Urban intelligent management service platform

Abstract

The present invention relates to the field of digital management technology, in particular to a city intelligent management service platform, including a data governance module, a resource library, a task creation module, an intelligent scheduling module and an event processing module, wherein the data governance module is used to obtain image data required for city management, and to add additional information to the image data for governance; the task creation module is used to create tasks according to image data and a preset task creation algorithm in an algorithm warehouse; the intelligent scheduling module calls the corresponding intelligent analysis algorithm in the algorithm warehouse according to the task, and outputs the task and the corresponding intelligent analysis algorithm to the event processing module; the event processing module executes the task and generates an event result. In the system architecture of the present invention, algorithms are centrally stored in the algorithm warehouse, the intelligent scheduling module calls the corresponding algorithm according to the task, and sets the priority of task processing, so that the task processing process is more efficient.

Description

Technical Field

[0001] The present invention relates to the field of digital management technology, in particular to a city intelligent management service platform. Background Art At present, more and more entities are applying new-generation information technology to urban relations, collecting the information needed for urban management through the Internet of Things infrastructure, cloud computing infrastructure, geographic space infrastructure, etc., forming a comprehensive and thorough perception and a large range of information interconnection, forming digital urban management.

[0002] The solution provided in the patent "System and Method for Digital Urban Management" (publication number CN116777711A), which has been published, improves the city management department's monitoring, processing and feedback capabilities of urban events through functions such as graphical user interface, abnormal data analysis, event message generation and sending, realizes intelligent urban management, and improves management efficiency compared with manual management.

[0003] However, through research, it was found that the core issue of current intelligent urban governance is not whether the perception coverage is comprehensive and rich, but how to improve the level of intelligent analysis of the perception sources and video sources that have been built in large quantities, and how to integrate the data of various information platforms that have been built to improve the efficiency of intelligent analysis. Summary of the invention In view of the above-mentioned deficiencies in the prior art, the present invention proposes a city intelligent management service platform by making full use of the existing perception and collection data and fully integrating various information systems, so as to make the intelligent city management more in line with the actual management situation and improve the utilization effect of the city intelligent management service platform.

[0004] In order to achieve the above objectives, the following technical solutions are proposed: The city intelligent management service platform includes data governance module, resource library, task creation module, intelligent scheduling module and event processing module. The resource library includes an algorithm warehouse and a data warehouse; The data management module is used to obtain image data required for urban management, add additional information to the image data for management, obtain the managed image data, and output the managed image data to the data warehouse; The task creation module is used to create tasks according to the managed image data and the task creation algorithm preset in the algorithm warehouse, and output the tasks to the intelligent scheduling module; The intelligent scheduling module calls the corresponding intelligent analysis algorithm in the algorithm warehouse according to the task, and outputs the task and the corresponding intelligent analysis algorithm to the event processing module; The event processing module executes the task according to the task and the corresponding intelligent analysis algorithm to generate event results.

[0005] As a preferred solution, the additional information includes point name, device directory, device type, camera IP, national standard code, longitude and latitude, online status, industry label, venue label, venue name, and scene label.

[0006] As a preferred solution, the creation task specifically includes the following steps: Initiate a task request; set the priority of the task using the task creation algorithm according to the image data; and complete the task creation.

[0007] As a preferred solution, the intelligent scheduling module processing flow includes: During the execution of the task by the event processing module, the corresponding intelligent analysis algorithm in the algorithm warehouse is called to the event processing module according to the task; and the computing power resources required for the task are preloaded, and the computing power resource scheduling strategy and the intelligent scheduling iterative optimization mechanism are executed during the loading process to continuously optimize resources. After the created task execution time expires, the corresponding computing power resources are released.

[0008] As a preferred solution, the event processing module performs tasks including the following steps: Generate event results based on image data and preset data processing flow; Filter the generated event results, remove duplicate and invalid events, and obtain valid events; Determine whether the result of the valid event is consistent with the facts. If it is consistent, it is a correct event, and the correct event and its result are input into the correct subject library. Otherwise, it is an incorrect event, and the incorrect event and its result are input into the incorrect subject library; output the correct event and its result.

[0009] As a preferred solution, the data used by the intelligent scheduling module to implement its functions include: algorithm scheduling request data, algorithm scheduling status data, device resource status data, and structured output data.

[0010] As a preferred solution, the intelligent scheduling module is also used to correct the erroneous events and their results in the error subject library to obtain corrected events and corrected results. The corrected events, corrected results, correct events, and results of correct events are used to optimize the intelligent analysis algorithm.

[0011] As a preferred solution, the algorithm warehouse stores algorithm model data and algorithm registration information data, and the registration information includes a description of the algorithm, the type of the algorithm, a label of the algorithm, resource requirements of the algorithm, and priority information of the algorithm.

[0012] As a preferred solution, the algorithm warehouse also includes algorithm training related data, and the algorithm training related data includes data to be labeled, labeling information data, labeling task data, training task data, training resource status data and model output related data.

[0013] Based on the same concept, a city intelligent management service system is also proposed, including at least one processor and a memory communicatively connected to the at least one processor; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can run any of the above-mentioned city intelligent management service platforms.

[0014] Compared with the prior art, the present invention has the following beneficial effects: the present invention provides a city intelligent management service platform and discloses the system architecture in the platform, in which algorithms are centrally stored in an algorithm warehouse, the intelligent scheduling module calls the corresponding algorithm according to the task and sets the priority of task processing, making the task processing process more efficient. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 This is the overall architecture diagram of the platform in Example 1; Figure 2 This is the data management module management flow chart in Example 1; Figure 3 A task creation flow chart of the task creation module in Example 1; Figure 4 This is a scheduling flow chart of the intelligent scheduling module in Example 1; Figure 5 is a processing flow chart of the event processing module in Example 1; Figure 6 This is a schematic diagram of the data composition of the platform in Example 1; Figure 7 is a detailed list of data categories in Example 1; Figure 8 This is the data management flow chart in Example 2; Fig. 9 This is a data processing flow chart in the demand response in Example 2; Fig.10 This is the event data ecology and processing flow chart in Example 2; Fig.11 This is a diagram of the MPP architecture supporting hybrid data storage in Example 2. DETAILED DESCRIPTION

[0016] The present invention is further described in detail below in conjunction with test examples and specific implementation methods. However, this should not be understood as the scope of the above subject matter of the present invention being limited to the following embodiments, and all technologies realized based on the content of the present invention belong to the scope of the present invention.

[0017] Example 1 The city intelligent management service platform includes data governance module, resource library, task creation module, intelligent scheduling module and event processing module. The overall architecture of the platform is as follows: Figure 1 As shown, The resource library includes an algorithm warehouse and a data warehouse; The data management module is used to obtain image data required for urban management, add additional information to the image data for management, obtain the managed image data, and output the managed image data to the data warehouse; The task creation module is used to create tasks according to the image data and the task creation algorithm preset in the algorithm warehouse, and output the tasks to the intelligent scheduling module; The intelligent scheduling module calls the corresponding intelligent analysis algorithm in the algorithm warehouse according to the task, and outputs the task and the corresponding intelligent analysis algorithm to the event processing module; The event processing module executes the task according to the task and the corresponding intelligent analysis algorithm to generate event results.

[0018] Furthermore, the data governance module governance flow chart is as follows: Figure 2 As shown, including: View data governance first needs to import the original video point equipment information list from the video aggregation and sharing platform, and then verify the equipment list information after importing. For points with incomplete basic information, feedback will be sent to the video aggregation and sharing platform for processing; for points with complete basic information, equipment online status governance, equipment catalog governance, point name governance, industry label governance, venue label governance, venue name governance, and scene label governance will be carried out manually. After the point governance is completed, it needs to be manually reviewed to ensure that the data is accurate, and the correct data will be stored in the data resource library. Data that fails the review will be rejected and needs to be governed again until it passes the review. Among them, industry label governance, venue label governance, venue name governance, scene label governance, and governance review are manually handled by governance personnel, and other process steps are automatically completed by the video analysis empowerment platform.

[0019] Further, the task creation flow chart of the task creation module is as follows Figure 3As shown in the figure, it includes: first, the user of the bureau and commission office initiates a task creation request. After the business and resource evaluation of the big data center, the task is named and the time for task execution is set. Then the camera that needs to execute the task is selected, the intelligent algorithm that needs to be analyzed is selected, and the priority of task execution is set. Then the platform arranges the task according to the current computing power resources, evaluates whether the required computing power meets the task operation requirements, and if it can support the task operation, the task creation is completed. If the computing power required for the evaluation does not meet the task operation requirements, the task priority is increased and the task is issued. The task is issued to the intelligent scheduling module for algorithm and computing power scheduling, and the task is successfully created. Among them, the initiation of the task request is established on the platform by the corresponding department through the interface, and the business, resource evaluation, task naming, setting the task execution time, selecting the camera, selecting the algorithm and setting the task priority are completed by the big data center. The big data center is the management department of the resource library, connected to this platform through the management interface, and completes the corresponding control. The video analysis empowerment platform performs tasks such as task arrangement, judgment on whether the computing power requirements are met, and task creation.

[0020] Furthermore, the scheduling flow chart of the intelligent scheduling module is as follows: Figure 4 As shown, it includes: after the task is successfully created, the task is sent to the intelligent scheduling module. First, according to the algorithm associated with the task, the corresponding intelligent analysis algorithm in the algorithm warehouse is called, and then the computing resources required for the task are preloaded. During the loading process, the computing resource scheduling strategy and the intelligent scheduling iterative optimization mechanism will be executed to continuously optimize resources and make full use of computing resources. After the execution time of the created task expires, the task execution will be terminated and the corresponding algorithm and computing resources will be released. The scheduling process of the entire intelligent scheduling module is completed by the video analysis empowerment platform.

[0021] Furthermore, the processing flow chart of the event processing module is as follows Figure 5 As shown, including: First, according to the results returned by the algorithm analysis, a series of rules are formulated, and the results or result sets that meet the rules will generate relevant social governance events. After the event is generated, the event is filtered to eliminate duplicate and invalid events, and then the event pictures are associated with the video. The event is then reviewed in the third-level operation center. Events that are judged to be wrong after manual review will be archived to the error topic library in the data resource library, and events that are judged to be correct will be archived to the corresponding topic library and pushed to the third-level operation center. At the same time, the correct events can also be used for subsequent algorithm training. The bureau and commission office obtains the correct event results by connecting to the third-level operation center, and then handles the event according to its own business situation.

[0022] In addition, each module involved in this platform not only has corresponding data flow but also data storage, so as to realize each data flow. The data structure diagram of this platform is as follows: Figure 6As shown in the figure, the view management data is mainly input into other data of intelligent scheduling for use; the task management data, equipment status data, video and image data of intelligent scheduling data are combined with the algorithm image data, algorithm description and specification information data related to the algorithm warehouse, etc., and finally generate event-related information in the event subject library; all processes generate data related to monitoring and operation. The detailed list of data categories is as follows Figure 7 As shown in the figure, structured data types are managed using relational databases, and unstructured data are managed using file storage, which supports horizontal expansion.

[0023] Example 2 From the perspective of data relationship and processing flow, the processes of each module are described as follows: Data governance processes such as Figure 8 As shown in Figure 2, the data processing flow in demand response is as follows: Fig. 9 As shown in the figure, the event data ecology and processing flow are as follows Fig.10 shown.

[0024] This system involves the storage of structured data such as information on collected data, equipment information, user information, etc., as well as unstructured performance such as feature data after structured processing of video images. In order to adapt to the scenario where multiple types of data coexist, the system uses a distributed database with an MPP architecture, which enables it to adapt to the following different types of data: A mixture of structured and unstructured data; structured data mainly refers to data such as the time, location, type of an event, as well as user data, permission data, log data, task data, etc. of the system operation; unstructured data mainly refers to image data, video data, and feature value data extracted from images and videos.

[0025] A mixture of massive data storage and lightweight data storage. For a large amount of collected data, it involves massive data storage, such as image or video collection records, algorithm event records, system log records, etc.; there are also some lightweight storage with relatively small storage capacity, such as user data, permission data, task data, etc. Storage needs to take into account the needs of two different levels of storage.

[0026] like Fig.11 The MPP architecture supporting hybrid data storage shown in the figure first expands the unstructured data on each traditional structured data node, making it an MPP storage node supporting hybrid data storage and query, and then implements data storage distribution, query routing, business balancing and other functions through the MPP architecture, thereby constructing an MPP storage solution.

[0027] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A city intelligent management service platform, characterized in that: It includes data governance module, resource library, task creation module, intelligent scheduling module and event processing module. The resource library includes an algorithm warehouse and a data warehouse; The data management module is used to obtain image data required for urban management, add additional information to the image data for management, obtain the managed image data, and output the managed image data to the data warehouse; The task creation module is used to create tasks according to the managed image data and the task creation algorithm preset in the algorithm warehouse, and output the tasks to the intelligent scheduling module; The intelligent scheduling module calls the corresponding intelligent analysis algorithm in the algorithm warehouse according to the task, and outputs the task and the corresponding intelligent analysis algorithm to the event processing module; The event processing module executes the task according to the task and the corresponding intelligent analysis algorithm to generate event results.

2. The city intelligent management service platform according to claim 1, characterized in that: The additional information includes point name, device directory, device type, camera IP, national standard code, longitude and latitude, online status, industry label, venue label, venue name, and scene label.

3. The city intelligent management service platform according to claim 1, characterized in that: The creation task specifically includes the following steps: Initiate a task request; set the priority of the task using the task creation algorithm according to the image data; and complete the task creation.

4. The city intelligent management service platform according to claim 1, characterized in that: The intelligent scheduling module processing flow includes: During the execution of the task by the event processing module, the corresponding intelligent analysis algorithm in the algorithm warehouse is called to the event processing module according to the task; and the computing power resources required for the task are preloaded, and the computing power resource scheduling strategy and the intelligent scheduling iterative optimization mechanism are executed during the loading process to continuously optimize resources. After the created task execution time expires, the corresponding computing power resources are released.

5. The city intelligent management service platform according to claim 1, characterized in that: The event processing module performs tasks including the following steps: Generate event results based on image data and preset data processing flow; Filter the generated event results, remove duplicate and invalid events, and obtain valid events; Determine whether the result of the valid event is consistent with the facts. If it is consistent, it is a correct event, and the correct event and its result are input into the correct subject library. Otherwise, it is an incorrect event, and the incorrect event and its result are input into the incorrect subject library; output the correct event and its result.

6. The city intelligent management service platform according to claim 1, characterized in that: The data used by the intelligent scheduling module to implement its functions include: algorithm scheduling request data, algorithm scheduling status data, device resource status data, and structured output data.

7. The city intelligent management service platform according to claim 5, characterized in that: The intelligent scheduling module is also used to correct the error events and their results in the error subject library to obtain corrected events and corrected results. The corrected events, corrected results, correct events, and results of correct events are used to optimize the intelligent analysis algorithm.

8. A city intelligent management service platform as described in claims 1-6, characterized in that: The algorithm warehouse stores algorithm model data and algorithm registration information data. The registration information includes a description of the algorithm, the type of the algorithm, the label of the algorithm, the resource requirements of the algorithm, and the priority information of the algorithm.

9. A city intelligent management service platform as described in claims 1-6, characterized in that: The algorithm warehouse also includes algorithm training related data, which includes data to be labeled, labeling information data, labeling task data, training task data, training resource status data and model output related data.

10. A city intelligent management service system, characterized in that: It includes at least one processor and a memory that is communicatively connected to the at least one processor; the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can run an urban intelligent management service platform as described in any one of claims 1 to 9.

Images