Urban operation and governance system based on global intelligent connection algorithm

Abstract

The invention discloses a city operation and treatment system based on a global intelligent connection algorithm, and relates to the technical field of smart cities, and the system comprises a data exchange center which is in communication connection with a data integration and sharing module, an intelligent analysis and decision support module and a safety service and user interface module; the data integration and sharing module collects video data of public place environment information through a public safety camera; the intelligent analysis and decision support module monitors environment information of a public place through a CNN, detects abnormal events, performs risk assessment in combination with a random forest, and analyzes the relation between different events through knowledge graph association; and a security protection layer of the security service and user interface module implements an access control mechanism through an encryption technology and a block chain technology, provides RESTful API for integration of third-party developers, and supports development of mobile applications for citizens and enterprises to use. According to the whole scheme, public and social participation is promoted, and the intelligent level of city management and the public safety guarantee capability are improved.

Description

Technical Field

[0001] The present invention relates to the field of smart city technology, and specifically to a city operation and governance system based on a global intelligent connection algorithm. Background Art

[0002] With the acceleration of urbanization, the challenges faced by cities are becoming increasingly complex, such as traffic congestion, environmental pollution, and public safety, which urgently require innovative solutions. By integrating cutting-edge technologies such as the Internet of Things, big data analysis, and artificial intelligence, the system realizes the efficient collaborative use and intelligent analysis of various video surveillance resources and other sensor data in the city, thereby providing real-time monitoring, prediction and early warning, and automated response capabilities. In addition, the global intelligent connection algorithm has also promoted the transformation of urban service models, from traditional passive management to active prevention and service pre-positioning. Citizens can obtain personalized life information and service recommendations through smartphone applications and enjoy a more convenient urban life experience. Enterprises can also benefit from a more transparent and efficient business environment and reduce operating costs. The government uses this platform to achieve refined governance and improve the scientific nature of decision-making and the level of public services. It is worth noting that this intelligent transformation not only improves the operating efficiency of the city, but also provides strong support for promoting economic growth and harmonious social development. In short, with the advancement of technology and the continuous expansion of application scenarios, the global intelligent connection algorithm will play an increasingly important role in the process of building smart cities, bringing unprecedented development opportunities to cities.

[0003] However, although the system has the ability to efficiently integrate various urban resources in theory, it still faces two core problems in actual operation. The first problem is that the various business departments have not been able to fully coordinate, resulting in the ineffective sharing of the valuable information source of urban video IoT resources. These video surveillance devices spread across every corner of the city should be an important bridge for cross-departmental collaboration and information sharing, but due to institutional barriers and inconsistent technical standards, they have become information islands, seriously affecting the overall effectiveness of urban management. Another major challenge is that the city operation and governance system of the global intelligent connection algorithm is currently unable to fully provide visual intelligence applications. As an important branch of artificial intelligence, visual intelligence technology has shown strong application potential in many fields and can significantly improve the accuracy and efficiency of urban management. However, due to considerations such as technology maturity, data security, and privacy protection, the system is still lagging behind in the development and application of visual intelligence applications and cannot give full play to its unique advantages in urban governance. Therefore, how to break down departmental barriers, promote the deep integration and sharing of urban video IoT resources, and accelerate the research and development and application of visual intelligence technology has become a key issue that needs to be urgently resolved in the current full-domain intelligent algorithm city operation and governance system. Summary of the invention

[0004] In view of the shortcomings of the prior art, the present invention provides an urban operation and governance system based on a global intelligent connection algorithm, which solves the problems in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: an urban operation and governance system based on a global intelligent connection algorithm, comprising a data exchange center, wherein the data exchange center is communicatively connected to a data integration and sharing module, an intelligent analysis and decision support module, and a security service and user interface module; The data integration and sharing module collects video data of public place environmental information through public security cameras; The intelligent analysis and decision support module monitors environmental information in public places through CNN, detects abnormal events, combines random forests for risk assessment, and analyzes the relationship between normal and abnormal events through knowledge graph association analysis; The security service and user interface module implements access control mechanisms through encryption technology and blockchain technology, provides RESTful APIs for third-party developers to integrate, and supports the development of mobile applications for operators to use.

[0006] Furthermore, in the data integration and sharing module, the process of collecting video data of public place environmental information through public security cameras includes: Public security cameras are installed in public places in cities to capture video images in real time. The cameras convert light signals into electrical signals through built-in optical lenses and image sensors, and generate digital video streams. Once the video data is captured, it immediately enters the preprocessing stage, and the preprocessing operations include removing noise, adjusting brightness and contrast, stabilizing the image, and compressing the video. After preprocessing, the video content is subjected to feature extraction, and edge detection, shape analysis, and motion vector analysis are used to identify the behavior of people, the movement trajectory of vehicles, and abnormal activities in the video; For complex tasks, convolutional neural network models are applied to automatically learn and extract high-level features, including: identifying behavioral patterns, understanding scenarios, detecting environmental changes, and identifying and classifying objects. The data after preprocessing and feature extraction is sent to the data exchange center, which is then distributed to the corresponding modules for further analysis and decision support.

[0007] Furthermore, in the intelligent analysis and decision support module, the process of monitoring environmental information of public places and detecting abnormal events by training the convolutional layer of the CNN model includes: The convolutional layer receives preprocessed video frames from the data exchange center as input, and slides through each video frame by applying an internal filter, performing element-by-element multiplication and summing, generating a feature map, automatically learning and extracting edge, texture, shape, human behavior, and vehicle motion trajectory features in the image. During the training process, the weights of the convolutional layer are continuously updated through the back-propagation algorithm to minimize the loss function.

[0008] Furthermore, in the intelligent analysis and decision support module, the process of monitoring environmental information of public places and detecting abnormal events by training the CNN model pooling layer includes: After each convolution operation, nonlinear characteristics are introduced through the activation function, and the pooling layer uses the maximum pooling method to select the maximum value of each local area to retain key features while ignoring unimportant features, expanding the system's focus range and identifying abnormally gathered people and abnormally moving vehicles.

[0009] Furthermore, in the intelligent analysis and decision support module, the process of monitoring environmental information of public places and detecting abnormal events by training the fully connected layer of the CNN model includes: The fully connected layer is located at the end of the network. Each neuron is connected to all neurons in the previous layer. The features extracted by the convolutional layer and the pooling layer are integrated, and finally a judgment is made based on the extracted features as to whether an abnormal event has occurred. The abnormal event includes a traffic accident or illegal intrusion, and a corresponding alarm signal is generated.

[0010] Furthermore, in the intelligent analysis and decision support module, the process of monitoring environmental information of public places and detecting abnormal events by training the CNN model output layer includes: The output layer uses a softmax function to convert the internal representation into actionable results, including normal, attention-demanding, and specific abnormal categories; The intelligent analysis and decision support module interacts with the access control mechanism and RESTful API interface in the security service and user interface module to achieve information transmission and timely sharing.

[0011] Furthermore, in the intelligent analysis and decision support module, the process of risk assessment combined with random forest includes: Constructing a random forest model for risk assessment, wherein the risk assessment process can be divided into a data preparation and feature engineering processing stage, a random forest model construction stage, and a risk score calculation stage; In the data preparation and feature engineering processing stage, the data exchange center passes the pre-processed real-time video data to the intelligent analysis module. The feature engineering component in the intelligent analysis module extracts key features from abnormal events and converts them into a format suitable for random forest model input. The key features include event type, occurrence time, geographic location, objects involved, and duration and scale of the event. By analyzing the results of similar events in the past, predicting the development path of the current event and its potential impact, and integrating the impact of external factors on the event, multi-dimensional information is provided for risk assessment; In the stage of building the random forest model, sub-datasets are created from the original video dataset through repeated sampling. A decision tree is trained independently for each sub-dataset so that each tree looks at the data from a different perspective. In the process of building the decision tree, only some randomly selected features are considered as candidate split features each time a node is split. By combining multiple decision trees, the random forest uses the majority voting method to integrate the prediction results of each tree and provide risk assessment. In the risk scoring stage, the system generates a risk score for each abnormal event based on comprehensive judgments of multiple decision trees. The factors include the severity of the event itself, the safety sensitivity of the location where the event occurred, and the time urgency. Different warning levels are set according to the risk scores, including low risk, medium risk, and high risk, and the corresponding preventive measures are matched.

[0012] Furthermore, in the intelligent analysis and decision support module, the process of analyzing the relationship between different events through the knowledge graph association includes: The system automatically extracts pedestrians, vehicles and building entities from video data, assigns unique identifiers, predefines relationship types and automatically establishes connections between entities, performs cross-data source integration and entity alignment, and constructs a knowledge graph containing static and dynamic information, where nodes represent entities and edges represent relationships between entities. Maintaining the latest status through incremental update mechanism, using path query and pattern matching techniques to analyze direct and indirect relationships between entities; Community detection algorithms and time series analysis methods are used to analyze group behaviors, event clusters, and examine the dynamic changes of events over time.

[0013] Furthermore, the security service and user interface module includes a security protection layer and a user interface layer, wherein the security protection layer implements the access control mechanism through encryption technology and blockchain technology, and the process includes: The security protection layer uses the AES-256 encryption algorithm for end-to-end encryption, uses transparent data to encrypt static and dynamic information in the knowledge graph, and regularly updates keys to reduce long-term risks; The distributed ledger built by blockchain technology records key operations and abnormal events. Each node saves a complete copy. Any modification requires consensus verification by the majority of nodes. Smart contracts automatically execute predefined rules, including access control and data sharing conditions. Through fine-grained access control mechanisms combined with multi-factor authentication and role-based access control, different user groups can only access data and services related to their responsibilities; Hash verification and digital signatures are used to prevent tampering and all important activities are recorded in logs.

[0014] Furthermore, the security service and user interface module includes a security protection layer and a user interface layer, wherein the user interface layer provides a RESTful API for third-party developers to integrate and supports the development of mobile applications for operators to use, including: The user interface layer follows the REST architectural style by providing a standardized RESTful API, using JSON format and HTTPS encrypted data transmission. Third-party developers use the API to develop mobile applications. Development documentation, sample codes and sandbox environment testing are used to assist developers in comprehensively testing applications, allowing citizens to view public environmental information in real time, receive emergency notifications, submit incident reports and query public service facilities. Enterprises can monitor the safety status of specific areas, manage employee travel arrangements and conduct risk management.

[0015] The present invention provides a city operation and governance system based on a global intelligent connection algorithm. It has the following beneficial effects:

[0016] First, in terms of data integration and sharing, the system can efficiently collect and process video data from public security cameras, achieving comprehensive information coverage and instant updates, and providing strong support for urban security monitoring. Secondly, the intelligent analysis and decision support module uses advanced CNN technology and random forest algorithms to not only improve the monitoring accuracy and response speed of public safety incidents, but also deeply analyze the correlation between events through knowledge graphs, providing a scientific basis and forward-looking prediction for urban management decisions.

[0017] In addition, the security service and user interface modules have built an indestructible data security protection network through the combination of encryption technology and blockchain technology, effectively ensuring the transmission and storage security of sensitive information. At the same time, the open RESTful API interface promotes the compatibility and scalability of the system, provides a broad space for innovation for third-party developers, and promotes the diversified development of smart city applications. This not only enriches the service options for citizens and enterprises, but also greatly improves the convenience and intelligence of urban life. In summary, the system has significantly improved the efficiency and governance capabilities of urban operations through comprehensive data integration, intelligent analysis, security assurance and user services, laying a solid foundation for building a safe, intelligent and livable modern urban environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a block diagram of a city operation and governance system based on a global intelligent connection algorithm in an embodiment of the present application. DETAILED DESCRIPTION

[0019] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0020] like Figure 1 As shown, the present invention provides a technical solution: an urban operation and governance system based on a global intelligent connection algorithm, comprising a data exchange center, wherein the data exchange center is communicatively connected with a data integration and sharing module, an intelligent analysis and decision support module, and a security service and user interface module; The data integration and sharing module collects video data of public place environmental information through public security cameras; The intelligent analysis and decision support module monitors environmental information in public places through CNN, detects abnormal events, combines random forests for risk assessment, and analyzes the relationship between normal and abnormal events through knowledge graph association analysis; The security service and user interface module implements access control mechanisms through encryption technology and blockchain technology, provides RESTful APIs for third-party developers to integrate, and supports the development of mobile applications for operators to use.

[0021] In the data integration and sharing module, the process of collecting video data of public place environment information through public security cameras includes: Public security cameras are installed in public places in cities to capture video images in real time. The cameras convert light signals into electrical signals through built-in optical lenses and image sensors, and generate digital video streams. Once the video data is captured, it immediately enters the preprocessing stage, and the preprocessing operations include removing noise, adjusting brightness and contrast, stabilizing the image, and compressing the video. After preprocessing, the video content is subjected to feature extraction, and edge detection, shape analysis, and motion vector analysis are used to identify the behavior of people, the movement trajectory of vehicles, and abnormal activities in the video; For complex tasks, convolutional neural network models are applied to automatically learn and extract high-level features, including: identifying behavioral patterns, understanding scenarios, detecting environmental changes, and identifying and classifying objects. The data after preprocessing and feature extraction is sent to the data exchange center, which is then distributed to the corresponding modules for further analysis and decision support.

[0022] In the intelligent analysis and decision support module, the process of monitoring environmental information of public places and detecting abnormal events by training the convolutional layer of the CNN model includes: The convolutional layer receives preprocessed video frames from the data exchange center as input, and slides through each video frame by applying an internal filter, performing element-by-element multiplication and summing, generating a feature map, automatically learning and extracting edge, texture, shape, human behavior, and vehicle motion trajectory features in the image. During the training process, the weights of the convolutional layer are continuously updated through the back-propagation algorithm to minimize the loss function.

[0023] In the intelligent analysis and decision support module, the process of monitoring environmental information of public places and detecting abnormal events by training the CNN model pooling layer includes: After each convolution operation, nonlinear characteristics are introduced through the activation function, and the pooling layer uses the maximum pooling method to select the maximum value of each local area to retain key features while ignoring unimportant features, expanding the system's focus range and identifying abnormally gathered people and abnormally moving vehicles.

[0024] In the intelligent analysis and decision support module, the process of monitoring environmental information of public places and detecting abnormal events by training the fully connected layer of the CNN model includes: The fully connected layer is located at the end of the network. Each neuron is connected to all neurons in the previous layer. The features extracted by the convolutional layer and the pooling layer are integrated, and finally a judgment is made based on the extracted features as to whether an abnormal event has occurred. The abnormal event includes a traffic accident or illegal intrusion, and a corresponding alarm signal is generated.

[0025] In the intelligent analysis and decision support module, the process of monitoring environmental information of public places and detecting abnormal events by training the CNN model output layer includes: The output layer uses a softmax function to convert the internal representation into actionable results, including normal, attention-demanding, and specific abnormal categories; The intelligent analysis and decision support module interacts with the access control mechanism and RESTful API interface in the security service and user interface module to achieve information transmission and timely sharing.

[0026] In the intelligent analysis and decision support module, the process of risk assessment combined with random forest includes: Constructing a random forest model for risk assessment, wherein the risk assessment process can be divided into a data preparation and feature engineering processing stage, a random forest model construction stage, and a risk score calculation stage; In the data preparation and feature engineering processing stage, the data exchange center passes the pre-processed real-time video data to the intelligent analysis module. The feature engineering component in the intelligent analysis module extracts key features from abnormal events and converts them into a format suitable for random forest model input. The key features include event type, occurrence time, geographic location, objects involved, and duration and scale of the event. By analyzing the results of similar events in the past, predicting the development path of the current event and its potential impact, and integrating the impact of external factors on the event, multi-dimensional information is provided for risk assessment; In the stage of building the random forest model, sub-datasets are created from the original video dataset through repeated sampling. A decision tree is trained independently for each sub-dataset so that each tree looks at the data from a different perspective. In the process of building the decision tree, only some randomly selected features are considered as candidate split features each time a node is split. By combining multiple decision trees, the random forest uses the majority voting method to integrate the prediction results of each tree and provide risk assessment. In the risk scoring stage, the system generates a risk score for each abnormal event based on comprehensive judgments of multiple decision trees. The factors include the severity of the event itself, the safety sensitivity of the location where the event occurred, and the time urgency. Different warning levels are set according to the risk scores, including low risk, medium risk, and high risk, and the corresponding preventive measures are matched.

[0027] In the intelligent analysis and decision support module, the process of analyzing the relationship between different events through the knowledge graph association includes: The system automatically extracts pedestrians, vehicles and building entities from video data, assigns unique identifiers, predefines relationship types and automatically establishes connections between entities, performs cross-data source integration and entity alignment, and constructs a knowledge graph containing static and dynamic information, where nodes represent entities and edges represent relationships between entities. Maintaining the latest status through incremental update mechanism, using path query and pattern matching techniques to analyze direct and indirect relationships between entities; Community detection algorithms and time series analysis methods are used to analyze group behaviors, event clusters, and examine the dynamic changes of events over time.

[0028] The security service and user interface module includes a security protection layer and a user interface layer, wherein the security protection layer implements the access control mechanism through encryption technology and blockchain technology, and the process includes: The security protection layer uses the AES-256 encryption algorithm for end-to-end encryption, uses transparent data to encrypt static and dynamic information in the knowledge graph, and regularly updates keys to reduce long-term risks; The distributed ledger built by blockchain technology records key operations and abnormal events. Each node saves a complete copy. Any modification requires consensus verification by the majority of nodes. Smart contracts automatically execute predefined rules, including access control and data sharing conditions. Through fine-grained access control mechanisms combined with multi-factor authentication and role-based access control, different user groups can only access data and services related to their responsibilities; Hash verification and digital signatures are used to prevent tampering and all important activities are recorded in logs.

[0029] The security service and user interface module includes a security protection layer and a user interface layer, wherein the user interface layer provides a RESTful API for third-party developers to integrate and supports the development of mobile applications. The process for operators to use includes: The user interface layer follows the REST architectural style by providing a standardized RESTful API, using JSON format and HTTPS encrypted data transmission. Third-party developers use the API to develop mobile applications. Development documentation, sample codes and sandbox environment testing are used to assist developers in comprehensively testing applications, allowing citizens to view public environmental information in real time, receive emergency notifications, submit incident reports and query public service facilities. Enterprises can monitor the safety status of specific areas, manage employee travel arrangements and conduct risk management.

[0030] The system first focuses on data collection and integration. Through the data integration and sharing module, the system uses the public security camera network to capture and collect video data of urban public places in real time. These data are then transmitted to the core data exchange center to provide raw materials for the system's subsequent analysis. Next, the intelligent analysis and decision support module uses convolutional neural network technology to conduct in-depth analysis of the video data received from the data exchange center, monitor the security status of public places in real time, and automatically detect potential threat events. At the same time, the system also combines the random forest algorithm to conduct risk assessment on the detected abnormal events to determine their severity and possible impact range. In addition, the intelligent analysis and decision support module also has the ability of knowledge graph association analysis. By analyzing the connection between different events, it reveals potential security risk trends and provides city managers with more comprehensive and in-depth insights. In terms of security, the security service and user interface module ensures the security and reliability of data transmission and storage processes by implementing encryption technology and blockchain technology. It also provides a RESTful API interface, allowing third-party developers to integrate system functions and develop mobile applications suitable for citizens and enterprises. The application not only facilitates users to view environmental information of public places in real time, but also provides personalized safety tips and warning services. In summary, the use of the system involves data collection, intelligent analysis, security assurance, user interface, etc. Through this series of steps, the system efficiently monitors the city's security situation and provides strong support for urban operation and governance.

[0031] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions. The sentence "includes an element defined by ... does not exclude the existence of other identical elements in the process, method, article or device including the element".

[0032] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A city operation and governance system based on a global intelligent connection algorithm, characterized by: It includes a data exchange center, which is communicatively connected with a data integration and sharing module, an intelligent analysis and decision support module, and a security service and user interface module; The data integration and sharing module collects video data of public place environmental information through public security cameras; The intelligent analysis and decision support module monitors environmental information in public places through CNN, detects abnormal events, combines random forests for risk assessment, and analyzes the relationship between normal and abnormal events through knowledge graph association analysis; The security service and user interface module implements access control mechanisms through encryption technology and blockchain technology, provides RESTful APIs for third-party developers to integrate, and supports the development of mobile applications for operators to use.

2. The city operation and governance system based on the global intelligent connection algorithm according to claim 1 is characterized by: In the data integration and sharing module, the process of collecting video data of public place environment information through public security cameras includes: Public security cameras are installed in public places in cities to capture video images in real time. The cameras convert light signals into electrical signals through built-in optical lenses and image sensors, and generate digital video streams. Once the video data is captured, it immediately enters the preprocessing stage, and the preprocessing operations include removing noise, adjusting brightness and contrast, stabilizing the image, and compressing the video. After preprocessing, the video content is subjected to feature extraction, and edge detection, shape analysis, and motion vector analysis are used to identify the behavior of people, the movement trajectory of vehicles, and abnormal activities in the video; For complex tasks, convolutional neural network models are applied to automatically learn and extract high-level features, including: identifying behavioral patterns, understanding scenarios, detecting environmental changes, and identifying and classifying objects. The data after preprocessing and feature extraction is sent to the data exchange center, which is then distributed to the corresponding modules for further analysis and decision support.

3. The city operation and governance system based on the global intelligent connection algorithm according to claim 2 is characterized by: In the intelligent analysis and decision support module, the process of monitoring environmental information of public places and detecting abnormal events by training the convolutional layer of the CNN model includes: The convolutional layer receives preprocessed video frames from the data exchange center as input, and slides through each video frame by applying an internal filter, performing element-by-element multiplication and summing, generating a feature map, automatically learning and extracting edge, texture, shape, human behavior, and vehicle motion trajectory features in the image. During the training process, the weights of the convolutional layer are continuously updated through the back-propagation algorithm to minimize the loss function.

4. The city operation and governance system based on the global intelligent connection algorithm according to claim 3 is characterized by: In the intelligent analysis and decision support module, the process of monitoring environmental information of public places and detecting abnormal events by training the CNN model pooling layer includes: After each convolution operation, nonlinear characteristics are introduced through the activation function, and the pooling layer uses the maximum pooling method to select the maximum value of each local area to retain key features while ignoring unimportant features, expanding the system's focus range and identifying abnormally gathered people and abnormally moving vehicles.

5. The city operation and governance system based on the global intelligent connection algorithm according to claim 4 is characterized by: In the intelligent analysis and decision support module, the process of monitoring environmental information of public places and detecting abnormal events by training the fully connected layer of the CNN model includes: The fully connected layer is located at the end of the network. Each neuron is connected to all neurons in the previous layer. The features extracted by the convolutional layer and the pooling layer are integrated, and finally a judgment is made based on the extracted features as to whether an abnormal event has occurred. The abnormal event includes a traffic accident or illegal intrusion, and a corresponding alarm signal is generated.

6. The city operation and governance system based on the global intelligent connection algorithm according to claim 5 is characterized by: In the intelligent analysis and decision support module, the process of monitoring environmental information of public places and detecting abnormal events by training the CNN model output layer includes: The output layer uses a softmax function to convert the internal representation into actionable results, including normal, attention-demanding, and specific abnormal categories; The intelligent analysis and decision support module interacts with the access control mechanism and RESTful API interface in the security service and user interface module to achieve information transmission and timely sharing.

7. The city operation and governance system based on the global intelligent connection algorithm according to claim 6 is characterized by: In the intelligent analysis and decision support module, the process of risk assessment combined with random forest includes: Constructing a random forest model for risk assessment, wherein the risk assessment process can be divided into a data preparation and feature engineering processing stage, a random forest model construction stage, and a risk score calculation stage; In the data preparation and feature engineering processing stage, the data exchange center passes the pre-processed real-time video data to the intelligent analysis module. The feature engineering component in the intelligent analysis module extracts key features from abnormal events and converts them into a format suitable for random forest model input. The key features include event type, occurrence time, geographic location, objects involved, and duration and scale of the event. By analyzing the results of similar events in the past, predicting the development path of the current event and its potential impact, and integrating the impact of external factors on the event, multi-dimensional information is provided for risk assessment; In the stage of building the random forest model, sub-datasets are created from the original video dataset through repeated sampling. A decision tree is trained independently for each sub-dataset so that each tree looks at the data from a different perspective. In the process of building the decision tree, only some randomly selected features are considered as candidate split features each time a node is split. By combining multiple decision trees, the random forest uses the majority voting method to integrate the prediction results of each tree and provide risk assessment. In the risk scoring stage, the system generates a risk score for each abnormal event based on comprehensive judgments of multiple decision trees. The factors include the severity of the event itself, the safety sensitivity of the location where the event occurred, and the time urgency. Different warning levels are set according to the risk scores, including low risk, medium risk, and high risk, and the corresponding preventive measures are matched.

8. The city operation and governance system based on the global intelligent connection algorithm according to claim 7 is characterized by: In the intelligent analysis and decision support module, the process of analyzing the relationship between different events through the knowledge graph association includes: The system automatically extracts pedestrians, vehicles and building entities from video data, assigns unique identifiers, predefines relationship types and automatically establishes connections between entities, performs cross-data source integration and entity alignment, and constructs a knowledge graph containing static and dynamic information, where nodes represent entities and edges represent relationships between entities. Maintaining the latest status through incremental update mechanism, using path query and pattern matching techniques to analyze direct and indirect relationships between entities; Community detection algorithms and time series analysis methods are used to analyze group behaviors, event clusters, and examine the dynamic changes of events over time.

9. The city operation and governance system based on the global intelligent connection algorithm according to claim 8 is characterized by: The security service and user interface module includes a security protection layer and a user interface layer, wherein the security protection layer implements the access control mechanism through encryption technology and blockchain technology, and the process includes: The security protection layer uses the AES-256 encryption algorithm for end-to-end encryption, uses transparent data to encrypt static and dynamic information in the knowledge graph, and regularly updates keys to reduce long-term risks; The distributed ledger built by blockchain technology records key operations and abnormal events. Each node saves a complete copy. Any modification requires consensus verification by the majority of nodes. Smart contracts automatically execute predefined rules, including access control and data sharing conditions. Through fine-grained access control mechanisms combined with multi-factor authentication and role-based access control, different user groups can only access data and services related to their responsibilities; Hash verification and digital signatures are used to prevent tampering and all important activities are recorded in logs.

10. The city operation and governance system based on the global intelligent connection algorithm according to claim 9 is characterized by: The security service and user interface module includes a security protection layer and a user interface layer, wherein the user interface layer provides a RESTful API for third-party developers to integrate and supports the development of mobile applications. The process for operators to use includes: The user interface layer provides a standardized RESTful API, using JSON format and HTTPS encrypted data transmission, and third-party developers use the API to develop mobile applications; Use development documentation, sample code, and sandbox environment testing to assist developers in fully testing applications, viewing public environmental information in real time, receiving emergency notifications, submitting incident reports, and querying public service facilities.

Images