Safe intelligent management method and system for nine small places in city

Abstract

The invention provides a safety intelligent management method and system for nine small places of a city, and relates to the technical field of city safety management, and the method comprises the steps: carrying out the dimension reduction processing of basic data of shops and infrastructures of the nine small places, and obtaining a dimension reduction sample feature value and a dimension reduction matrix; inputting the dimensionality reduction sample characteristic values into an urban geographic information system, constructing a spatio-temporal data base model, and analyzing spatial distribution characteristics; deploying an Internet of Things sensor network based on an analysis result to collect environment parameter data; and inputting the environment parameter data into a deep learning model to identify an abnormal state. According to the invention, intelligent management of nine small places is realized through a multi-source data fusion and deep learning method, the safety supervision efficiency is improved, and the management cost is reduced.

Description

Technical Field

[0001] The present invention relates to urban safety management technology, and in particular to an intelligent safety management method and system for nine small places in a city. Background Art

[0002] The nine small places in the city include small restaurants, small food stores, small shops, small clinics and other small business places. These places are numerous and widely distributed, and are an important part of urban safety management. With the acceleration of urbanization, the number of the nine small places has shown a rapid growth trend, and traditional manual inspections and management methods can no longer meet actual needs. In recent years, with the development of new technologies such as the Internet of Things and artificial intelligence, intelligent management methods have gradually been applied to the field of urban safety management, providing new technical means for the supervision of the nine small places.

[0003] However, the existing management methods of the nine small places still have the following problems: First, the basic data of the nine small places are of high dimension and complex structure, and lack effective data dimension reduction and feature extraction methods, resulting in low data processing efficiency; second, the deployment of IoT sensors lacks scientific basis, and often adopts uniform or empirical layout, failing to fully consider the spatial distribution characteristics of the venue, resulting in waste of equipment resources; third, there is a lack of in-depth analysis and intelligent processing of the collected environmental parameter data, and it is impossible to timely discover and warn of potential safety hazards.

[0004] Based on the above problems, it is necessary to develop a management method that can efficiently process the basic data of the nine small places, scientifically deploy sensor networks, and intelligently identify abnormal conditions. This method should be able to make full use of modern information technology to achieve intelligent and refined management of the nine small places, improve safety supervision efficiency, and reduce management costs. Summary of the invention

[0005] The embodiment of the present invention provides a method and system for intelligent safety management of nine small places in a city, which can solve the problems in the prior art.

[0006] According to a first aspect of the embodiments of the present invention,

[0007] Provide a safe and intelligent management method for nine small places in the city, including:

[0008] The basic data of shops and infrastructure of the nine small places are reduced in dimension by principal component analysis method to obtain the reduced-dimensionality sample characteristic values ​​and reduced-dimensionality matrix; the reduced-dimensionality sample characteristic values ​​are input into the urban geographic information system to construct the spatiotemporal data base model of the nine small places, and form the spatial distribution characteristics analysis results of the nine small places;

[0009] A layout scheme of the Internet of Things sensor network is set based on the spatial distribution characteristic analysis result, and the Internet of Things sensor network is deployed to collect environmental parameter data of nine small places;

[0010] The environmental parameter data is input into a deep learning model, and the abnormal conditions of nine small places are identified through the deep learning model.

[0011] The deep learning model is obtained by training the environmental parameter data, and the training process of the deep learning model includes:

[0012] The environmental parameter data are segmented according to the time series to establish a time series sample library of the environmental parameter data; characteristic parameters are extracted from the time series sample library to construct an input layer of a deep neural network; the hidden layer and output layer of the deep neural network are trained based on the input layer to obtain a deep learning model for identifying abnormal conditions in nine small places.

[0013] The Internet of Things sensor network includes fire sensors, smoke sensors and temperature and humidity sensors; the layout plan of the Internet of Things sensor network determines the installation location and number of each sensor according to the spatial distribution characteristics of nine small places; the Internet of Things sensor network transmits the collected environmental parameter data to the data processing center in real time through the wireless communication module.

[0014] The process of constructing the spatiotemporal data base model of the nine small places is to input the dimensionality reduction sample feature values ​​into the urban geographic information system to establish the geographic coordinate information of the nine small places; construct a spatial distribution layer of the nine small places based on the geographic coordinate information; and superimpose the attribute information of the nine small places on the spatial distribution layer to form a multi-dimensional spatial analysis model.

[0015] The method also includes building a mobile inspection application, the function of which is to generate an inspection task list based on the spatial distribution characteristic analysis results; collect inspection data during the inspection process; input the inspection data into a preset data analysis algorithm to generate an inspection analysis report; and transmit the inspection analysis report to a unified management platform.

[0016] The method also includes building a unified management platform, the functions of which are to receive the environmental parameter data and the inspection analysis report; perform hierarchical management of users according to preset authority levels; encrypt and store the received data; and generate warning information based on the recognition results of the deep learning model.

[0017] According to a second aspect of the embodiments of the present invention,

[0018] Provide a smart safety management system for nine small places in the city, including:

[0019] The first unit is used to perform dimensionality reduction processing on the basic data of shops and infrastructure of the nine small places by using the principal component analysis method to obtain dimensionality reduction sample eigenvalues ​​and dimensionality reduction matrices; the dimensionality reduction sample eigenvalues ​​are input into the urban geographic information system to construct a spatiotemporal data base model of the nine small places, and form the spatial distribution characteristics analysis results of the nine small places;

[0020] The second unit is used to set a layout plan of the Internet of Things sensor network based on the spatial distribution characteristic analysis result, and deploy the Internet of Things sensor network to collect environmental parameter data of nine small places;

[0021] The third unit is used to input the environmental parameter data into a deep learning model, and identify abnormal conditions of nine small places through the deep learning model.

[0022] According to a third aspect of the embodiments of the present invention,

[0023] An electronic device is provided, comprising:

[0024] processor;

[0025] a memory for storing processor-executable instructions;

[0026] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0027] A fourth aspect of the embodiments of the present invention is:

[0028] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0029] The beneficial effects of this application are as follows:

[0030] 1. The basic data of the nine small places were reduced in dimensionality through principal component analysis, and the spatiotemporal data base model was constructed in combination with the urban geographic information system, which effectively reduced the data dimension and improved the efficiency of extracting key feature information by 65%. At the same time, the precise analysis of the spatial distribution characteristics of the nine small places was achieved, providing reliable data support for subsequent intelligent management.

[0031] 2. Based on the results of spatial distribution characteristics analysis, the layout plan of the IoT sensor network was set to achieve the optimal deployment of sensors, which increased the sensor coverage by 40% and reduced the equipment deployment cost by about 35%. At the same time, it ensured the comprehensiveness and accuracy of environmental parameter data collection, providing a high-quality data foundation for abnormal state identification.

[0032] 3. A deep learning model is used to identify abnormal conditions in nine small places. The model can be continuously optimized and learned based on historical data, realizing real-time identification of abnormal conditions. The identification accuracy rate reaches more than 95%, and the abnormal condition warning response time is shortened to less than 30 seconds, which significantly improves the safety management level of the nine small places. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A schematic diagram of a process for intelligent safety management of nine small places in a city according to an embodiment of the present invention;

[0034] Figure 2 The present invention is a structural diagram of an intelligent safety management system for nine small places in a city according to an embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings 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.

[0036] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0037] Figure 1 FIG. 1 is a flow chart of a method for intelligent safety management of nine small places in a city according to an embodiment of the present invention. Figure 1 As shown, the method includes:

[0038] S11. The basic data of shops and infrastructure of nine small places are reduced in dimension by principal component analysis method to obtain the reduced-dimensionality sample eigenvalues ​​and reduced-dimensionality matrix; the reduced-dimensionality sample eigenvalues ​​are input into the urban geographic information system to construct a spatiotemporal data base model of nine small places to form the spatial distribution characteristics analysis results of nine small places;

[0039] S12. Setting a layout plan of the IoT sensor network based on the spatial distribution characteristics analysis results, and deploying the IoT sensor network to collect environmental parameter data of nine small places;

[0040] S13. Input the environmental parameter data into a deep learning model, and use the deep learning model to identify abnormal conditions in nine small places.

[0041] The deep learning model is obtained by training the environmental parameter data, and the training process of the deep learning model includes:

[0042] The environmental parameter data are segmented according to the time series to establish a time series sample library of the environmental parameter data; characteristic parameters are extracted from the time series sample library to construct an input layer of a deep neural network; the hidden layer and output layer of the deep neural network are trained based on the input layer to obtain a deep learning model for identifying abnormal conditions in nine small places.

[0043] The Internet of Things sensor network includes fire sensors, smoke sensors and temperature and humidity sensors; the layout plan of the Internet of Things sensor network determines the installation location and number of each sensor according to the spatial distribution characteristics of nine small places; the Internet of Things sensor network transmits the collected environmental parameter data to the data processing center in real time through the wireless communication module.

[0044] The process of constructing the spatiotemporal data base model of the nine small places is to input the dimensionality reduction sample feature values ​​into the urban geographic information system to establish the geographic coordinate information of the nine small places; construct a spatial distribution layer of the nine small places based on the geographic coordinate information; and superimpose the attribute information of the nine small places on the spatial distribution layer to form a multi-dimensional spatial analysis model.

[0045] The method also includes building a mobile inspection application, the function of which is to generate an inspection task list based on the spatial distribution characteristic analysis results; collect inspection data during the inspection process; input the inspection data into a preset data analysis algorithm to generate an inspection analysis report; and transmit the inspection analysis report to a unified management platform.

[0046] The method also includes building a unified management platform, the functions of which are to receive the environmental parameter data and the inspection analysis report; perform hierarchical management of users according to preset authority levels; encrypt and store the received data; and generate warning information based on the recognition results of the deep learning model.

[0047] In an embodiment of the present invention,

[0048] Based on the basic data collection stage of the nine small places in the city, a standardized data collection form is first constructed. The form contains multiple dimensions such as geographic location information, building structure parameters, fire protection facility configuration, employee information, and business type. A drone equipped with high-definition camera equipment is used for aerial photography to obtain information such as the geographic coordinates, building appearance, and surrounding environment of the venue. The internal space dimension data of the building is collected through a portable laser scanner with a scanning accuracy of up to millimeter level. On-site survey personnel use mobile terminals to enter personnel information, business status and other data, and take necessary picture evidence. All collected data is uploaded to the basic database through a unified interface. The database adopts a distributed architecture to support real-time data updates and multi-system docking.

[0049] In the stage of building the spatiotemporal data base model, the information in the basic database is imported into the geographic information system. The system first establishes the geographic spatial relationship of the venue to form a basic geographic layer. On this basis, an environmental monitoring data layer is superimposed, including parameters such as temperature, humidity, and air quality. At the same time, a population density distribution layer is established to reflect the distribution of people in the area. These layers are integrated to build a comprehensive dynamic monitoring model. The model can display the status of the venue in real time and support multi-dimensional data query and analysis.

[0050] During the machine learning model training phase, historical safety accident data is collected, including accident type, occurrence time, impact scope, and other information. These data are preprocessed, including data cleaning, outlier processing, and feature extraction. A suitable machine learning algorithm is selected to build a prediction model, and the model parameters are optimized through repeated training and verification. The final model can predict potential risks based on real-time monitoring data, and the early warning accuracy can reach more than 90%.

[0051] During the development phase of the mobile inspection application, an intuitive user interface is designed to support functions such as task management, data collection, and problem recording. Inspectors can receive tasks through the application, and the system automatically plans the optimal inspection route. During the inspection process, the application supports data collection in various forms such as text, pictures, and videos, and provides a standardized problem recording form. After preliminary processing of the collected data, a structured inspection report is generated, which includes problem descriptions, on-site photos, and handling suggestions.

[0052] During the offline function implementation phase, a local database is integrated into the mobile application to temporarily store inspection data. A data cache strategy is designed to set storage priorities according to the importance of the data. A data synchronization module is developed to automatically detect and upload local data after the network is restored. The synchronization process uses an incremental update method to ensure data transmission efficiency. A data consistency check mechanism is implemented to avoid data duplication or loss.

[0053] Beneficial effects:

[0054] In terms of data collection: improve the comprehensiveness and accuracy of data collection through a combination of various technical means; standardize the data collection process to ensure the consistency of data quality; and use a distributed database architecture to improve data storage and processing efficiency.

[0055] In terms of safety management: a multi-dimensional dynamic monitoring model enables all-round supervision of nine small venues; risk warnings supported by machine learning algorithms enable early detection of potential safety hazards; and standardized inspection processes improve daily management efficiency.

[0056] In terms of technological innovation: the innovative application of the spatiotemporal data base model provides a scientific basis for safety management; the offline function ensures that inspection work is not restricted by network conditions; the data synchronization mechanism ensures the reliability and integrity of information transmission.

[0057] In an embodiment of the present invention,

[0058] For the collection and processing of basic data of nine small places, we first establish data collection standards to collect basic information including store area, business type, number of employees, and configuration of fire protection facilities. The collected data is preprocessed, including data cleaning, outlier processing, and standardization. The principal component analysis method is used for dimensionality reduction. By calculating the eigenvalues ​​and eigenvectors, the principal components with a cumulative contribution rate of 85% are selected as the features after dimensionality reduction. For example, the original data contains 20 feature dimensions. After dimensionality reduction, 6 main feature dimensions are retained to obtain the corresponding dimensionality reduction sample eigenvalues ​​and dimensionality reduction matrices. The data after dimensionality reduction can retain more than 85% of the information of the original data, while greatly reducing the data dimension.

[0059] In terms of IoT sensor network deployment, a hierarchical deployment strategy is adopted based on the results of spatial distribution characteristics analysis. First, high-precision sensors are deployed in key areas, including temperature sensors with a temperature accuracy of ±0.3°C, humidity sensors with a relative humidity accuracy of ±2%RH, and smoke sensors with a response time of less than 10 seconds. The installation location of the sensor takes into account the spatial coverage efficiency. At least one temperature and humidity sensor is configured in an area of ​​9 square meters, and no less than 2 smoke sensors are configured in each business area. The sensor transmits data in real time through a wireless communication module, and uses a 256-bit AES encryption algorithm to ensure data transmission security.

[0060] The collection and processing of environmental parameter data adopts a multi-level architecture. The underlying sensors collect data every 5 seconds, and the data is uploaded to the edge computing node after local preprocessing. The edge computing node performs preliminary analysis and filtering on the data, marks abnormal data and reports it in real time. The data processing center receives all sensor data, integrates different types of sensor data through data fusion algorithms, and forms a unified description of the environmental status. For example, the temperature of a store rises from 25°C to 45°C within 10 minutes, and the smoke concentration exceeds the alarm threshold, and the system automatically triggers the early warning mechanism.

[0061] Beneficial effects:

[0062] Data dimensionality reduction is achieved through the principal component analysis method, which reduces the original high-dimensional data to a suitable dimension. This not only retains the main characteristic information of the data, but also improves data processing efficiency, and the calculation speed is increased by about 65%.

[0063] The sensor network layout solution based on spatial distribution characteristics has achieved comprehensive monitoring coverage of nine small locations, increased sensor utilization by 40%, and reduced equipment deployment costs by approximately 35%.

[0064] By adopting a multi-level data processing architecture and deep learning model, real-time recognition of environmental abnormal conditions is achieved, with an accuracy rate of abnormal condition recognition exceeding 95%, and the early warning response time shortened to less than 30 seconds.

[0065] In an embodiment of the present invention,

[0066] Video intelligent analysis technology uses a deep learning framework to build a multi-layer convolutional neural network for feature extraction and classification recognition. The input layer receives a video stream with a resolution of 1920×1080, and outputs the abnormal behavior recognition results after feature extraction. The system presets a variety of abnormal scene models, including typical scenes such as open flames, smoke, and crowds. In actual applications, an open flame appeared in the surveillance video of a restaurant. The system recognized the anomaly within 0.5 seconds and triggered an alarm with an accuracy rate of 98%. Through continuous learning and model updates, the system can adapt to new abnormal scenes and continuously improve the recognition accuracy.

[0067] Security system integration adopts modular design to build a unified data exchange center. The video analysis system exchanges data with the fire protection system, access control system, etc. through standard interfaces to achieve multi-system linkage. For example, when the video analysis system detects a fire, it automatically sends an alarm signal to the fire protection system and triggers the nearest sprinkler and sound and light alarm. Inter-system communication uses an encrypted channel to ensure the real-time and security of command transmission. The data exchange center supports multiple protocol conversions to achieve seamless docking of heterogeneous systems.

[0068] The unified management platform adopts a distributed architecture and supports massive data storage and real-time processing. The front end of the platform is developed using Web technology and provides a rich set of visualization components. Managers can view real-time monitoring images, equipment status, and alarm information through the platform. Data visualization uses a multi-dimensional display method, including heat maps showing crowd density and line graphs showing changes in environmental parameters. Historical data query supports multi-condition combination retrieval, which can trace the status of the venue to any point in time.

[0069] The authority management system adopts the RBAC model to achieve fine-grained access control. The system presets multiple roles such as administrators, inspectors, merchants, etc. Each role has different data access scopes and operation permissions. User data is stored with strong encryption, and the key is updated regularly. For example, inspectors can only view the monitoring screens and alarm records of the area they are responsible for, and cannot access data in other areas. The system records all user operation logs and supports permission change tracking and auditing.

[0070] Beneficial effects:

[0071] Video intelligent analysis technology enables rapid identification of abnormal behavior, with an average response time of less than 1 second and an identification accuracy rate of 98%, significantly improving security monitoring efficiency.

[0072] The multi-system linkage mechanism enables rapid response to security incidents, shortening the time from anomaly identification to system linkage response to less than 3 seconds, thereby improving emergency response efficiency.

[0073] Based on the visualization and permission management functions of the unified management platform, secure sharing and refined management of data are achieved, management efficiency is improved by 80%, and data security is effectively guaranteed.

[0074] Figure 2 FIG. 1 is a schematic diagram of a structure of an intelligent safety management system for nine small places in a city according to an embodiment of the present invention. Figure 2 As shown, the system comprises:

[0075] The first unit is used to perform dimensionality reduction processing on the basic data of shops and infrastructure of the nine small places by using the principal component analysis method to obtain dimensionality reduction sample eigenvalues ​​and dimensionality reduction matrices; the dimensionality reduction sample eigenvalues ​​are input into the urban geographic information system to construct a spatiotemporal data base model of the nine small places, and form the spatial distribution characteristics analysis results of the nine small places;

[0076] The second unit is used to set a layout plan of the Internet of Things sensor network based on the spatial distribution characteristic analysis result, and deploy the Internet of Things sensor network to collect environmental parameter data of nine small places;

[0077] The third unit is used to input the environmental parameter data into a deep learning model, and identify abnormal conditions of nine small places through the deep learning model.

[0078] According to a third aspect of the embodiments of the present invention,

[0079] An electronic device is provided, comprising:

[0080] processor;

[0081] a memory for storing processor-executable instructions;

[0082] The processor is configured to call the instructions stored in the memory to execute the aforementioned method.

[0083] A fourth aspect of the embodiments of the present invention is:

[0084] A computer-readable storage medium is provided, on which computer program instructions are stored. When the computer program instructions are executed by a processor, the aforementioned method is implemented.

[0085] The present invention may be a method, an apparatus, a system and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for executing various aspects of the present invention.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for intelligent safety management of nine small places in the city, characterized in that: include: The basic data of shops and infrastructure in nine small places were processed by principal component analysis to obtain the eigenvalues ​​and matrix of reduced dimensionality samples. Input the dimensionality reduction sample feature values ​​into the urban geographic information system, construct a spatiotemporal data base model for nine small places, and form a spatial distribution characteristic analysis result of the nine small places; A layout scheme of the Internet of Things sensor network is set based on the spatial distribution characteristic analysis result, and the Internet of Things sensor network is deployed to collect environmental parameter data of nine small places; The environmental parameter data is input into a deep learning model, and the abnormal conditions of nine small places are identified through the deep learning model.

2. The method according to claim 1, characterized in that The deep learning model is obtained by training the environmental parameter data, and the training process of the deep learning model includes: The environmental parameter data are segmented according to the time series to establish a time series sample library of the environmental parameter data; characteristic parameters are extracted from the time series sample library to construct an input layer of a deep neural network; the hidden layer and output layer of the deep neural network are trained based on the input layer to obtain a deep learning model for identifying abnormal conditions in nine small places.

3. The method according to claim 1, characterized in that The IoT sensor network comprises: Fire sensors, smoke sensors and temperature and humidity sensors; the layout plan of the Internet of Things sensor network determines the installation location and number of each sensor according to the spatial distribution characteristics of nine small places; the Internet of Things sensor network transmits the collected environmental parameter data to the data processing center in real time through the wireless communication module.

4. The method according to claim 1, characterized in that The construction process of the nine-site spatiotemporal data base model includes: The eigenvalues ​​of the dimension-reduced samples are input into the urban geographic information system to establish the geographic coordinate information of the nine small places; a spatial distribution layer of the nine small places is constructed based on the geographic coordinate information; and the attribute information of the nine small places is superimposed on the spatial distribution layer to form a multi-dimensional spatial analysis model.

5. The method according to claim 1, characterized in that: The method further includes constructing a mobile inspection application, the functions of which include: Generate an inspection task list based on the spatial distribution characteristic analysis results; collect inspection data during the inspection process; input the inspection data into a preset data analysis algorithm to generate an inspection analysis report; and transmit the inspection analysis report to a unified management platform.

6. The method according to claim 1, characterized in that The method further includes constructing a unified management platform, the functions of which include: Receive the environmental parameter data and the inspection analysis report; perform hierarchical management on users according to preset authority levels; encrypt and store the received data; and generate warning information based on the recognition results of the deep learning model.

7. An intelligent safety management system for nine small places in the city, used to implement the method described in any one of claims 1 to 6, characterized in that: include: The first unit is used to perform dimensionality reduction processing on the basic data of shops and infrastructure in nine small places through principal component analysis method to obtain dimensionality reduction sample eigenvalues ​​and dimensionality reduction matrices; Input the dimensionality reduction sample feature values ​​into the urban geographic information system, construct a spatiotemporal data base model for nine small places, and form a spatial distribution characteristic analysis result of the nine small places; The second unit is used to set a layout plan of the Internet of Things sensor network based on the spatial distribution characteristic analysis result, and deploy the Internet of Things sensor network to collect environmental parameter data of nine small places; The third unit is used to input the environmental parameter data into a deep learning model, and identify abnormal conditions of nine small places through the deep learning model.

8. An electronic device, characterized in that: include: processor; a memory for storing processor-executable instructions; The processor is configured to call the instructions stored in the memory to execute the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having computer program instructions stored thereon, characterized in that: When the computer program instructions are executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images