Regional floating population density analysis and prediction service system using city public transportation data
The regional floating population density analysis system uses deep learning on public transportation data to predict and prevent accidents by classifying and predicting density, enhancing safety management and infrastructure planning.
Patent Information
- Application Number
- KR1020230158132
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2023-11-15
- Publication Date
- 2026-11-11
- Estimated Expiration
- 2043-11-15
AI Technical Summary
Conventional systems for analyzing and predicting population density are unable to provide proactive measures to prevent accidents, crimes, and natural disasters due to their inability to predict crowd density in real-time.
A regional floating population density analysis and prediction service system utilizing urban public transportation data, employing a deep learning algorithm to classify and predict density based on bus and subway boarding data, and providing this information as a web service to police and local governments.
Enables proactive measures to prevent accidents and natural disasters by predicting potential risks, improving safety management efficiency and resource allocation, and identifying areas requiring infrastructure development.
Smart Images

Figure 112023126386331-PAT00001_ABST
Abstract
Description
Technology Field
[0001] The present invention relates to a density analysis system, and more specifically, to a service system for analyzing and predicting the density of floating population by zone using urban public transportation data. Background Technology
[0003] Population density is a key factor affecting safety management in various ways and refers to the number of people living and working in a specific area. This indicator can lead to overcrowding, traffic congestion, and increased crime rates, and can also affect the environment, including air quality, water resources, and waste management.
[0004] Therefore, there is a growing need to analyze population density to identify potential safety risks and to implement proactive measures to mitigate them.
[0005] Local governments are establishing measures to prevent accidents caused by crowd congestion; for example, crowd density analysis systems using CCTV, mobile communication networks for analyzing floating population density, congestion verification systems using mobile networks, and floating population services for commercial area analysis are being implemented.
[0006] These conventional systems utilize communication networks and CCTVs to check density in real time, but they have the disadvantage of being unable to predict it, making them vulnerable to rapid response to specific situations. Prior art literature
[0008] KR10-2536579B The problem to be solved
[0009] The present invention is proposed to solve the technical problems described above and provides a regional pedestrian traffic density analysis and prediction service system utilizing urban public transportation data, which enables the prediction of potential risks and preventive measures to prevent accidents, crimes, and natural disasters by analyzing pedestrian traffic density based on data from various situations. means of solving the problem
[0011] According to one embodiment of the present invention for solving the above problems, a regional floating population density analysis and prediction service system is provided, comprising: a public data processing unit that processes public transportation boarding data by bus stop to extract valid data; a deep learning learning unit that inputs the data provided by the public data processing unit into a preset deep learning algorithm to automatically classify the floating population density by region; and a control center that provides the regional floating population density classified by the deep learning learning unit as a web service.
[0012] In addition, the public transportation boarding data by stop in the present invention is characterized by including bus boarding data and subway boarding data.
[0013] In addition, the pre-set deep learning algorithm in the present invention is characterized as being a DNN (Deep neural network).
[0014] In addition, the control center in the present invention is characterized by providing the density of floating population by region as a web service by mapping information on the expected occurrence of safety accidents and providing it to police terminals and local government terminals nationwide. Effects of the invention
[0016] The regional population density analysis and prediction service system (1) of the present invention can predict the level of congestion by analyzing public transportation boarding and alighting data for specific days, times, and places. That is, by analyzing population density based on data from various situations, it is possible to predict potential risks and take preventive measures to prevent accidents, crimes, and natural disasters.
[0017] That is, the regional population density analysis and prediction service system (1) utilizes bus and subway user type and trend analysis for population prediction, which has many variables depending on various events such as date, time, event, and weather.
[0018] By utilizing the system of the present invention, safety accidents such as stampede can be prevented through data-based population density analysis, and the occurrence of safety accidents can be blocked by sharing safety information with the police, local governments, etc.
[0019] In addition, by analyzing population density, potential safety risks can be identified, and proactive measures can be taken to prevent accidents, crimes, and natural disasters.
[0020] In addition, by analyzing population density, the police and local governments can determine the most efficient routes and allocate resources more effectively, enabling them to reduce response times in emergency situations.
[0021] In addition, urban planning departments, such as local governments, can identify areas requiring the development of additional infrastructure, such as roads, public transportation, and emergency services, thereby enabling efficient transportation systems, better emergency services, and improved quality of life for citizens.
[0022] In addition, by analyzing population density, potential health risks such as waste generation, air pollution, and water pollution can be identified, and proactive measures to mitigate them can be taken, thereby improving public health and urban services. Brief explanation of the drawing
[0024] FIG. 1 is a conceptual diagram of the regional population density analysis and prediction service system (1) of the present invention. FIG. 2 is a configuration diagram according to an embodiment of the regional population density analysis and prediction service system (1) of FIG. 1. FIG. 3 is an example diagram of the data collection and processing process in the regional population density analysis and prediction service system (1). FIG. 4 is an example diagram of the artificial neural network construction process in the regional population density analysis and prediction service system (1). FIG. 5 is an example diagram of the web service construction process in the regional population density analysis and prediction service system (1). Specific details for implementing the invention
[0025] Hereinafter, in order to explain in detail enough for a person skilled in the art to easily implement the technical concept of the present invention, embodiments of the present invention will be described with reference to the attached drawings.
[0027] FIG. 1 is a conceptual diagram of the regional population density analysis and prediction service system (1) of the present invention, and FIG. 2 is a configuration diagram according to an embodiment of the regional population density analysis and prediction service system (1) of FIG. 1.
[0029] The regional population density analysis and prediction service system (1) according to the present embodiment includes only a brief configuration to clearly explain the technical concept to be proposed.
[0030] Referring to FIGS. 1 and 2, the regional population density analysis and prediction service system (1) is configured to include a public data processing unit (100), a deep learning learning unit (200), and a control center (300).
[0032] The main operations of the regional population density analysis and prediction service system (1) configured as described above are as follows.
[0035] The public data processing unit (100) processes public transportation boarding data by stop and extracts valid data.
[0036] In this embodiment, public transportation boarding data by stop includes bus boarding data and subway boarding data. If public transportation such as a ground tram exists, all boarding data for that public transportation is also included.
[0038] The deep learning training unit (200) inputs data provided by the public data processing unit (100) into a preset deep learning algorithm to automatically predict the density of the floating population by region.
[0039] In this embodiment, the pre-set deep learning algorithm is a Deep Neural Network (DNN), and other deep learning algorithms may be used depending on the embodiment.
[0041] The control center (300) provides a web service of the density of floating population by region classified by the deep learning learning unit (200), and can map information on the expected occurrence of safety accidents and provide it to police terminals and local government terminals nationwide. That is, the system of the present invention operates in the form of a control service by linking a web service and an artificial intelligence model, and can contribute to the prevention of safety accidents by distributing information on the expected occurrence of safety accidents to the police and local governments nationwide.
[0043] The regional population density analysis and prediction service system (1) of the present invention described above provides a service that can prevent safety accidents by utilizing big data and artificial intelligence technology to realize a safe smart city free from safety accidents. In addition, it can predict the possibility of safety accidents based on population density big data and share the prediction information with the police and local governments to prevent accidents from occurring at the source.
[0045] Figure 3 is an example of the data collection and processing process in the regional population density analysis and prediction service system (1).
[0046] Referring to Fig. 3, the data collection and processing process includes a data collection step, a DB creation and format standardization step, a data cleansing step, a data analysis and labeling step, and a monitoring step.
[0047] During the data collection phase, population density data is collected, which includes bus stop information and the number of passengers boarding and alighting by time of day.
[0048] In the DB creation and format standardization phase, data is organized into a comprehensive database that is easily accessible to all relevant stakeholders, and can be structured, for example, in JSON format.
[0049] In the data cleansing stage, data is organized and verified through a cleansing process to remove all inaccurate or duplicate data. Specifically, tasks such as checking for missing values, correcting errors, and standardizing data are performed.
[0050] In the data analysis and labeling stage, data is analyzed to identify patterns and trends of product defects, and elements suitable for checking labeling data and AI training are selected.
[0051] In the monitoring phase, the database is continuously monitored and updated to track progress and verify that the execution plan is being effectively implemented. This involves regularly reviewing data, identifying new defects, and adjusting the execution plan as necessary.
[0053] Figure 4 is an example diagram of the artificial neural network construction process in the regional population density analysis and prediction service system (1).
[0054] In the artificial neural network construction process, training is performed based on the previously processed training dataset. Deep Neural Network (DNN) and Convolutional Neural Network (CNN) models are used for predicting defect rates and causes, although other models may be used depending on the embodiment. Input data, the number of hidden layers, activation functions, etc., are determined based on the processed training data and reflected in the neural network design, and are modified according to the prediction results. While the output values are planned to be defect rates and causes, other aspects may be added during the service development process. Python may be used as the development language, and frameworks such as TensorFlow and Keras may be utilized.
[0056] Referring to Fig. 4, the artificial neural network construction process includes a model selection stage, a model configuration stage, a model training stage, a model evaluation stage, a model deployment stage, and a model maintenance stage.
[0057] In the model selection phase, a neural network architecture to be used for the prediction model is chosen by considering data volume and type, complexity, and GPU resources.
[0058] In the model configuration phase, the neural network is configured by setting the number of hidden layers, activation function, and optimization function, and the training rate and batch size are adjusted based on the results of initial tests and general cases.
[0059] In the model training phase, neural network training is performed using training data. The progress of the model's training, such as loss and accuracy, is monitored, hyperparameters are adjusted as needed, overfitting is monitored using validation data, and training is stopped when performance improvement ceases.
[0060] In the model evaluation phase, the model trained on the validation data is evaluated. Model performance is assessed by calculating metrics such as accuracy, precision, recall, and F1 score, and if the performance is unsatisfactory, the process returns to the previous step to adjust the model configuration.
[0061] In the model deployment phase, the model trained in an environment suitable for web services is deployed to predict product defect rates and causes, and the performance of the deployed model is monitored and updated as necessary.
[0062] In the model maintenance phase, the model is updated with new data, retrained as needed, and issues are resolved; regular maintenance is performed to ensure the model remains accurate and effective over time.
[0064] Figure 5 is an example of the web service construction process in the regional population density analysis and prediction service system (1).
[0065] In establishing a web service integrated with an AI model for the advance prediction of population density and safety accidents, the web service is built by linking with the previously developed AI model; key functions include population density and safety accident prediction and dashboards, and Flask or Django frameworks can be utilized for integration with the AI model.
[0067] Referring to Fig. 5, the web service construction process includes a backend development phase, a frontend development phase, and an integration and deployment phase.
[0068] In the backend development phase, the backend system includes an ANN model and APIs for prediction, and the backend can be built using Python, Flask, or Django frameworks.
[0069] In the frontend development phase, you can build the frontend using React and JavaScript frameworks to interact with web services.
[0070] In the integration and deployment phase, you can use REST or GraphQL to connect APIs to the frontend for system integration and deploy, for example, through Naver Cloud Platform.
[0072] As described above, the regional population density analysis and prediction service system (1) of the present invention can prevent safety accidents and improve citizen convenience by utilizing urban data, and can improve safety accident risk and safety management efficiency by predicting regional population density and safety accidents before potential safety accidents occur through AI services, and can reduce the risk of human error and increase accuracy by automating the AI-based population density calculation and safety accident prediction process, and can accelerate problem solving and reduce costs by analyzing big data related to population density and identifying the root causes of safety accidents in real time. In addition, it can improve safety management efficiency by analyzing the causes of safety accidents based on population density and identifying areas that need improvement.
[0074] Furthermore, the system of the present invention can predict congestion levels by analyzing public transportation boarding and alighting data for specific days, times, and locations. In other words, by analyzing the density of the floating population based on data from various situations, it is possible to predict potential risks and take proactive preventive measures to prevent accidents, crimes, and natural disasters.
[0075] That is, the regional population density analysis and prediction service system (1) utilizes bus and subway user type and trend analysis for population prediction, which has many variables depending on various events such as date, time, event, and weather.
[0076] By utilizing the system of the present invention, safety accidents such as stampede can be prevented through data-based population density analysis, and the occurrence of safety accidents can be blocked by sharing safety information with the police, local governments, etc.
[0077] In addition, by analyzing population density, potential safety risks can be identified, and proactive measures can be taken to prevent accidents, crimes, and natural disasters.
[0078] In addition, by analyzing population density, the police and local governments can determine the most efficient routes and allocate resources more effectively, enabling them to reduce response times in emergency situations.
[0079] In addition, urban planning departments, such as local governments, can identify areas requiring the development of additional infrastructure, such as roads, public transportation, and emergency services, thereby enabling efficient transportation systems, better emergency services, and improved quality of life for citizens.
[0080] In addition, by analyzing population density, potential health risks such as waste generation, air pollution, and water pollution can be identified, and proactive measures to mitigate them can be taken, thereby improving public health and urban services.
[0082] As such, those skilled in the art to which the present invention pertains will understand that the present invention may be implemented in other specific forms without altering its technical concept or essential features. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. The scope of the present invention is defined by the claims set forth below rather than by the detailed description above, and all modifications or variations derived from the meaning and scope of the claims and their equivalents should be interpreted as being included within the scope of the present invention. Explanation of the symbols
[0084] 100 : Public Data Processing Department 200 : Deep learning training unit 300 : Control Center
Claims
Claim 1 A public data processing unit that processes public transportation boarding data by bus stop to extract valid data; a deep learning learning unit that inputs the data provided by the public data processing unit into a preset deep learning algorithm to classify the density of floating population by region; A regional population density analysis and prediction service system comprising: a control center that provides a web service of the regional population density classified by the deep learning unit; wherein the public data processing unit collects public transportation boarding data by stop, performs a data collection and processing process that removes inaccurate or duplicate data, organizes, and verifies the boarding data through a cleansing process; wherein the deep learning unit receives data provided by the public data processing unit and classifies the population density by specific day, time, and region by performing population prediction based on date, time, event, and weather using the analysis of public transportation user patterns and trends by stop; and wherein the control center, in providing the regional population density as a web service, maps information on the expected occurrence of safety accidents by identifying potential safety risks and provides it to police terminals and local government terminals nationwide via a cloud platform. Claim 2 A regional population density analysis and prediction service system according to claim 1, characterized in that the public transportation boarding data by stop includes bus boarding data and subway boarding data. Claim 3 A regional population density analysis and prediction service system according to claim 1, characterized in that the preset deep learning algorithm is a Deep Neural Network (DNN). Claim 4 delete
Citation Information
Patent Citations
Method and system of predicting infectious diseases using big data and artificial intelligence
KR1020230020779A
Commercial area analysis system using public data and open API
KR102429139B1
System for providing real time public transportation congestion
KR1020220110399A
Method for forcastiing the distribution and density of pedestrian
KR102098305B1
