Calling system based on public safety field
By integrating monitoring, data collection, analysis and AI emergency response modules, the problems of narrow monitoring range and information islands in traditional public safety systems are solved, and real-time monitoring of emergencies and efficient emergency response are achieved.
Patent Information
- Application Number
- CN202510487502.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-18
- Publication Date
- 2025-09-16
AI Technical Summary
Traditional public safety response systems rely on fixed monitoring stations and manual inspections, with a narrow monitoring scope, low data collection frequency, and poor timeliness. It is difficult to fully and real-timely grasp the public safety situation, information sharing and collaborative work are imperfect, emergency response efficiency is low, and there is a lack of flexibility and accuracy.
It adopts monitoring module, data acquisition module, data analysis and processing module, AI intelligent emergency response module, early warning and alarm module, central control module and communication module, combined with multi-source fusion emergency response system and AI model, to achieve real-time monitoring, comprehensive data collection, analysis and processing, automatic information release, and centralized coordination of emergency actions.
It realizes real-time monitoring of emergencies and in-depth data mining, improves emergency response efficiency, shortens rescue time, reduces accident losses, and improves system operation efficiency and stability.
Smart Images
Figure CN120656304A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of public safety technology, and in particular to a call response system based on the public safety field. Background Art
[0002] Traditional public safety response systems primarily rely on a limited number of fixed monitoring sites and manual inspections. This approach suffers from problems such as a narrow monitoring range, low frequency of data collection, and poor timeliness, making it difficult to comprehensively and real-timely grasp changes in the public safety situation. Secondly, traditional methods often rely on manual experience and simple statistical analysis, making it difficult to extract valuable information and underlying patterns from massive amounts of data, and unable to provide a quick and accurate basis for emergency decision-making. At the same time, imperfect information sharing and collaborative working mechanisms between departments lead to information silos, poor command, and uncoordinated actions during the emergency response process, seriously affecting the efficiency and effectiveness of emergency handling. In terms of emergency notifications, traditional response systems are usually based on fixed processes and models, lacking flexibility and accuracy, and are difficult to adapt to complex and changing public safety incidents. Summary of the Invention
[0003] The purpose of the present invention is to provide a call response system based on the public security field to solve the technical problems existing in the prior art.
[0004] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:
[0005] A public safety call response system includes the following modules: a monitoring module, which uses preset sensors, cameras and monitoring equipment to conduct real-time monitoring of the surrounding environment data, image information, meteorological data, geological activities, and crowd density data of the emergency scene; a data acquisition module, which collects the data collected by the monitoring module and then performs preliminary processing on the collected data; a data analysis and processing module, which analyzes and processes the data after the preliminary processing through multiple sets of algorithms and corresponding models, and combines the multi-source fusion emergency response system to identify potential safety risks and emergencies; an AI intelligent emergency response module, which can quickly analyze the potential safety risks identified by the data analysis and processing module through the preset trained AI model and the real-time updated machine learning algorithm. The system can understand the nature, scale and coverage of all risks and emergencies; exchange information with multiple relevant departments, and automatically release accurate and clear information to the public to guide the public to cooperate with emergency work; the early warning and alarm module receives information from the data analysis and processing module, and judges the level of risk, and then manually releases the alarm through multiple communication channels; the central control module establishes a control center to coordinate the allocation of the situation at the scene of the incident and the currently available resources, unites different departments, and conducts orderly response actions; the communication module establishes a communication network system to transmit and share information from each module in real time; the management module monitors and manages each module, records the data of each work, and regularly updates and optimizes the entire system.
[0006] Furthermore, the preliminary processing in the data acquisition module includes: data cleaning, which is to denoise the data collected by the monitoring module, remove duplicate data, and piece together incomplete data, so as to process the image of the emergency scene in high definition; data integration, which is to integrate the public safety system, fire protection system and medical system, and unify the cleaned data to form a visual view, and store it at the same time; data aggregation, which is to summarize and count the data at the emergency scene, so as to help the final decision maker to take corresponding measures; data encryption, which is to encrypt the data after data cleaning.
[0007] Furthermore, the multi-source fusion emergency response system in the data analysis and processing module includes: natural disaster emergency cases, public safety incident handling cases, and major event security cases; the natural disaster emergency cases are composed of: earthquakes, floods, tsunamis, mountain torrents, and typhoons; the public safety incident handling cases are composed of: attacks, fires, and explosions.
[0008] Furthermore, the multiple groups of algorithms in the data analysis and processing module include: decision tree algorithm, scenario simulation, random forest algorithm, clustering algorithm, prediction algorithm and anomaly detection algorithm; the models include: decision tree model, random forest model, clustering model and deep learning model.
[0009] Furthermore, the pre-trained AI model in the AI intelligent emergency response module includes the following learning steps: S1: data collection and processing, collecting data related to various public safety incidents, the relevant data coming from relevant department reports, news reports, academic research, social media and sensor records; then, cleaning, denoising and labeling the collected relevant data, and converting the data into multiple formats suitable for input; S2: model training, using the processed data to train the transformer model and the integration model, continuously adjusting the model parameters according to different problem properties and data characteristics, and applying stochastic gradient descent and AdaGrad algorithms to the model so that the model can capture basic language structures and patterns; S3: adjustment, after the model training after S2 is completed, it is adjusted again and iterated multiple times using data from the emergency management field, and the model is evaluated after each iteration; S4: prompt word training, using multiple sets of prompt words to train the model so that the model gradually begins to understand and respond to descriptions of different emergency situations, and form corresponding emergency measures and voice broadcasts; S5: application, deploying the model after S4 into actual applications, and regularly updating and training the model.
[0010] Furthermore, the multiple communication channels in the early warning and alarm module include: text messages, voice calls, mobile application push, email, radio and television, social media, public alarms, and public display screens.
[0011] Compared with the prior art, the present invention has the following beneficial effects:
[0012] (1) The present invention, through the provision of a monitoring module, a data acquisition module, and a data analysis and processing module, can monitor the crime scene in real time and comprehensively collect data, thereby deeply mining the potential information and patterns in the transmitted data and ensuring the accuracy and integrity of the data;
[0013] (2) The present invention, through the provision of an AI intelligent emergency response module, can quickly respond to emergencies and generate optimal emergency plans and resource scheduling based on real-time data, thereby shortening rescue time and improving the efficiency of emergency response;
[0014] (3) The present invention combines manual notification and automatic notification through the setting of early warning and alarm modules and central control modules, thereby reducing errors, reducing losses and hazards caused by accidents, and improving the overall operating efficiency and stability of the system through centralized coordination and control by the central control module. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] Figure 1 Flowchart of the present invention. DETAILED DESCRIPTION
[0016] To make the contents of the present invention more clearly understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings of the embodiments of the present invention. Identical parts are denoted by the same reference numerals. It should be noted that the terms "front," "rear," "left," "right," "up," and "down" used in the following description refer to directions in the accompanying drawings, and the terms "inward" and "outward" refer to directions toward or away from the geometric center of a particular component, respectively.
[0017] like Figure 1 As shown, this embodiment provides a call response system based on the public security field, including the following modules:
[0018] The monitoring module uses preset sensors, cameras, and monitoring equipment to monitor the surrounding environmental data, image information, meteorological data, geological activities, and crowd density data of the emergency scene in real time. The sensors monitor temperature, humidity, fog, and gas concentration. It uses high-precision and high-sensitivity smoke detectors and flame detectors to cover the emergency scene. The sensors, cameras, and monitoring equipment all use 5G networks, satellite positioning, and Beidou short messages to transmit data to ensure that data transmission can still be achieved in the event of power outages or network disconnections.
[0019] The data acquisition module collects the data collected by the monitoring module and then performs preliminary processing on the collected data. The preliminary processing in the data acquisition module includes: data cleaning, which performs denoising, deduplication, and piecing together incomplete data on the data collected by the monitoring module, removing invalid data and interference signals, thereby processing the image of the emergency scene in high definition and reducing unnecessary data transmission burden; data integration, which integrates the public safety system, fire protection system, and medical system, and unifies and integrates the cleaned data to form a visual view, and converts the collected data into a unified format and unit for storage; data aggregation, which summarizes and compiles the data at the emergency scene to help the final decision maker make corresponding measures; data encryption, which encrypts the cleaned data;
[0020] The data analysis and processing module analyzes and processes the data after the preliminary processing through multiple sets of algorithms and corresponding models, and combines them with the multi-source fusion emergency response system to identify potential safety risks and emergencies; the multiple sets of algorithms in the data analysis and processing module include: decision tree algorithm, scenario simulation, random forest algorithm, clustering algorithm, prediction algorithm and anomaly detection algorithm. The decision tree algorithm, scenario simulation, random forest algorithm, clustering algorithm, prediction algorithm and anomaly detection algorithm are all existing technologies and will not be described in detail here; the models include: decision tree model, random forest model, clustering model and deep learning model. The deep learning model uses a complex neural network structure, which can be used to analyze a large amount of historical data and real-time data. Hidden patterns and laws can be found in the data, which is applicable to the field of public safety. One or more models work simultaneously, and their respective advantages are used to enhance the overall prediction ability and stability, thereby improving the efficiency and effect of model training. The multi-source fusion emergency response system includes: natural disaster emergency cases, public safety incident handling cases, and major event security cases. The natural disaster emergency cases are composed of earthquakes, floods, tsunamis, mountain torrents, and typhoons; the public safety incident handling cases are composed of attacks, fires, and explosions. The multi-source fusion emergency response system can integrate cases from different fields to provide comprehensive and multi-dimensional public safety situation awareness for the data analysis and processing modules, thereby more accurately predicting existing public safety risks and crises.
[0021] The AI intelligent emergency response module, through pre-trained AI models and real-time updated machine learning algorithms, extracts and analyzes features from data processed by the data analysis and processing module. It can quickly analyze the nature, scale, and coverage of potential safety risks and emergencies identified by the data analysis and processing module, and, combined with the built-in emergency knowledge base, it can interact with multiple relevant departments and automatically release accurate and clear information to the public, guiding the public to cooperate with emergency work, shortening rescue time, and improving the efficiency of emergency response;
[0022] The early warning and alarm module receives information from the data analysis and processing module, determines the risk level, and then manually issues the alarm through multiple communication channels, including text messages, voice calls, mobile application push notifications, email, radio and television, social media, public alarms, and public display screens. The early warning and alarm module is equipped with one-click group voice intelligent calling and text message notification functions for emergency personnel. It also sets the number of unanswered calls and records missed calls, disconnected calls, and connection time, thereby ensuring that the alarm information is properly issued.
[0023] The central control module establishes a control center and performs system initialization, including loading configuration files, establishing communication with each module, issuing instructions to each module based on real-time data and needs, and unifying the dispatch and allocation of the incident scene and currently available resources, connecting different departments, and carrying out orderly response actions. At this time, the system is in an encrypted state to protect against possible network attacks;
[0024] Communication module, establishes a communication network system to transmit and share information of each module in real time;
[0025] The management module monitors and manages each module, records each working data, and regularly updates and optimizes the entire system.
[0026] The pre-trained AI model in the AI intelligent emergency response module includes the following learning steps:
[0027] S1: Data collection and processing: This involves collecting data related to various public safety incidents, including detailed records of historical events, emergency response plans, relevant laws and policies, environmental data, and socioeconomic data. This data is sourced from departmental reports, news reports, academic research, social media, and sensor records. The collected data is then cleaned, denoised, and labeled to extract meaningful features that effectively represent the data's key information and facilitate model learning and prediction. The data is then converted into various formats suitable for input.
[0028] S2: Model training: Use the processed data to train the transformer model and ensemble model. Model parameters are continuously adjusted based on different problem properties and data features. Stochastic gradient descent and AdaGrad algorithms are applied to the model to enable it to capture basic language structures and patterns.
[0029] S3: Adjustment: After the model training is completed after S2, it is adjusted again and iterated multiple times using data from the emergency management field. After each iteration, the model is evaluated to improve its performance;
[0030] S4: Prompt word training: The model is trained using multiple sets of prompt words, such as disaster type (earthquake, tsunami, flash flood, fire, etc.), emergency measures (evacuation, rescue, material deployment, etc.), geographic location, and event severity. This allows the model to gradually understand and respond to descriptions of different emergency situations and develop corresponding emergency measures and voice broadcasts.
[0031] S5: Application: Deploy the model that has passed S4 to actual applications and regularly update and train the model, such as adjusting model parameters, increasing the amount of training data, and using more complex training techniques.
[0032] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention shall be included in the scope of protection of the present invention.
Claims
1. A call response system based on the public security field, characterized by: Includes the following modules: The monitoring module uses preset sensors, cameras, and monitoring equipment to conduct real-time monitoring of the surrounding environmental data, image information, meteorological data, geological activities, and crowd density data at the emergency scene; The data acquisition module collects the data collected by the monitoring module and then performs preliminary processing on the collected data; The data analysis and processing module analyzes and processes the data after the preliminary processing through multiple sets of algorithms and corresponding models, and combines them with the multi-source fusion emergency response system to identify potential safety risks and emergencies; The AI Intelligent Emergency Response Module, through pre-trained AI models and real-time updated machine learning algorithms, can quickly analyze the nature, scale, and coverage of potential security risks and emergencies identified by the data analysis and processing module; exchange information with multiple relevant departments, and automatically release accurate and clear information to the public to guide public cooperation in emergency response work; The early warning and alarm module receives information from the data analysis and processing module, determines the level of risk, and then manually issues an alarm through various communication channels; Central control module: establish a control center to coordinate and allocate the situation at the scene of the incident and currently available resources, connect different departments, and carry out orderly response actions; Communication module, establishes a communication network system to transmit and share information of each module in real time; The management module monitors and manages each module, records each working data, and regularly updates and optimizes the entire system.
2. The public security call response system according to claim 1, characterized in that: The preliminary processing in the data acquisition module includes: Data cleaning: denoising, removing duplicate data, and piecing together incomplete data from the data collected by the monitoring module, thereby processing high-definition images of the emergency scene; Data integration: integrating public safety systems, fire protection systems, and medical systems, and integrating the cleaned data to form a visual view and store it simultaneously; Data aggregation: Summarize and compile statistics on the data at the scene of the emergency to help final decision makers take corresponding measures; Data encryption: encrypt the data after data cleaning.
3. The public security call response system according to claim 1, characterized in that: The multi-source fusion emergency response system in the data analysis and processing module includes: natural disaster emergency cases, public safety incident handling cases, and major event security cases; the natural disaster emergency cases are composed of: earthquakes, floods, tsunamis, mountain torrents, and typhoons; the public safety incident handling cases are composed of: attacks, fires, and explosions.
4. The public security call response system according to claim 3, characterized in that: The multiple groups of algorithms in the data analysis and processing module include: decision tree algorithm, scenario simulation, random forest algorithm, clustering algorithm, prediction algorithm and anomaly detection algorithm; the models include: decision tree model, random forest model, clustering model and deep learning model.
5. The public security call response system according to claim 1, characterized in that: The pre-trained AI model in the AI intelligent emergency response module includes the following learning steps: S1: Data collection and processing, which collects data related to various public safety incidents. The data comes from relevant department reports, news reports, academic research, social media, and sensor records; Then, the collected relevant data is cleaned, denoised and labeled, and the data is converted into various formats suitable for input; S2: Model training: Use the processed data to train the transformer model and ensemble model. Model parameters are continuously adjusted based on different problem properties and data features. Stochastic gradient descent and AdaGrad algorithms are applied to the model to enable it to capture basic language structures and patterns. S3: Adjustment: After the model training is completed in S2, it is adjusted again and iterated multiple times using data from the emergency management field. After each iteration, the model is evaluated. S4: Prompt word training: Use multiple sets of prompt words to train the model, so that the model gradually begins to understand and respond to descriptions of different emergency situations, and formulate corresponding emergency measures and voice broadcasts; S5: Application: Deploy the model obtained in S4 to actual applications and regularly update and train the model.
6. The public security call response system according to claim 1, characterized in that: The multiple communication channels in the warning and alarm module include: text messages, voice calls, mobile application push, email, radio and television, social media, public alarms, and public display screens.