manner. This application first uses a pre-trained deep learning model for object detection and feature extraction, and then compares it with the known features in the knowledge base to determine the target identity, and then performs behavioral analysis. Finally, based on the analysis results, it determines whether an early warning needs to be triggered; it not only greatly improves the efficiency and accuracy of video analysis, but also realizes the transition from passive monitoring to active early warning. It can automatically identify suspicious persons and behaviors, generate alarms in a timely manner and push them to relevant personnel, greatly enhancing the response capabilities of security personnel.
[0015] Optionally, an adaptive resource scheduling module is also included, and the adaptive resource scheduling module includes: The load monitoring unit is used to monitor the load conditions of each component of the system in real time and obtain load data; A performance calculation unit, used to calculate the performance index of the current system according to the load data; An optimization judgment unit, used to judge whether the current system is in an optimal state based on the performance index, and obtain an optimization judgment result; A resource determination unit, used to determine the resource type that needs to be adjusted when the resource is not in an optimal state, and obtain the adjusted resource type; A resource adjustment unit, configured to dynamically adjust the allocation of computing resources, storage resources or network bandwidth according to the adjustment resource type to obtain an adjustment result; An effect recording unit, used to continue monitoring system performance after adjustment, and record the adjustment effect to obtain adjustment effect data; A strategy optimization unit is used to optimize future resource scheduling strategies based on the adjustment effect data.
[0016] By adopting the above technical solution, during the security of large-scale events, the system load may fluctuate dramatically. The traditional static resource allocation method often leads to resource waste or performance bottlenecks. This application first monitors the load of each component in real time and calculates performance indicators to determine whether it is in the optimal state. When optimization is required, it determines the adjustment resource type, dynamically adjusts resource allocation, and continuously monitors the adjustment effect to optimize future strategies. It significantly improves the resource utilization efficiency of the system, achieves rapid response and adaptive adjustment to load changes, avoids resource waste while ensuring performance, and greatly enhances the scalability and stability of the system.
[0017] In a second aspect, the present application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: Receive video image data and IoT sensing data from the public security video transmission network to obtain original data; Determine a preliminary processing strategy according to the type of the original data to obtain a processing strategy; Processing the original data according to the processing strategy to obtain processed data; Associating and fusing the processed data with existing business data in the public security information network to obtain fused data; Determine the storage category of the data according to the characteristics of the fused data, and obtain a classification result; Based on the classification result, the fused data is stored in an original library, a resource library, a subject library, a business library or a knowledge library respectively.
[0018] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented: Receive video image data and IoT sensing data from the public security video transmission network to obtain original data; Determine a preliminary processing strategy according to the type of the original data to obtain a processing strategy; Processing the original data according to the processing strategy to obtain processed data; Associating and fusing the processed data with existing business data in the public security information network to obtain fused data; Determine the storage category of the data according to the characteristics of the fused data and obtain a classification result; Based on the classification result, the fused data is stored in an original library, a resource library, a subject library, a business library or a knowledge library respectively.
[0019] In summary, the present application includes at least one of the following beneficial technical effects: In large-scale urban security, videos, images and IoT data from different sources are often processed and stored in isolation, making it difficult to achieve effective association and value mining between data. This application first receives multi-source data from the public security video transmission network, then performs intelligent preliminary processing based on the data type, then associates and integrates the processed data with the existing business data in the public security information network, and finally stores the data in a classified manner based on data features; it not only significantly improves the efficiency of data processing and storage, but also achieves deep integration and association analysis of multi-source data; it can automatically process different types of data, intelligently identify potential connections between data, and achieve efficient retrieval and utilization through classified storage; In practical applications, such as cross-regional crime tracking, data sharing between superior and subordinate departments is often delayed and inaccurate, resulting in the inability to coordinate actions in a timely and effective manner. This application first receives the data demand strategy of the superior department, and determines the type of data that needs to be extracted and aggregated based on this. Then, relevant data is aggregated from multi-level databases and subordinate departments, and the data fusion module is used for in-depth analysis to generate new data sets and associate and merge them with existing business data. Finally, an accurate data distribution strategy is formulated to distribute the new data sets to relevant subordinate departments and store them. This not only greatly improves the efficiency and accuracy of data sharing, but also realizes the two-way flow of data, significantly enhances the cross-departmental and cross-regional collaborative combat capabilities, and provides more intelligent and efficient technical support for public safety management. Traditional public security video surveillance systems face network bandwidth pressure and response delay problems when processing large-scale real-time data. For example, in urban traffic monitoring, traditional systems often transmit all video data to the central server for processing, resulting in network congestion and analysis delays, and are unable to respond to emergencies in a timely manner. This application first receives and preliminarily analyzes real-time data at the edge node, and transmits the analysis results to the cloud. The cloud then aggregates the results of multiple edge nodes for in-depth analysis and cross-regional data fusion, and updates the edge processing strategy based on the fusion results and sends it to the corresponding nodes. This not only significantly reduces the amount of data transmission and reduces network bandwidth pressure, but also realizes the coordinated optimization of edge intelligence and cloud intelligence, greatly improving the real-time and intelligence level of the system. BRIEF DESCRIPTION OF THE DRAWINGS
[0020] Figure 1 This is a module diagram of a public safety video image intelligent application system according to an embodiment of the present application; Figure 2 It is a module schematic diagram of a data sharing module in a public security video image intelligent application system according to an embodiment of the present application; Figure 3 It is a module schematic diagram of a cloud-edge collaborative processing module in a public security video image intelligent application system according to an embodiment of the present application; Figure 4It is a module schematic diagram of an intelligent retrieval module in a public safety video image intelligent application system according to an embodiment of the present application; Figure 5 It is a module schematic diagram of an intelligent analysis module in a public safety video image intelligent application system according to an embodiment of the present application; Figure 6 It is a module schematic diagram of an adaptive resource scheduling module in a public safety video image intelligent application system according to an embodiment of the present application; Figure 7 It is a diagram of the internal structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION
[0021] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to be used as limitations to the present application. As used in the specification and appended claims of the present application, the singular expressions "one", "a kind of", "said", "above", "the" and "this" are intended to also include plural expressions, unless there is a clear indication to the contrary in the context. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations comprising one or more listed items.
[0022] In the following, the terms "first" and "second" are used for descriptive purposes only and are not to be understood as suggesting or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features, and in the description of the embodiments of the present application, unless otherwise specified, "plurality" means two or more.
[0023] The embodiments of the present application are further described in detail below in conjunction with the drawings in the specification.
[0024] In the first aspect, the present application provides a public security video image intelligent application system, referring to Figure 1 ,The public security video image intelligent application system includes data ,acquisition module, data processing module, data fusion module, storage classification module and data storage module.
[0025] The data acquisition module is used to receive video image data and IoT perception data from the public security video transmission network to obtain original data.
[0026] Specifically, the data acquisition module is responsible for receiving multi-source data from the public security video transmission network. The public security video transmission network is a network system dedicated to transmitting public safety-related video and image data, including urban surveillance cameras, traffic cameras and other devices. IoT sensing data comes from various IoT devices, such as temperature sensors and pedestrian flow monitors. The data acquisition module receives the data streams transmitted by these devices in real time through standardized interface protocols, and converts them into a format that can be processed by the system to form raw data.
[0027] The data processing module is used to determine a preliminary processing strategy according to the type of the original data, and process the original data according to the preliminary processing strategy to obtain processed data.
[0028] In this embodiment, the data processing module automatically selects an appropriate preliminary processing strategy according to the type of original data received.
[0029] Specifically, the data processing module first classifies the original data into video data, image data, IoT data, etc. For video data, the processing strategies adopted include video compression, key frame extraction, moving target detection, etc. For image data, the strategies adopted include image enhancement, target recognition, feature extraction, etc. For IoT data, data cleaning, outlier detection, data standardization, etc. are performed.
[0030] The data fusion module is used to associate and fuse the processed data with the existing business data in the public security information network to obtain fused data.
[0031] In this embodiment, the data fusion module associates and fuses the processed data with the existing business data in the public security information network. The public security information network is a dedicated network for storing various public security business data, including various data such as population information, vehicle information, case information, etc.
[0032] Specifically, the data fusion module first identifies key elements in the processed data, including facial features, license plate numbers, and geographic location information. Then, it searches for relevant matching information in the existing business data of the public security information network. For example, the license plate number identified in the surveillance video is matched with the vehicle information database, or the identified face is compared with the population information database. In this way, the system can associate isolated video image data with rich background information to form more valuable fused data.
[0033] The storage classification module is used to determine the storage category of the data according to the characteristics of the fused data and obtain the classification result.
[0034] In this embodiment, the storage classification module determines the appropriate storage category of the fused data according to its characteristics.
[0035] Specifically, the storage classification module considers the type, importance, and frequency of use of data to determine the storage category, and screens according to preset conditions. For example, the original unprocessed video is classified into the original library; the processed and integrated high-value data is classified into the resource library; the data related to specific topics (such as traffic safety, public security prevention and control, etc.) is classified into the subject library; the data that directly supports daily public security business is classified into the business library; and the data that has been deeply analyzed and mined is classified into the knowledge base.
[0036] The data storage module is used to store the fused data into the original library, resource library, subject library, business library or knowledge library based on the classification results.
[0037] Among them, the data storage module stores the fused data in different databases based on the classification results of the storage classification module, which can improve the access efficiency and security of the data.
[0038] Specifically, the data storage module uses different storage technologies and strategies to adapt to the characteristics of different types of data. In this embodiment, the original library uses a large-capacity disk array to store massive amounts of original video data; the resource library uses a high-performance solid-state hard drive to store frequently accessed important data; the subject library and business library use a relational database to store structured information; the knowledge base uses a graph database to store complex related information. In this way, the system can strike a balance between storage efficiency, access speed, and data security, and provide the most suitable data support for different application scenarios.
[0039] In one embodiment, referring to Figure 2 ,The system also includes a data sharing module, which works in coordination with the data fusion module and the data storage module, including a demand receiving unit, a data type determining unit, a data aggregation unit, a fusion analysis unit, a data set generating unit, a policy making unit and a data distributing unit.
[0040] The demand receiving unit is used to receive data demand strategies from superior departments.
[0041] In this embodiment, the demand receiving unit is responsible for receiving the data demand strategy from the superior department, including information such as demand description for specific types of data, data usage purpose, timeliness requirements, etc.
[0042] Specifically, the demand receiving unit receives these strategies through a dedicated data sharing platform or secure communication channel. For example, a higher-level department may issue a data demand strategy for the analysis of recent traffic violations in the jurisdiction, requiring the aggregation of traffic violation data in various regions and comprehensive analysis. The demand receiving unit will convert these strategies into a format that the system can recognize, providing clear guidance for subsequent data processing.
[0043] The data type determination unit is used to determine the data type that needs to be extracted from the data storage module and the data type that needs to be aggregated from the lower-level departments according to the data demand strategy, and obtain the aggregated data type.
[0044] Specifically, the data type determination unit first analyzes the keywords and requirements in the demand strategy, and then matches them with the existing data types in the system. For example, for the needs of traffic violation analysis, the system may need to extract traffic monitoring video data, violation record data, etc. from the storage module, and may also need to collect the latest traffic violation statistics from the public security departments at the district and county levels.
[0045] The data aggregation unit is used to extract relevant data from the original library, resource library, subject library, business library or knowledge library of the data storage module based on the aggregation data type, and aggregate designated data from subordinate departments to obtain aggregated data.
[0046] Specifically, the data aggregation unit obtains data from multiple data sources at the same time. For example, it extracts the original traffic monitoring video from the original database, extracts the structured violation records from the business database, and obtains the latest traffic violation statistics from the district and county public security departments through a secure data exchange interface. During the aggregation process, the system will perform necessary data format conversion and quality checks to ensure the consistency and availability of the aggregated data.
[0047] The fusion analysis unit is used to perform fusion analysis on the aggregated data using the data fusion module to obtain a fusion analysis result.
[0048] Specifically, the fusion analysis unit performs data analysis through preset machine learning algorithms or preset rules. For example, it correlates the vehicle behavior data in the traffic surveillance video with the violation record data to identify the high-incidence violation sections and time periods; it compares and analyzes the traffic violation data in different regions to identify regional traffic management problems.
[0049] The data set generation unit is used to generate a new data set according to the fusion analysis results, and associate and fuse the new data set with the existing business data through the data fusion module.
[0050] Specifically, the dataset generation unit creates multiple thematic datasets based on the analysis results. For example, the "Traffic Signal Optimization Suggestions" dataset contains information such as signal light duration adjustment suggestions and new signal light locations based on traffic flow analysis; the "Traffic Accident Risk Warning" dataset combines historical accident data, road condition information, and weather forecasts to provide accident risk level predictions for specific road sections in different time periods. These newly generated datasets will then be associated and integrated with existing business data (such as real-time road condition data, traffic facility distribution data, etc.) to form a more comprehensive and predictive data resource.
[0051] The delivery strategy formulation unit is used to formulate a data delivery strategy based on a new data set.
[0052] Specifically, the distribution strategy formulation unit formulates different distribution strategies for different types of data sets according to the characteristics and uses of the data sets. For example, for the "traffic signal optimization suggestion" data set, a regular update and hierarchical distribution strategy is formulated to distribute the optimization suggestions for different areas to the corresponding traffic management departments; for the "traffic accident risk warning" data set, a real-time push strategy is formulated to ensure that relevant departments can obtain the latest risk warning information in a timely manner. At the same time, the strategy may also include the update cycle and validity period of the data set to ensure that each department uses the latest and most relevant data.
[0053] The data distribution unit is used to distribute the new data set to the relevant subordinate departments according to the data distribution strategy, and store it in the corresponding database through the storage classification module and the data storage module.
[0054] Specifically, the data distribution unit may use a secure data transmission protocol to distribute data to designated subordinate departments. For example, the "traffic accident risk warning" data set will be distributed to traffic management departments in various regions to guide them in formulating targeted control measures. At the same time, the system will classify and store these new data sets, such as storing comprehensive analysis results in the knowledge base and storing specific statistical data in the business database.
[0055] In one embodiment, referring to Figure 3 ,The system also includes a cloud-edge collaborative processing module, which includes an edge receiving unit, an edge analysis unit, a data transmission unit, a cloud receiving unit, a deep analysis unit, a policy updating unit and a policy issuing unit.
[0056] The edge receiving unit is used to receive real-time video image data and IoT perception data from the data acquisition module at the edge node to obtain real-time data.
[0057] For example, at a busy traffic intersection, an edge receiving unit may simultaneously receive video streams from multiple high-definition cameras and data from multiple traffic flow sensors.
[0058] The edge analysis unit is used to perform preliminary analysis on real-time data according to a preset edge processing strategy to obtain preliminary analysis results.
[0059] Specifically, the edge analysis unit uses lightweight machine learning models or rule engines to perform fast analysis. For example, for traffic monitoring data, the edge analysis unit performs tasks such as vehicle counting, speed estimation, and license plate recognition. For abnormal events such as traffic accidents or violations, the edge analysis unit may immediately generate alarm information.
[0060] The data transmission unit is used to transmit the preliminary analysis results to the cloud to obtain the analysis results received by the cloud.
[0061] Specifically, the data transmission unit adopts an incremental transmission strategy, which only transmits changed data or high-value analysis results. For example, for normal traffic flow data, only statistical summaries are transmitted; while for detected traffic accidents or severe congestion, more detailed data and related video clips are transmitted.
[0062] The cloud receiving unit is used to receive analysis results from multiple edge nodes in the cloud to obtain multi-node analysis results.
[0063] The deep analysis unit is used to perform deep analysis and cross-regional data fusion based on multi-node analysis results to obtain deep fusion results.
[0064] Specifically, the deep analysis unit uses advanced machine learning algorithms and big data analysis techniques, such as conducting spatiotemporal analysis of traffic flow data for the entire city to identify the propagation patterns of traffic congestion; or using multi-source data (such as traffic data, weather data, accident data, etc.) to establish a predictive model to predict future traffic conditions.
[0065] The strategy updating unit is used to determine the new edge processing strategy and parameters according to the deep fusion result, obtain the new edge strategy and store it in the data storage module.
[0066] Specifically, the policy update unit may optimize the edge processing strategy based on the results of the deep analysis. For example, if the deep analysis finds that certain types of traffic incidents are more likely to occur in certain time periods, the policy update unit will adjust the data sampling rate or analysis sensitivity of the edge node in these time periods.
[0067] The policy delivery unit is used to extract the new edge policy from the data storage module and deliver it to the corresponding edge node.
[0068] In one embodiment, referring to Figure 4 The system also includes an intelligent retrieval module, which includes a request receiving unit, a database determination unit, a parallel retrieval unit, a result fusion unit, a relevance calculation unit, a sorting unit and a result returning unit.
[0069] The request receiving unit is used to receive a search request from a user and obtain the search request.
[0070] In this embodiment, the request receiving unit receives the user's search request and is the first interface for the user to interact with the system. It is responsible for parsing and preliminarily processing the user's search intention.
[0071] Specifically, the request receiving unit may use natural language processing technology to understand the user's search intent. For example, when the user enters "highway accident-prone areas in the past week", the request receiving unit will identify the keywords "week", "highway", "accident" and "high-incidence areas", and convert this information into a structured query that the system can process.
[0072] The database determination unit is used to determine the type of database to be queried according to the content of the search request and the classification result of the storage classification module, and obtain the queried database type.
[0073] Specifically, the database determination unit uses a rule engine or a machine learning model to determine the most suitable database type. For example, for the above retrieval request, the system determines that it is necessary to query the subject database (containing accident statistics) and the business database (containing road information). If the user is identified as a staff member of the traffic management department, it will also decide to query the knowledge base to obtain relevant accident prevention suggestions.
[0074] The parallel retrieval unit is used to perform retrieval in parallel in the original library, resource library, subject library, business library and knowledge library of the data storage module based on the query database type to obtain the retrieval results.
[0075] Specifically, the parallel retrieval unit uses a distributed computing framework to achieve efficient parallel retrieval, for example, to simultaneously retrieve accident data for the past week in the subject database, to retrieve highway information in the business database, and to retrieve relevant accident analysis reports in the knowledge database.
[0076] The result fusion unit is used to use the data fusion module to fuse the search results according to preset fusion rules to obtain a fused search result.
[0077] Specifically, the result fusion unit uses semantic fusion technology to integrate data from different sources. For example, it associates accident data with road information to generate a comprehensive report containing the accident location, time, type, and corresponding road characteristics.
[0078] The relevance calculation unit is used to calculate the relevance score of each search item based on the fusion search result and the associated information provided by the data fusion module to obtain the relevance score.
[0079] Specifically, the relevance calculation unit considers time relevance (the last week), spatial relevance (highways), topic relevance (accidents), and the importance and reliability of the data. For each search item, a relevance score is calculated comprehensively.
[0080] The sorting unit is used to sort the search results according to the relevance scores to obtain the sorting results.
[0081] Specifically, the sorting unit considers the relevance score and sorts the search results to obtain the sorting results. Furthermore, other factors are combined for intelligent sorting. For example, considering the timeliness of the data, the latest information is ranked first; or considering the user's permission level to ensure that users can only see the information they have access to.
[0082] The result returning unit is used to return the sorting result to the user.
[0083] In one embodiment, referring to Figure 5 The system also includes an intelligent analysis module, which includes a data receiving unit, an object detection unit, a feature extraction unit, a feature comparison unit, an identity determination unit, a behavior analysis unit, a warning judgment unit and an alarm generation unit.
[0084] The data receiving unit is used to receive video image data and obtain data to be analyzed.
[0085] The object detection unit is used to perform object detection on the data to be analyzed using a pre-trained deep learning model to obtain an object detection result.
[0086] In this embodiment, the object detection unit uses a pre-trained deep learning model to perform object detection on the data to be analyzed.
[0087] Specifically, the object detection unit uses a preset deep learning model to detect multiple categories of objects in real time, such as pedestrians, vehicles, objects, etc. For example, in a traffic monitoring scenario, the system may detect all vehicles, pedestrians, and traffic signs on the road at the same time.
[0088] The feature extraction unit is used to extract the features of the target object according to the object detection result to obtain the target feature.
[0089] Specifically, the feature extraction unit uses a variety of feature extraction algorithms based on predefined content. For example, for human targets, facial features, body features, and clothing features are extracted; for vehicle targets, vehicle model, color, and license plate information are extracted. These features may be encoded into high-dimensional vectors for subsequent comparison and analysis.
[0090] The feature comparison unit is used to compare the target feature with the known features in the knowledge base to obtain a comparison result.
[0091] Specifically, the feature comparison unit uses an efficient similarity calculation algorithm, such as cosine similarity or Euclidean distance, to quickly find the most similar matches in a large-scale feature database.
[0092] The identity determination unit is used to determine the identity of the target object based on the comparison result to obtain the target identity.
[0093] The behavior analysis unit is used to perform behavior analysis according to the target identity and obtain behavior analysis results.
[0094] Specifically, the behavior analysis unit uses time series analysis and pattern recognition technology to analyze the target according to predefined content. For pedestrian targets, the system analyzes their movement trajectory, dwell time, and interaction with other objects; for vehicle targets, the system analyzes their driving speed, lane change behavior, and whether they comply with traffic rules.
[0095] The early warning judgment unit is used to judge whether it is necessary to trigger an event early warning based on the behavior analysis result and obtain the early warning judgment result.
[0096] For example, some basic warning rules are set in advance, such as "vehicle speeding" or "pedestrian entering a restricted area"; at the same time, anomaly detection algorithms are used to identify behaviors that deviate significantly from normal patterns.
[0097] The alarm generation unit is used to generate an alarm and push relevant information to the designated operator when an early warning is required.
[0098] In one embodiment, referring to Figure 6 The system also includes an adaptive resource scheduling module, which includes a load monitoring unit, a performance calculation unit, an optimization judgment unit, a resource determination unit, a resource adjustment unit, an effect recording unit and a strategy optimization unit.
[0099] The load monitoring unit is used to monitor the load conditions of each component of the system in real time and obtain load data.
[0100] Specifically, the load monitoring unit may use distributed monitoring technology to collect indicators such as CPU usage, memory usage, disk I / O, and network traffic in real time.
[0101] The performance calculation unit is used to calculate the performance indicators of the current system according to the load data.
[0102] The optimization judgment unit is used to judge whether the current system is in an optimal state based on the performance index and obtain the optimization judgment result.
[0103] In this embodiment, the optimization judgment unit uses a multi-threshold judgment method to set multiple levels of thresholds for different performance indicators, such as "normal", "warning" and "critical". When one or more indicators exceed a specific threshold, the system will determine that optimization is required.
[0104] The resource determination unit is used to determine the resource type that needs to be adjusted when the resource is not in an optimal state, and obtain the adjusted resource type.
[0105] Specifically, when the resource determination unit finds that the system response time is slow, it may analyze whether it is caused by CPU overload, insufficient memory or limited network bandwidth. Based on this analysis, it determines the type of resources that need to be adjusted, such as increasing the number of CPU cores, expanding memory capacity or increasing network bandwidth.
[0106] The resource adjustment unit is used to dynamically adjust the allocation of computing resources, storage resources or network bandwidth according to the adjustment resource type to obtain an adjustment result.
[0107] Specifically, the number of virtual machine instances may be automatically increased or decreased through API calls; in a containerized environment, the resource limits or number of replicas of containers may be adjusted; for storage resources, the size of storage volumes may be dynamically adjusted or data may be migrated, and so on.
[0108] The effect recording unit is used to continue monitoring the system performance after the adjustment, and record the adjustment effect to obtain the adjustment effect data.
[0109] The strategy optimization unit is used to optimize future resource scheduling strategies based on adjustment effect data.
[0110] It should be understood that the size of the serial numbers of the steps in the above embodiments does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.
[0111] In one embodiment, the present application provides an electronic device, which may be a server, and its internal structure diagram may be as follows: Figure 7 As shown. The electronic device includes a processor, a memory and a network interface connected via a system bus. The processor of the electronic device is used to provide computing and control capabilities. The memory of the electronic device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the electronic device is used to store data. The network interface of the electronic device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, the following steps are implemented: Receive video image data and IoT sensing data from the public security video transmission network to obtain original data; According to the type of original data, determine the preliminary processing strategy and obtain the processing strategy; Processing the original data according to the processing strategy to obtain processed data; Associating and fusing the processed data with the existing business data in the public security information network to obtain fused data; According to the characteristics of the fused data, the storage category of the data is determined to obtain the classification result; Based on the classification results, the fused data is stored in the original library, resource library, subject library, business library or knowledge library respectively.
[0112] Those skilled in the art will understand that Figure 7 The structure shown in the figure is merely a block diagram of a partial structure related to the scheme of the present application, and does not constitute a limitation on the electronic device to which the scheme of the present application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have a different arrangement of components.
[0113] In one embodiment, an electronic device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented: Receive video image data and IoT sensing data from the public security video transmission network to obtain original data; According to the type of original data, determine the preliminary processing strategy and obtain the processing strategy; Processing the original data according to the processing strategy to obtain processed data; Associating and fusing the processed data with the existing business data in the public security information network to obtain fused data; According to the characteristics of the fused data, the storage category of the data is determined to obtain the classification result; Based on the classification results, the fused data is stored in the original library, resource library, subject library, business library or knowledge library respectively.
[0114] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the above-mentioned computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).
[0115] The above are all preferred embodiments of the present application, and the protection scope of the present application is not limited thereto. Therefore, any equivalent changes made according to the structure, shape, and principle of the present application should be included in the protection scope of the present application.
Claims
1. A public safety video image intelligent application system, characterized in that: include: The data acquisition module is used to receive video image data and IoT sensing data from the public security video transmission network to obtain original data; A data processing module, used to determine a preliminary processing strategy according to the type of the original data, and process the original data according to the preliminary processing strategy to obtain processed data; A data fusion module, used to associate and fuse the processed data with existing business data in the public security information network to obtain fused data; A storage classification module, used to determine the storage category of the data according to the characteristics of the fused data, and obtain a classification result; The data storage module is used to store the fused data into an original library, a resource library, a subject library, a business library or a knowledge library based on the classification result.
2. The public security video image intelligent application system according to claim 1 is characterized in that: It also includes a data sharing module, which works in coordination with the data fusion module and the data storage module, and specifically includes: Demand receiving unit, used to receive data demand strategies from superior departments; A data type determination unit, used to determine the data type that needs to be extracted from the data storage module and the data type that needs to be aggregated from the lower-level departments according to the data demand strategy, and obtain the aggregated data type; A data aggregation unit, for extracting relevant data from an original library, a resource library, a subject library, a business library or a knowledge library of the data storage module based on the aggregate data type, and aggregating designated data from subordinate departments to obtain aggregated data; A fusion analysis unit, used to perform fusion analysis on the aggregated data using the data fusion module to obtain a fusion analysis result; A data set generation unit, used to generate a new data set according to the fusion analysis result, and associate and fuse the new data set with the existing business data through the data fusion module; A delivery strategy formulation unit, used to formulate a data delivery strategy based on the new data set; The data distribution unit is used to distribute the new data set to relevant subordinate departments according to the data distribution strategy, and store it in the corresponding database through the storage classification module and the data storage module.
3. The public security video image intelligent application system according to claim 1 is characterized in that: It also includes a cloud-edge collaborative processing module, which includes: An edge receiving unit, used to receive the real-time video image data and IoT sensing data from the data acquisition module at the edge node to obtain real-time data; An edge analysis unit, used to perform preliminary analysis on the real-time data according to a preset edge processing strategy to obtain preliminary analysis results; A data transmission unit, used for transmitting the preliminary analysis result to the cloud to obtain the analysis result received by the cloud; A cloud receiving unit, used to receive analysis results from multiple edge nodes in the cloud to obtain multi-node analysis results; A deep analysis unit, used to perform deep analysis and cross-region data fusion based on the multi-node analysis results to obtain a deep fusion result; A strategy updating unit, configured to determine a new edge processing strategy and parameters according to the deep fusion result, obtain a new edge strategy and store it in the data storage module; The policy issuing unit is used to extract the new edge policy from the data storage module and issue it to the corresponding edge node.
4. The public security video image intelligent application system according to claim 1 is characterized in that: It also includes an intelligent retrieval module, which includes: A request receiving unit, used to receive a search request from a user and obtain the search request; A database determination unit, configured to determine the type of database to be queried according to the content of the retrieval request and the classification result of the storage classification module, and obtain the queried database type; A parallel search unit, used to perform a search in parallel in the original library, resource library, subject library, business library and knowledge library of the data storage module based on the query database type to obtain a search result; A result fusion unit, used to use the data fusion module to fuse the search results according to a preset fusion rule to obtain a fused search result; A relevance calculation unit, used to calculate the relevance score of each search item based on the fused search result and the associated information provided by the data fusion module to obtain a relevance score; A sorting unit, used to sort the search results according to the relevance scores to obtain a sorting result; The result returning unit is used to return the sorting result to the user.
5. The public security video image intelligent application system according to claim 1 is characterized in that: It also includes an intelligent analysis module, which includes: A data receiving unit, used for receiving video image data and obtaining data to be analyzed; An object detection unit, configured to perform object detection on the data to be analyzed using a pre-trained deep learning model to obtain an object detection result; A feature extraction unit, used to extract the features of the target object according to the object detection result to obtain the target feature; A feature comparison unit, used to compare the target feature with known features in the knowledge base to obtain a comparison result; An identity determination unit, configured to determine the identity of the target object based on the comparison result to obtain a target identity; A behavior analysis unit, configured to perform behavior analysis according to the target identity and obtain a behavior analysis result; An early warning judgment unit, used to judge whether it is necessary to trigger an event early warning based on the behavior analysis result, and obtain an early warning judgment result; The alarm generation unit is used to generate an alarm and push relevant information to the designated operator when an early warning is required.
6. The public security video image intelligent application system according to claim 1 is characterized in that: It also includes an adaptive resource scheduling module, the adaptive resource scheduling module includes: The load monitoring unit is used to monitor the load conditions of each component of the system in real time and obtain load data; A performance calculation unit, used to calculate the performance index of the current system according to the load data; An optimization judgment unit, used to judge whether the current system is in an optimal state based on the performance index, and obtain an optimization judgment result; A resource determination unit, used to determine the resource type that needs to be adjusted when the resource is not in an optimal state, and obtain the adjusted resource type; A resource adjustment unit, configured to dynamically adjust the allocation of computing resources, storage resources or network bandwidth according to the adjustment resource type to obtain an adjustment result; An effect recording unit, used to continue monitoring system performance after adjustment, and record the adjustment effect to obtain adjustment effect data; A strategy optimization unit is used to optimize future resource scheduling strategies based on the adjustment effect data.
7. An electronic device, characterized in that: The invention comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program: Receive video image data and IoT sensing data from the public security video transmission network to obtain original data; Determine a preliminary processing strategy according to the type of the original data to obtain a processing strategy; Processing the original data according to the processing strategy to obtain processed data; Associating and fusing the processed data with existing business data in the public security information network to obtain fused data; Determine the storage category of the data according to the characteristics of the fused data and obtain a classification result; Based on the classification result, the fused data is stored in an original library, a resource library, a subject library, a business library or a knowledge library respectively.
8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the following steps are implemented: Receive video image data and IoT sensing data from the public security video transmission network to obtain original data; Determine a preliminary processing strategy according to the type of the original data to obtain a processing strategy; Processing the original data according to the processing strategy to obtain processed data; Associating and fusing the processed data with existing business data in the public security information network to obtain fused data; Determine the storage category of the data according to the characteristics of the fused data and obtain a classification result; Based on the classification result, the fused data is stored in an original library, a resource library, a subject library, a business library or a knowledge library respectively.
Citation Information
Patent Citations
Unified information resource management method and system based on cross-level and heterogeneous data aggregation
CN110968629A
Power distribution Internet of Things system capable of processing multi-source data based on edge computing
CN113703969A
Multi-modal information composite perception and fusion architecture and method based on cloud edge collaboration
CN114971574A
Multi-source heterogeneous data fusion and mining method based on security space-time big data
CN117992508A
Case investigation auxiliary system based on multi-source data association analysis
CN118643465A
