Park safety early warning response management method and system based on artificial intelligence

Abstract

According to the park safety early warning response management method and system based on artificial intelligence, a complex event is decomposed into a plurality of single simple events in a mode of carrying out feature difference on a safety event, and the problems that in the prior art, a system is difficult to rapidly extract useful information from a large amount of monitoring data, and the safety of a park is influenced are solved. And the intelligent analysis and dynamic adjustment capability is lacked, so that the recognition and response of complex or variable security events are not timely.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic digital data processing, and in particular to an artificial intelligence-based park safety early warning response management method and system. Background Art

[0002] With the acceleration of urbanization and the increase in the complexity of parks, all types of parks require more efficient security management. The advancement of emerging technologies such as artificial intelligence, the Internet of Things, and big data has made automated and intelligent security management possible. However, the existing management methods have limited data processing capabilities, making it difficult for the system to quickly extract useful information from a large amount of monitoring data. They lack intelligent analysis and dynamic adjustment capabilities, resulting in untimely identification and response to complex or variable security incidents. Summary of the invention

[0003] In view of the technical problems existing in the above technical background, the present invention proposes a park safety early warning response management method and system based on artificial intelligence, and the technical solutions adopted are as follows: A park safety early warning response management method based on artificial intelligence, the method comprising: S1: All security events are stored in a large database, which is then broken down into multiple security features. The multiple security features are then classified in a pyramid-like manner according to the degree of impact on security. S2: The sensor acquires security events occurring in the park in real time, and decomposes the security events into multiple security features. According to the pyramid levels where the multiple security features are located, the security level of the event is obtained. The security level indicates the degree to which the event affects security, and the events are prioritized according to the security level. S3: When multiple events occur in the park, the emergency handling system triggers the corresponding emergency warning mechanism according to the priority order of the events; S4: Record the time consumed in each step of resolving the incident, combine it with the security level of the incident, obtain the response time of the incident, and compare the response time with the standard response time. When the response time exceeds the standard response time, analyze the time consumed in each step of resolving the incident in segments, and optimize the time-consuming steps based on the analysis results. Finally, record the events that occurred and upload them to the big database for storage to increase the sample diversity of the big database.

[0004] Preferably, the S1 includes: S11: Import historical security events into a big database, which stores all security events; S12: Extract features of the security incidents, the feature extraction method comprising: for video data, extracting features and behavior patterns of moving objects; for audio data, extracting audio features; for environmental data, extracting sensor degree feature values ​​and change trends; for text data, extracting keywords, and classifying the security features in the form of a pyramid according to the degree of impact on park security, wherein the top layer of the pyramid represents a major security threat, the middle layer of the pyramid represents a medium threat, and the bottom layer of the pyramid represents a minor threat.

[0005] Preferably, the S2 includes: S21: The sensor collects data in the park in real time and uploads the data to the central processing system. After receiving the data, the central processing system identifies the security events in the data. The specific method includes: for video data, identifying events through image processing algorithms; for audio data, identifying events by comparing special audio, the special audio includes but is not limited to the sound of breaking a door, gunshots and screams; for environmental data, comparing various parameters of the current environment with the environmental parameters under normal conditions to determine whether the current environmental parameters are abnormal; for text data, identifying security-related keywords and phrases through natural language processing of text; S22: After identifying the security events in the data, perform feature extraction on the security events in the data, decompose the composite events into a set of security features, make each security feature in the set correspond to the security feature in the pyramid, and superimpose each security feature with different degrees of impact on the security of the park to obtain the security level of the security event, thereby classifying each security event into a security level, and prioritizing all security events according to the security level.

[0006] Preferably, S3 includes: S31: The emergency handling system starts corresponding emergency measures in order of priority from high to low. At the same time, the emergency handling system uploads the information of the security incident and the emergency measures taken to the emergency dispatch system, and dispatches emergency handling personnel to handle the incident through the emergency dispatch system; S32: The emergency response system monitors the development of security incidents in real time during the processing process. When the security incident is resolved, the security incident is recorded and transmitted to the recording system.

[0007] Preferably, the S4 includes: S41: Record the time consumed by each step in the emergency handling system to obtain the total time consumed to resolve the security incident, and combine the security incident level to obtain the response time of the emergency handling system to resolve the security incident; S42: Compare the response time with the standard processing time of security incidents with the same security level. When it is found that the total response time of the security incident exceeds the set standard response time, perform segmented analysis on the time consumption of each step, analyze the problem of excessive time consumption in the specific step and the cause of the problem, and optimize the specific step according to the analysis result; S43: Generate a processing report of the security incident, the report including problem identification, analysis process and optimization suggestions, and upload the processing report to a big data database.

[0008] An artificial intelligence-based park safety early warning response management system, the system comprising: Big database: All security events are stored in the big database, which is then broken down into multiple security features. The multiple security features are then classified in a pyramid-like manner according to the degree of impact on security. Classification system: The sensor acquires security events occurring in the park in real time and decomposes the security events into multiple security features. According to the pyramid levels where the multiple security features are located, the security level of the event is obtained. The security level indicates the degree to which the event affects security. The events are prioritized according to the security level. Emergency response system: When multiple events occur in the park, the emergency response system triggers the corresponding emergency warning mechanism according to the priority order of the events; Optimization system: record the time consumed in each step of resolving the incident, combine the security level of the incident, obtain the response time of the incident, and compare the response time with the standard response time. When the response time exceeds the standard response time, analyze the time consumed in each step of resolving the incident in segments, and optimize the time-consuming steps based on the analysis results. Finally, record the events that occurred and upload them to the big database for storage to increase the sample diversity of the big database.

[0009] Preferably, the large database includes: Data storage system: import historical security events into a big database, which stores all security events; Feature extraction system: extract features of the security incidents, and the feature extraction method includes: for video data, extracting features and behavior patterns of moving objects; for audio data, extracting audio features; for environmental data, extracting sensor degree feature values ​​and change trends; for text data, extracting keywords, and classifying the security features in the form of a pyramid according to the degree of impact on park security, wherein the top layer of the pyramid represents a major security threat, the middle layer of the pyramid represents a medium threat, and the bottom layer of the pyramid represents a minor threat.

[0010] Preferably, the classification system comprises: Identification system: The sensor collects data in the park in real time and uploads the data to the central processing system. After receiving the data, the central processing system identifies the security events in the data. The specific methods include: for video data, identifying events through image processing algorithms; for audio data, identifying events by comparing special audio, which includes but is not limited to the sound of breaking doors, gunshots and screams; for environmental data, comparing various parameters of the current environment with the environmental parameters under normal conditions to determine whether there are abnormalities in the current environmental parameters; for text data, natural language processing of text to identify security-related keywords and phrases; Decomposition system: After identifying the security events in the data, perform feature extraction on the security events in the data, decompose the composite events into a set of security features, make each security feature in the set correspond to the security feature in the pyramid, and superimpose each security feature with different degrees of impact on the security of the park to obtain the security level of the security event. In this way, each security event is divided into a security level, and all security events are prioritized according to the security level.

[0011] Preferably, the emergency handling system comprises: Emergency dispatch system: The emergency handling system starts corresponding emergency measures in order of priority from high to low. At the same time, the emergency handling system uploads the information of security events and the emergency measures taken to the emergency dispatch system, and dispatches emergency handling personnel to handle the events through the emergency dispatch system; Real-time monitoring system: The emergency response system monitors the development of security incidents in real time during the processing process. When the security incident is resolved, the security incident will be recorded and transmitted to the recording system.

[0012] Preferably, the optimization system comprises: Recording system: Record the time consumed in each step of the emergency response system to obtain the total time consumed to resolve the security incident. At the same time, combined with the security incident level, obtain the response time of the emergency response system to resolve the security incident. Analysis system: Compare the response time with the standard processing time of security incidents with the same security level. When it is found that the total response time of a security incident exceeds the set standard response time, the time consumption of each step is analyzed in sections to find out the time consumption problem in the specific step and the cause of the problem. Based on the analysis results, the specific steps are optimized. Reporting system: Generates a security incident processing report, which includes problem identification, analysis process and optimization suggestions, and uploads the processing report to a big database.

[0013] Beneficial effects of the present invention: The present invention realizes efficient management of security incidents by integrating a large database, a classification system, an emergency processing system and an optimization system. The system uses AI to monitor and automatically classify events in real time, and uses a pyramid classification method to prioritize security threats, making resource scheduling more accurate and emergency response faster. In addition, by recording and analyzing the time consumption of each processing step, the system can identify and optimize time-consuming steps and improve overall response efficiency. Finally, the generated processing report provides data support for continuous improvement. In summary, the system has significantly improved the security management level of the park, and can increase the accuracy and efficiency of event processing while reducing response time. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 This is a flow chart of the park safety early warning response management method based on artificial intelligence described in the present invention. DETAILED DESCRIPTION

[0015] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0016] One embodiment of the present invention provides a campus safety early warning response management method based on artificial intelligence, the method comprising: S1: All security events are stored in a large database, which is then broken down into multiple security features. The multiple security features are then classified in a pyramid-like manner according to the degree of impact on security. S2: The sensor acquires security events occurring in the park in real time, and decomposes the security events into multiple security features. According to the pyramid levels where the multiple security features are located, the security level of the event is obtained. The security level indicates the degree to which the event affects security, and the events are prioritized according to the security level. S3: When multiple events occur in the park, the emergency handling system triggers the corresponding emergency warning mechanism according to the priority order of the events; S4: Record the time consumed in each step of resolving the incident, combine it with the security level of the incident, obtain the response time of the incident, and compare the response time with the standard response time. When the response time exceeds the standard response time, analyze the time consumed in each step of resolving the incident in segments, and optimize the time-consuming steps based on the analysis results. Finally, record the events that occurred and upload them to the big database for storage to increase the sample diversity of the big database.

[0017] The working principle and effect of the above technical solution are as follows: First, the system stores all historical security events through a large database, analyzes these events, and extracts several representative security features. These features are classified hierarchically using a pyramid model according to the severity of their impact on security, with the top layer of the pyramid representing the highest severity and the bottom layer representing the lowest. The sensor system installed in the park monitors field data in real time and automatically detects and identifies new security events. The event is decomposed into multiple established security features, and each feature is assigned a corresponding security level according to its position in the pyramid. In addition, the algorithm compares all ongoing events, determines its priority based on the security level of the event, and provides input for emergency handling. In the event of multiple events occurring at the same time, the emergency handling system responds in order of priority, giving priority to high-level serious events. The system will act as a response coordination center, automatically trigger the corresponding early warning mechanism, and dispatch appropriate emergency measures and resources to minimize the impact of the event. The system records the time taken for each step in all response processes in detail and compares it with the preset standard response time; if there is a timeout, the response time taken for each stage will be analyzed to identify efficiency bottlenecks. Based on the analysis, the system will provide optimization suggestions for time-consuming steps to ensure more efficient future responses. At the same time, the solutions and relevant data of all events will be recorded and uploaded to the big database to enrich its data samples and further enhance the system's learning ability and response effect. This method improves the speed and accuracy of event identification and response through real-time monitoring, intelligent analysis and data-driven optimization process. The use of big databases and pyramid-style security feature classification effectively ensures the rational allocation and priority processing of resources, thereby reducing delays in responding to emergency events. In addition, it has an automatic recording and feedback mechanism, which can continuously optimize the response steps, enhance the system's learning ability and processing efficiency, and ultimately improve the overall security management level and reduce potential risks.

[0018] In one embodiment of the present invention, the S1 includes: S11: Import historical security events into a big database, which stores all security events; S12: Extract features of the security incidents, the feature extraction method comprising: for video data, extracting features and behavior patterns of moving objects; for audio data, extracting audio features; for environmental data, extracting sensor degree feature values ​​and change trends; for text data, extracting keywords, and classifying the security features in the form of a pyramid according to the degree of impact on park security, wherein the top layer of the pyramid represents a major security threat, the middle layer of the pyramid represents a medium threat, and the bottom layer of the pyramid represents a minor threat.

[0019] The working principle and effect of the above technical solution are as follows: First, historical security events are systematically imported into a large database. By extracting features from different types of raw data, this includes: for video data: extracting features of moving objects through computer vision technology, identifying their motion trajectories and behavior patterns; for audio data: applying audio analysis technology to identify special audio features, such as sharp or explosive sounds; for environmental data: extracting changing features through analysis of sensor data, such as significant changes in temperature, humidity, and smoke concentration; for text data: using natural language processing (NLP) to identify keywords and phrases, the text features come from reports, complaints or social media, helping to reveal potential risks. The extracted features are classified into a pyramid model according to the degree of impact on park safety: the top layer (major threats) includes event features that can directly affect life safety or cause major property losses. The middle layer (medium threats) includes small-scale events that cause operational interruptions or local risks. The bottom layer (minor threats) covers anomalies that have little impact on daily operations and only require routine monitoring. This method improves the system's ability to predict and respond to future events through deep storage and feature extraction of historical security event data. Through comprehensive analysis of video, audio, environment and text data, the system realizes multi-angle monitoring of security threats. At the same time, based on the hierarchical classification of the pyramid model, the system can quickly identify and sort threats of different degrees, which not only optimizes the efficiency of resource allocation and emergency response, but also improves the accuracy and reliability of overall security management.

[0020] In one embodiment of the present invention, S2 includes: S21: The sensor collects data in the park in real time and uploads the data to the central processing system. After receiving the data, the central processing system identifies the security events in the data. The specific method includes: for video data, identifying events through image processing algorithms; for audio data, identifying events by comparing special audio, the special audio includes but is not limited to the sound of breaking a door, gunshots and screams; for environmental data, comparing various parameters of the current environment with the environmental parameters under normal conditions to determine whether the current environmental parameters are abnormal; for text data, identifying security-related keywords and phrases through natural language processing of text; S22: After identifying the security events in the data, perform feature extraction on the security events in the data, decompose the composite events into a set of security features, make each security feature in the set correspond to the security feature in the pyramid, and superimpose each security feature with different degrees of impact on the security of the park to obtain the security level of the security event, thereby classifying each security event into a security level, and prioritizing all security events according to the security level.

[0021] And, the security level is obtained by the following formula: ;

[0022] Among them, P represents the safety level, It represents the population density per unit area. represents the speed of the moving object, represents a special sound feature, and the special sound feature is normalized and quantized, Indicates the volume change per unit time. Indicates the change in gas concentration per unit time. It represents the temperature change per unit time. represents the humidity change per unit time, T represents the keyword feature score, express and The weight at the i-th level of the pyramid, i can be 1, 2, 3, and =0.2, Corresponding to the bottom of the pyramid, =0.3, corresponding to the middle layer of the pyramid, =0.5, corresponding to the top layer of the pyramid, , as well as Similarly; And, the keyword feature score T is obtained by the following formula: ; in, Indicates the frequency of keyword i appearing. Indicates the weight of keyword i. The greater the impact of the keyword on security, The bigger.

[0023] The working principle and effect of the above technical solution are as follows: various sensors, including cameras, microphones, environmental monitors, etc., are deployed in the park, and the data streams collected by the sensors in real time are uploaded to the central processing system, which is responsible for the centralized storage and processing of data. The video stream is analyzed by the image processing algorithm system to identify suspicious activities or behavior patterns. Through the sound pattern recognition technology, the audio features appearing in the park are compared to identify abnormal events, such as breaking doors, gunshots, screams, etc., to indicate possible security threats. Real-time environmental data parameters are compared with normal values ​​to detect abnormal changes that trigger security incidents. Natural language processing technology is applied to analyze text input and identify key security words and phrases, such as the content of alarms, complaints or social media posts, to reveal potential security risks. Once a security incident is identified, the system will further analyze the incident and extract specific security features from different data types. The extracted security features are matched with the existing pyramid classification system, and the contribution of each feature to the security risk is determined through the feature impact degree model. According to the addition and combination of each security feature, the comprehensive security level of the current event is calculated. The security level of the incident is sorted to determine the priority of the response, thereby optimizing the allocation and scheduling of emergency resources, so that more serious incidents can be handled in a timely manner. This method improves the detection speed and response accuracy of park security incidents, and ensures more efficient threat identification and processing capabilities through real-time data collection and intelligent analysis; at the same time, it avoids the problem of emergency measures delayed due to data lag or inaccurate identification, and realizes the reasonable allocation of resources through rapid priority sorting of incidents.

[0024] In this formula, square roots are used to handle crowd density and aggregation. This can smooth the impact of extreme values ​​and prevent some abnormally high crowd densities from directly leading to disproportionately high scores, thereby stabilizing model evaluation. Logarithmic functions are used for the speed and direction of moving objects. The impact of abnormally high speeds can be handled, so that its growth tends to be stable at high values, avoiding excessive impact caused by a single abnormal speed. The changes in gas concentration and temperature and humidity are combined to solve the problem that the impact of gas depends on ambient temperature or humidity, thereby balancing the overall risk value. The keyword feature score is directly added to the final score, which simply reflects the contribution of important keywords detected in the text. This formula quantitatively analyzes security features, improves the ability to identify multi-dimensional risks, and uses nonlinear processing technology to reduce the risk of false alarms caused by instantaneous extreme values. By comprehensively evaluating the interactive effects between video, audio, environment, and text data, the sensitivity to complex security situations is improved, ensuring that more accurate security level assessments can be provided when different features show abnormalities.

[0025] In one embodiment of the present invention, S3 includes: S31: The emergency handling system starts corresponding emergency measures in order of priority from high to low. At the same time, the emergency handling system uploads the information of the security incident and the emergency measures taken to the emergency dispatch system, and dispatches emergency handling personnel to handle the incident through the emergency dispatch system; S32: The emergency response system monitors the development of security incidents in real time during the processing process. When the security incident is resolved, the security incident is recorded and transmitted to the recording system.

[0026] The working principle and effect of the above technical solution are as follows: after identifying and evaluating security incidents, the incidents have been sorted from high to low according to the security level. The system first handles high-priority security incidents to maximize resource efficiency and minimize the impact of the incident. For each detected high-priority incident, the emergency handling system initiates specific emergency measures according to the preset response strategy. The detailed information of the security incident, as well as the emergency measures taken, are uploaded to the emergency dispatch system. This system is responsible for connecting the event requirements with the available emergency handling personnel to ensure that the relevant personnel are in place quickly and further implement the detailed response plan. The emergency handling system continuously monitors the dynamic development of each security incident. Once the incident is resolved, the system will fully record the entire process information from identification, response to resolution, including the steps taken, timeline, resource use and resolution results. This method improves the response efficiency and effectiveness of security incidents, and ensures that resources can be connected and utilized in a timely manner through the rapid coordination of the emergency dispatch system; at the same time, it avoids improper emergency handling due to information asymmetry or slow response, and enhances the transparency of the incident resolution process and the integrity of data feedback through real-time dynamic monitoring and recording mechanisms.

[0027] In one embodiment of the present invention, the S4 includes: S41: Record the time consumed by each step in the emergency handling system to obtain the total time consumed to resolve the security incident, and combine the security incident level to obtain the response time of the emergency handling system to resolve the security incident; S42: Compare the response time with the standard processing time of security incidents with the same security level. When it is found that the total response time of the security incident exceeds the set standard response time, perform segmented analysis on the time consumption of each step, analyze the problem of excessive time consumption in the specific step and the cause of the problem, and optimize the specific step according to the analysis result; S43: Generate a processing report of the security incident, the report including problem identification, analysis process and optimization suggestions, and upload the processing report to a big data database.

[0028] The working principle and effect of the above technical solution are as follows: During the event handling process, the emergency handling system records the time consumption of each processing step in detail. This includes all stages of identification, response and resolution. By accumulating the time spent in each step, the total time required to resolve a specific security incident is calculated. Combining the response time of an event with its security level can provide a more comprehensive understanding of the speed of event processing at different risk levels. Compare the calculated total response time with the pre-set standard processing time (the standard is set based on historical data). If it is found that the processing time exceeds the expected standard, the time consumption of each step should be systematically analyzed to find out the cause of the delay in the specific stage. After identifying the problem, the system can implement targeted adjustments and optimizations. This method improves the efficiency and transparency of security incident handling, ensures the optimal configuration of resources and processes by recording and analyzing the time consumption of each step in detail; at the same time, it avoids response delays caused by process bottlenecks or improper resource allocation, and promotes the self-improvement and knowledge accumulation of the system by generating detailed processing reports and continuous optimization suggestions.

[0029] One embodiment of the present invention is a park safety early warning response management system based on artificial intelligence, the system comprising: Big database: All security events are stored in the big database, which is then broken down into multiple security features. The multiple security features are then classified in a pyramid-like manner according to the degree of impact on security. Classification system: The sensor acquires security events occurring in the park in real time and decomposes the security events into multiple security features. According to the pyramid levels where the multiple security features are located, the security level of the event is obtained. The security level indicates the degree to which the event affects security. The events are prioritized according to the security level. Emergency response system: When multiple events occur in the park, the emergency response system triggers the corresponding emergency warning mechanism according to the priority order of the events; Optimization system: record the time consumed in each step of resolving the incident, combine the security level of the incident, obtain the response time of the incident, and compare the response time with the standard response time. When the response time exceeds the standard response time, analyze the time consumed in each step of resolving the incident in segments, and optimize the time-consuming steps based on the analysis results. Finally, record the events that occurred and upload them to the big database for storage to increase the sample diversity of the big database.

[0030] The working principle and effect of the above technical solution are as follows: First, the system stores all historical security events through a large database, analyzes these events, and extracts several representative security features. These features are classified hierarchically using a pyramid model according to the severity of their impact on security, with the top layer of the pyramid representing the highest severity and the bottom layer representing the lowest. The sensor system installed in the park monitors field data in real time and automatically detects and identifies new security events. The event is decomposed into multiple established security features, and each feature is assigned a corresponding security level according to its position in the pyramid. In addition, the algorithm compares all ongoing events, determines its priority based on the security level of the event, and provides input for emergency handling. In the event of multiple events occurring at the same time, the emergency handling system responds in order of priority, giving priority to high-level serious events. The system will act as a response coordination center, automatically trigger the corresponding early warning mechanism, and dispatch appropriate emergency measures and resources to minimize the impact of the event. The system records the time taken for each step in all response processes in detail and compares it with the preset standard response time; if there is a timeout, the response time taken for each stage will be analyzed to identify efficiency bottlenecks. Based on the analysis, the system will provide optimization suggestions for time-consuming steps to ensure more efficient future responses. At the same time, the solutions and relevant data of all events will be recorded and uploaded to the big database to enrich its data samples and further enhance the system's learning ability and response effect. This method improves the speed and accuracy of event identification and response through real-time monitoring, intelligent analysis and data-driven optimization process. The use of big databases and pyramid-style security feature classification effectively ensures the rational allocation and priority processing of resources, thereby reducing delays in responding to emergency events. In addition, it has an automatic recording and feedback mechanism, which can continuously optimize the response steps, enhance the system's learning ability and processing efficiency, and ultimately improve the overall security management level and reduce potential risks.

[0031] In one embodiment of the present invention, the large database includes: Data storage system: import historical security events into a big database, which stores all security events; Feature extraction system: extract features of the security incidents, and the feature extraction method includes: for video data, extracting features and behavior patterns of moving objects; for audio data, extracting audio features; for environmental data, extracting sensor degree feature values ​​and change trends; for text data, extracting keywords, and classifying the security features in the form of a pyramid according to the degree of impact on park security, wherein the top layer of the pyramid represents a major security threat, the middle layer of the pyramid represents a medium threat, and the bottom layer of the pyramid represents a minor threat.

[0032] The working principle and effect of the above technical solution are as follows: First, historical security events are systematically imported into a large database. By extracting features from different types of raw data, this includes: for video data: extracting features of moving objects through computer vision technology, identifying their motion trajectories and behavior patterns; for audio data: applying audio analysis technology to identify special audio features, such as sharp or explosive sounds; for environmental data: extracting changing features through analysis of sensor data, such as significant changes in temperature, humidity, and smoke concentration; for text data: using natural language processing (NLP) to identify keywords and phrases, the text features come from reports, complaints or social media, helping to reveal potential risks. The extracted features are classified into a pyramid model according to the degree of impact on park safety: the top layer (major threats) includes event features that can directly affect life safety or cause major property losses. The middle layer (medium threats) includes small-scale events that cause operational interruptions or local risks. The bottom layer (minor threats) covers anomalies that have little impact on daily operations and only require routine monitoring. This method improves the system's ability to predict and respond to future events through deep storage and feature extraction of historical security event data. Through comprehensive analysis of video, audio, environment and text data, the system realizes multi-angle monitoring of security threats. At the same time, based on the hierarchical classification of the pyramid model, the system can quickly identify and sort threats of different degrees, which not only optimizes the efficiency of resource allocation and emergency response, but also improves the accuracy and reliability of overall security management.

[0033] In one embodiment of the present invention, the classification system comprises: Identification system: The sensor collects data in the park in real time and uploads the data to the central processing system. After receiving the data, the central processing system identifies the security events in the data. The specific methods include: for video data, identifying events through image processing algorithms; for audio data, identifying events by comparing special audio, which includes but is not limited to the sound of breaking doors, gunshots and screams; for environmental data, comparing various parameters of the current environment with the environmental parameters under normal conditions to determine whether there are abnormalities in the current environmental parameters; for text data, natural language processing of text to identify security-related keywords and phrases; Decomposition system: After identifying the security events in the data, perform feature extraction on the security events in the data, decompose the composite events into a set of security features, make each security feature in the set correspond to the security feature in the pyramid, and superimpose each security feature with different degrees of impact on the security of the park to obtain the security level of the security event. In this way, each security event is divided into a security level, and all security events are prioritized according to the security level.

[0034] The working principle and effect of the above technical solution are as follows: various sensors, including cameras, microphones, environmental monitors, etc., are deployed in the park, and the data streams collected by the sensors in real time are uploaded to the central processing system, which is responsible for the centralized storage and processing of data. The video stream is analyzed by the image processing algorithm system to identify suspicious activities or behavior patterns. Through the sound pattern recognition technology, the audio features appearing in the park are compared to identify abnormal events, such as breaking doors, gunshots, screams, etc., to indicate possible security threats. Real-time environmental data parameters are compared with normal values ​​to detect abnormal changes that trigger security incidents. Natural language processing technology is applied to analyze text input and identify key security words and phrases, such as the content of alarms, complaints or social media posts, to reveal potential security risks. Once a security incident is identified, the system will further analyze the incident and extract specific security features from different data types. The extracted security features are matched with the existing pyramid classification system, and the contribution of each feature to the security risk is determined through the feature impact degree model. According to the addition and combination of each security feature, the comprehensive security level of the current event is calculated. The security level of the incident is sorted to determine the priority of the response, thereby optimizing the allocation and scheduling of emergency resources, so that more serious incidents can be handled in a timely manner. This method improves the detection speed and response accuracy of park security incidents, and ensures more efficient threat identification and processing capabilities through real-time data collection and intelligent analysis; at the same time, it avoids the problem of emergency measures delayed due to data lag or inaccurate identification, and realizes the reasonable allocation of resources through rapid priority sorting of incidents.

[0035] In one embodiment of the present invention, the emergency handling system comprises: Emergency dispatch system: The emergency handling system starts corresponding emergency measures in order of priority from high to low. At the same time, the emergency handling system uploads the information of security events and the emergency measures taken to the emergency dispatch system, and dispatches emergency handling personnel to handle the events through the emergency dispatch system; Real-time monitoring system: The emergency response system monitors the development of security incidents in real time during the processing process. When the security incident is resolved, the security incident will be recorded and transmitted to the recording system.

[0036] The working principle and effect of the above technical solution are as follows: after identifying and evaluating security incidents, the incidents have been sorted from high to low according to the security level. The system first handles high-priority security incidents to maximize resource efficiency and minimize the impact of the incident. For each detected high-priority incident, the emergency handling system initiates specific emergency measures according to the preset response strategy. The detailed information of the security incident, as well as the emergency measures taken, are uploaded to the emergency dispatch system. This system is responsible for connecting the event requirements with the available emergency handling personnel to ensure that the relevant personnel are in place quickly and further implement the detailed response plan. The emergency handling system continuously monitors the dynamic development of each security incident. Once the incident is resolved, the system will fully record the entire process information from identification, response to resolution, including the steps taken, timeline, resource use and resolution results. This method improves the response efficiency and effectiveness of security incidents, and ensures that resources can be connected and utilized in a timely manner through the rapid coordination of the emergency dispatch system; at the same time, it avoids improper emergency handling due to information asymmetry or slow response, and enhances the transparency of the incident resolution process and the integrity of data feedback through real-time dynamic monitoring and recording mechanisms.

[0037] In one embodiment of the present invention, the optimization system comprises: Recording system: Record the time consumed in each step of the emergency response system to obtain the total time consumed to resolve the security incident. At the same time, combined with the security incident level, obtain the response time of the emergency response system to resolve the security incident. Analysis system: Compare the response time with the standard processing time of security incidents with the same security level. When it is found that the total response time of a security incident exceeds the set standard response time, the time consumption of each step is analyzed in sections to find out the time consumption problem in the specific step and the cause of the problem. Based on the analysis results, the specific steps are optimized. Reporting system: Generates a security incident processing report, which includes problem identification, analysis process and optimization suggestions, and uploads the processing report to a big database.

[0038] The working principle and effect of the above technical solution are as follows: During the event handling process, the emergency handling system records the time consumption of each processing step in detail. This includes all stages of identification, response and resolution. By accumulating the time spent in each step, the total time required to resolve a specific security incident is calculated. Combining the response time of an event with its security level can provide a more comprehensive understanding of the speed of event processing at different risk levels. Compare the calculated total response time with the pre-set standard processing time (the standard is set based on historical data). If it is found that the processing time exceeds the expected standard, the time consumption of each step should be systematically analyzed to find out the cause of the delay in the specific stage. After identifying the problem, the system can implement targeted adjustments and optimizations. This method improves the efficiency and transparency of security incident handling, ensures the optimal configuration of resources and processes by recording and analyzing the time consumption of each step in detail; at the same time, it avoids response delays caused by process bottlenecks or improper resource allocation, and promotes the self-improvement and knowledge accumulation of the system by generating detailed processing reports and continuous optimization suggestions.

[0039] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A park safety early warning response management method based on artificial intelligence, characterized in that: The method comprises: S1: All security events are stored in a large database, which is then broken down into multiple security features. The multiple security features are then classified in a pyramid-like manner according to the degree of impact on security. S2: The sensor acquires security events occurring in the park in real time, and decomposes the security events into multiple security features. According to the pyramid levels where the multiple security features are located, the security level of the event is obtained. The security level indicates the degree to which the event affects security, and the events are prioritized according to the security level. S3: When multiple events occur in the park, the emergency handling system triggers the corresponding emergency warning mechanism according to the priority order of the events; S4: Record the time consumed in each step of resolving the incident, combine it with the security level of the incident, obtain the response time of the incident, and compare the response time with the standard response time. When the response time exceeds the standard response time, analyze the time consumed in each step of resolving the incident in segments, and optimize the time-consuming steps based on the analysis results. Finally, record the events that occurred and upload them to the big database for storage to increase the sample diversity of the big database.

2. According to the artificial intelligence-based park safety early warning response management method of claim 1, it is characterized in that: The S1 includes: S11: Import historical security events into a big database, which stores all security events; S12: Extract features of the security incidents, the feature extraction method comprising: for video data, extracting features and behavior patterns of moving objects; for audio data, extracting audio features; for environmental data, extracting sensor degree feature values ​​and change trends; for text data, extracting keywords, and classifying the security features in the form of a pyramid according to the degree of impact on park security, wherein the top layer of the pyramid represents a major security threat, the middle layer of the pyramid represents a medium threat, and the bottom layer of the pyramid represents a minor threat.

3. According to the artificial intelligence-based park safety early warning response management method of claim 1, it is characterized in that: The S2 includes: S21: The sensor collects data in the park in real time and uploads the data to the central processing system. After receiving the data, the central processing system identifies the security events in the data. The specific method includes: for video data, identifying events through image processing algorithms; for audio data, identifying events by comparing special audio, the special audio includes but is not limited to the sound of breaking a door, gunshots and screams; for environmental data, comparing various parameters of the current environment with the environmental parameters under normal conditions to determine whether the current environmental parameters are abnormal; for text data, identifying security-related keywords and phrases through natural language processing of text; S22: After identifying the security events in the data, perform feature extraction on the security events in the data, decompose the composite events into a set of security features, make each security feature in the set correspond to the security feature in the pyramid, and superimpose each security feature with different degrees of impact on the security of the park to obtain the security level of the security event, thereby classifying each security event into a security level, and prioritizing all security events according to the security level.

4. According to the artificial intelligence-based park safety early warning response management method of claim 1, it is characterized in that: The S3 includes: S31: The emergency handling system starts corresponding emergency measures in order of priority from high to low. At the same time, the emergency handling system uploads the information of the security incident and the emergency measures taken to the emergency dispatch system, and dispatches emergency handling personnel to handle the incident through the emergency dispatch system; S32: The emergency response system monitors the development of security incidents in real time during the processing process. When the security incident is resolved, the security incident is recorded and transmitted to the recording system.

5. According to the artificial intelligence-based park safety early warning response management method of claim 1, it is characterized in that: The S4 includes: S41: Record the time consumed by each step in the emergency handling system to obtain the total time consumed to resolve the security incident, and combine the security incident level to obtain the response time of the emergency handling system to resolve the security incident; S42: Compare the response time with the standard processing time of security incidents with the same security level. When it is found that the total response time of the security incident exceeds the set standard response time, perform segmented analysis on the time consumption of each step, analyze the problem of excessive time consumption in the specific step and the cause of the problem, and optimize the specific step according to the analysis result; S43: Generate a processing report of the security incident, the report including problem identification, analysis process and optimization suggestions, and upload the processing report to a big data database.

6. A park safety early warning response management system based on artificial intelligence, characterized in that: The system comprises: Big database: All security events are stored in the big database, which is then broken down into multiple security features. The multiple security features are then classified in a pyramid-like manner according to the degree of impact on security. Classification system: The sensor acquires security events occurring in the park in real time and decomposes the security events into multiple security features. According to the pyramid levels where the multiple security features are located, the security level of the event is obtained. The security level indicates the degree to which the event affects security. The events are prioritized according to the security level. Emergency response system: When multiple events occur in the park, the emergency response system triggers the corresponding emergency warning mechanism according to the priority order of the events; Optimization system: record the time consumed in each step of resolving the incident, combine the security level of the incident, obtain the response time of the incident, and compare the response time with the standard response time. When the response time exceeds the standard response time, analyze the time consumed in each step of resolving the incident in segments, and optimize the time-consuming steps based on the analysis results. Finally, record the events that occurred and upload them to the big database for storage to increase the sample diversity of the big database.

7. The artificial intelligence-based park safety early warning response management system according to claim 6 is characterized in that: The large database includes: Data storage system: import historical security events into a big database, which stores all security events; Feature extraction system: extract features of the security incidents, and the feature extraction method includes: for video data, extracting features and behavior patterns of moving objects; for audio data, extracting audio features; for environmental data, extracting sensor degree feature values ​​and change trends; for text data, extracting keywords, and classifying the security features in the form of a pyramid according to the degree of impact on park security, wherein the top layer of the pyramid represents a major security threat, the middle layer of the pyramid represents a medium threat, and the bottom layer of the pyramid represents a minor threat.

8. The artificial intelligence-based park safety early warning response management system according to claim 6 is characterized in that: The classification system includes: Identification system: The sensor collects data in the park in real time and uploads the data to the central processing system. After receiving the data, the central processing system identifies the security events in the data. The specific methods include: for video data, identifying events through image processing algorithms; for audio data, identifying events by comparing special audio, which includes but is not limited to the sound of breaking doors, gunshots and screams; for environmental data, comparing various parameters of the current environment with the environmental parameters under normal conditions to determine whether there are abnormalities in the current environmental parameters; for text data, natural language processing of text to identify security-related keywords and phrases; Decomposition system: After identifying the security events in the data, perform feature extraction on the security events in the data, decompose the composite events into a set of security features, make each security feature in the set correspond to the security feature in the pyramid, and superimpose each security feature with different degrees of impact on the security of the park to obtain the security level of the security event. In this way, each security event is divided into a security level, and all security events are prioritized according to the security level.

9. The artificial intelligence-based park safety early warning response management system according to claim 6 is characterized in that: The emergency handling system comprises: Emergency dispatch system: The emergency handling system starts corresponding emergency measures in order of priority from high to low. At the same time, the emergency handling system uploads the information of security events and the emergency measures taken to the emergency dispatch system, and dispatches emergency handling personnel to handle the events through the emergency dispatch system; Real-time monitoring system: The emergency response system monitors the development of security incidents in real time during the processing process. When the security incident is resolved, the security incident will be recorded and transmitted to the recording system.

10. The artificial intelligence-based park safety early warning response management system according to claim 6 is characterized in that: The optimization system comprises: Recording system: Record the time consumed in each step of the emergency response system to obtain the total time consumed to resolve the security incident. At the same time, combined with the security incident level, obtain the response time of the emergency response system to resolve the security incident. Analysis system: Compare the response time with the standard processing time of security incidents with the same security level. When it is found that the total response time of a security incident exceeds the set standard response time, the time consumption of each step is analyzed in sections to find out the time consumption problem in the specific step and the cause of the problem. Based on the analysis results, the specific steps are optimized. Reporting system: Generates a security incident processing report, which includes problem identification, analysis process and optimization suggestions, and uploads the processing report to a big database.

Images