Information provision system

Abstract

[Problem] To provide an information provision system that is capable of effectively performing risk prediction by visualizing various situations in a real space in real time and analyzing collected data. [Solution] An information provision system (1) comprising: a situation data acquisition unit (2) that acquires situation data which has been obtained by imaging a condition in a real space; a virtual space display unit (3) that displays, on the basis of the situation data, information pertaining to the situation at coordinates in a virtual space corresponding to the real space, the coordinates corresponding to a position in the real space where the situation has occurred; an analysis unit (5) that calculates the degree of risk on the basis of the situation data; and a notification unit (6) that transmits a warning notification when the degree of risk exceeds a threshold value, wherein the situation data includes information pertaining to at least one of a traffic accident, a crime, an anomaly of an infrastructure facility, and a natural disaster.

Description

Information Provision System

[0001] The present invention relates to an information provision system that visualizes conditions in real space on a virtual space.

[0002] Conventionally, techniques for scanning a real space using a sensor and generating a map corresponding to the real space are known. For example, the system described in Patent Document 1 generates a three-dimensional map based on scan data obtained by using an omnidirectional laser rangefinder mounted on a moving body while moving the moving body with a remote control device.

[0003] Japanese Unexamined Patent Publication No. 2017-198517

[0004] In recent years, social issues such as the aging of infrastructure equipment, the intensification of natural disasters, traffic accidents, and crime have become increasingly serious. To address these issues, it is important to grasp various conditions in real space in real time and predict risks.

[0005] However, conventional systems such as that described in Patent Document 1 require special equipment such as an omnidirectional laser rangefinder, and are limited to periodic acquisition of scan data, making it difficult to grasp conditions in real time and predict risks.

[0006] Accordingly, an object of the present invention is to provide an information provision system capable of visualizing various conditions in real space in real time, and performing effective risk prediction by analyzing collected data.

[0007] To achieve the above object, the present invention provides an information provision system comprising: a condition data acquisition unit that acquires condition data obtained by imaging conditions in a real space; and a virtual space display unit that, based on the condition data, displays information related to the condition at coordinates on a virtual space corresponding to the real space, where the coordinates correspond to the location of the condition occurrence in the real space, wherein the condition data includes information related to at least any one of traffic accidents, crime, abnormalities in infrastructure equipment, and natural disasters.

[0008] According to the present invention, by using world map data provided by satellite image and map data providers, and combining it with video data collected from user-submitted videos and apps such as dashcams, it becomes possible to grasp the situation in real time regarding traffic accidents, crimes, infrastructure malfunctions, natural disasters, etc.

[0009] Furthermore, the situation data includes time information regarding the time when the situation occurred, allowing for the accumulation of data in a time series. This enables analysis of temporal changes and anomaly detection through comparison with past data.

[0010] Furthermore, the system includes an analysis unit that calculates the degree of risk in the real world based on the situational data, and the virtual space display unit can display the degree of risk in the virtual space. This allows for a quantitative evaluation of the degree of risk in each region and enables the implementation of preventive measures.

[0011] Furthermore, the analysis unit can calculate the level of risk for each time period based on the situational data. This makes it possible to predict risks and take countermeasures according to the time of day.

[0012] Furthermore, the analysis unit can calculate the degree of risk in the real world by comparing accumulated past situation data with current situation data. This makes it possible to detect deviations from normal conditions and to detect abnormalities early.

[0013] Furthermore, the situation data includes meteorological information at the location where the situation occurs, and the analysis unit can analyze the correlation between the meteorological information and the degree of risk. This allows for the quantification of the influence of meteorological conditions on the degree of risk, enabling predictions linked to weather forecasts.

[0014] Furthermore, the analysis unit can analyze the correlation between geographical characteristics and the risk level, and predict the risk level of other areas with similar geographical characteristics. This makes it possible to take preventative measures even in areas where events have not yet occurred.

[0015] According to the present invention, it becomes possible to visualize various situations in real space in real time and analyze the collected data to make effective risk predictions. This makes it possible to take preventative measures against social issues such as infrastructure malfunctions and natural disasters.

[0016] This figure shows an overview of the processing in one embodiment of the information provision system according to the present invention. This is a configuration diagram of one embodiment of the information provision system according to the present invention. This is a flowchart showing the processing flow in one embodiment of the information provision system according to the present invention. This is a flowchart showing the detailed analysis processing flow in one embodiment of the information provision system according to the present invention.

[0017] Next, embodiments for carrying out the present invention will be described in detail with reference to the drawings.

[0018] As shown in Figure 1, the information provision system 1 according to the present invention acquires situation data in the real world and displays it in a virtual world. Situation data in the real world includes traffic accidents, crimes, infrastructure malfunctions, natural disasters, etc. This situation data is acquired by the situation data acquisition unit 2 and displayed in the virtual world by the virtual world display unit 3.

[0019] Figure 2 shows a detailed configuration of the information provision system 1 described above. This information provision system 1 comprises a status data acquisition unit 2, a virtual space display unit 3, a storage unit 4, an analysis unit 5, a notification unit 6, and a control unit 7. The information provision system 1 is connected to the user terminal 10 and the information processing system 11 via a network for communication.

[0020] The situation data acquisition unit 2 is an interface for acquiring data on various events in the real world. Situation data, including traffic accident information, crime occurrence information, infrastructure equipment malfunction information, and natural disaster information, can be collected from various sensors, external databases, user input, etc.

[0021] The data sources acquired by the situation data acquisition unit 2 include data obtained from various mobile objects and business activities. Specifically, this includes videos and photographs of road conditions acquired from cameras and sensors mounted on bicycles, video data of the region acquired from vehicles used in postal delivery and courier services, airplanes or drones, and videos and photographs taken by tourists at tourist destinations. In addition, 2D data such as design drawings and CAD data related to buildings and urban infrastructure, as well as 3D data such as 3D scan data and 3D model data, can also be acquired as situation data.

[0022] Situation data can include attribute information such as the time of occurrence, location information (latitude and longitude), type, and severity, and the data format can support standard formats such as JSON, XML, and CSV. For example, in the case of traffic accident data, situation data would include information such as "February 25, 2025, 14:35, Minato Ward, Tokyo, collision accident at an intersection, 2 people with minor injuries."

[0023] The virtual space display unit 3 has the function of mapping and displaying the acquired situation data onto the virtual space. The virtual space reproduces the real geographic space and can be either a two-dimensional map or a three-dimensional metaverse space. In addition, each situation data is placed at a geographic location corresponding to the real space.

[0024] Methods for displaying data in the virtual space include icon display (different icons for each type of event), heatmap display (representing density and frequency with varying shades of color), and color-coding display (color coding according to the level of danger). Furthermore, a function to display changes over time as animations can also be implemented.

[0025] The memory unit 4 is a database that stores situational data and its analysis results in chronological order. It can employ a relational database and an unstructured database configuration, and can be designed to process large amounts of data.

[0026] The memory unit 4 can store situation data and its attribute information, time-series analysis results, risk assessment results, etc. The data can be stored as time information. In addition, data compression functions for long-term storage and data redundancy functions for use in the event of failure can also be implemented.

[0027] The analysis unit 5 has the function of analyzing accumulated situational data and calculating the degree of risk. Analysis methods such as statistical time series analysis, machine learning-based predictive analysis, spatial clustering analysis, and correlation analysis can be implemented.

[0028] To calculate the risk level, an evaluation algorithm can be employed that considers factors such as the frequency of the event, the severity of the event, its geographical concentration, the rate of change over time, and the degree of similarity with similar patterns. These factors can then be weighted and quantified.

[0029] The notification unit 6 has the function of sending a warning notification to the user, etc., if the risk level exceeds a threshold as a result of the analysis. Notification methods include push notifications (for mobile applications), email notifications, SMS notifications, and notifications to external systems via API integration.

[0030] Notification content can include the type of hazard, location, expected impact, and recommended countermeasures. It is also possible to customize the notification content according to the recipient's role. Notification priority can be set according to the level of hazard.

[0031] The control unit 7 has the function of controlling the operation of the entire system and managing the coordination between each component. Specific functions include data processing scheduling, resource allocation optimization, fault detection and recovery, and security management.

[0032] The hardware configuration of this system can include, for example, a cloud-based server environment, a data collection infrastructure using IoT devices and sensor networks, storage using distributed databases, and processors for high-speed analytical processing.

[0033] Next, the processing flow in one embodiment of the present invention will be described with reference to Figure 3. Figure 3 is a flowchart showing the basic monitoring flow in one embodiment of the present invention.

[0034] As shown in Figure 3, the process begins at Start, and in step S1, the system is initialized. The initialization process includes system diagnosis, database connection confirmation, and external system integration confirmation.

[0035] In step S2, the situation data acquisition unit 2 acquires data on various events in the real world. Data sources include real-time data from public APIs, sensing data from IoT sensors, anomaly detection from SNS data, data provided by local governments and companies, user-submitted information, videos and photos taken while cycling, video data acquired from postal service vehicles, automobiles, airplanes or drones, videos and photos taken by tourists and employees, and 2D and 3D data such as architectural drawings of buildings and facilities.

[0036] The data is converted to a standardized format, undergoes quality checks (missing value detection, outlier detection, etc.), and can then be imported into the system. At this stage, data reliability can also be evaluated, and unreliable data can be appropriately weighted.

[0037] In step S3, the acquired data is stored in the storage unit 4 as time information. The data is saved at time intervals according to the type of situation, and aggregation can also be performed at the mesh level based on geographic coordinates.

[0038] In step S4, the analysis unit 5 calculates the risk level based on the accumulated data. The calculation method involves calculating the standard risk level, the time change coefficient, the spatial concentration level, etc., and the risk level can be derived by combining these.

[0039] The calculated risk level can be normalized to a numerical value and classified into stages such as normal, caution, alert, dangerous, and emergency. Based on this classification, subsequent actions can be determined.

[0040] In step S5, the virtual space display unit 3 updates the display of the virtual space based on the calculated risk level. The display method can include, for example, color display according to the risk level, changes in icon size, animation effects, 3D representation, etc.

[0041] The update can be performed at fixed intervals, but immediate update is also possible when a rapid increase in the risk level is detected. In addition, the display content can also be appropriately adjusted according to user operations (enlargement, reduction, viewpoint movement, etc.).

[0042] In step S6, it is evaluated whether the calculated risk level exceeds a set threshold. The threshold can be set based on the statistical distribution of risk levels in the normal state of the target area, fluctuation patterns by time zone, seasonal factors, special conditions, etc.

[0043] A basic value can be set for the threshold, and the threshold can be adjusted according to the situation. For example, the threshold can be adjusted in time zones where accidents have frequently occurred in the past to optimize the warning timing.

[0044] When the risk level exceeds the threshold, the process proceeds to step S7, and the notification unit 6 transmits a warning notification to a user or the like. The notification content can include the type and location information of the risk, the current risk level and predicted changes, recommended countermeasures, a link to detailed information, etc.

[0045] The notification can be customized based on the receiver's profile, and a priority can be set according to the importance. In addition, a function of controlling the frequency of notifications regarding the same event and preventing notification fatigue can be implemented.

[0046] After the notification processing in step S7 is completed, or when the risk level does not exceed the threshold, the system terminates its operation.

[0047] Furthermore, the system can also be configured as a continuous monitoring cycle. In this case, after the notification processing in step S7 is completed, or when the risk level does not exceed the threshold, the processing returns to step S2 of acquiring situation data for the next monitoring cycle, thereby enabling continuous monitoring of the situation in the real space and continuous evaluation of the risk level. The system can continue to operate until an explicit termination instruction is received from an external source such as a user operation or periodic maintenance.

[0048] Next, a detailed analysis processing flow in an embodiment of the present invention will be described with reference to FIG. 4. This processing can be executed when more advanced analysis is required. Furthermore, this detailed analysis processing can also be executed as part of step S4 in FIG. 3.

[0049] As shown in FIG. 4, the detailed analysis process starts from a start step, and the detailed analysis process is initialized in step S11. This process can be executed according to a regular schedule, or activated by a specific trigger (such as an increase in risk level, a request from a user, or the like).

[0050] In step S12, past situation data to be analyzed is read out from the storage unit 4. The read range can be determined based on a time range (a past period from the present), a spatial range (a target area and its surrounding area), a data type (a type of related situation data), and the like.

[0051] In the case of a large-scale dataset, sampling and filtering can be applied for efficient processing. However, special processing can be performed on important events and abnormal values to prevent them from being omitted due to sampling.

[0052] In step S13, comparative analysis is performed on current situation data and past data. As a comparison method, analysis by statistical test, similarity analysis of time-series patterns, application of an anomaly detection algorithm, periodicity analysis, or the like can be used. This makes it possible to evaluate how much the current situation deviates from past patterns. For example, this includes evaluation such as comparing a current occurrence rate with a past average value.

[0053] In step S14, a more detailed risk level can be calculated based on the comparison results with past data. Factors such as statistical deviation, past results of similar events, temporal proximity, and spatial diffusion prediction can be considered.

[0054] By combining multiple analytical models (statistical models, machine learning models, etc.), prediction accuracy can be improved. The outputs of each model can be weighted and integrated into a final risk assessment.

[0055] In step S15, the correlation between meteorological data (temperature, humidity, precipitation, wind speed, atmospheric pressure, etc.) and situational data can be analyzed. Correlation analysis can utilize methods such as linear correlation analysis, nonlinear correlation analysis, information-theoretic correlation measurement, and delayed correlation analysis considering time lag. This analysis can reveal relationships between meteorological conditions and situational data, such as the relationship between precipitation and accident rates. These findings can be used for future predictions using weather forecast data.

[0056] Step S16 allows for the analysis of the correlation between geographical features such as topography, land use, population density, and building distribution, and situational data. Geographic information system technology and spatial statistical methods can be utilized for this analysis.

[0057] By conducting hotspot analysis, spatial autocorrelation analysis, topographic factor analysis, and urban structure analysis, it is possible to identify correlation patterns with geographical features. This allows for the acquisition of insights such as the relationship between topographic conditions and accident risk.

[0058] These analysis results can be used to determine whether the risk level exceeds a threshold. If a risk level exceeding the threshold is detected, the process proceeds to step S17 and a warning notification is sent. If the risk level is below the threshold, the process proceeds to step S18, where the analysis results are visualized and the process ends.

[0059] Next, examples of applications of the present invention will be described.

[0060] Application Example 1: Application to a Traffic Accident Monitoring System One application example of the present invention is its application to a traffic accident monitoring and early warning system in urban areas.

[0061] Traffic accident situation data can be obtained from sources such as the police's real-time accident reporting system, video analysis from traffic surveillance cameras, vehicle telematics data, road sensors, and user reports. This data, along with attribute information such as location, time, accident type, number of vehicles involved, and whether or not there were injuries, can be collected and stored in a time-series database.

[0062] In calculating the risk level, factors such as accident frequency, accident severity, relationship with traffic volume, specific patterns, and correlation with weather conditions can be considered.

[0063] If the analysis reveals, for example, that the accident rate increases at a particular intersection during rainy evenings, then warnings can be sent to drivers passing through the area via navigation apps, notifications can be sent to traffic management centers, and warnings can be sent to nearby residents.

[0064] Furthermore, in the virtual space, the intersection in question can be highlighted with a color corresponding to its level of danger, and a warning icon can be displayed. Based on past accident patterns and weather forecasts, future danger levels can also be visualized.

[0065] Application Example 2: Application to Natural Disaster Monitoring Systems Another application example of the present invention is its use in monitoring natural disasters such as floods and landslides.

[0066] Data on natural disasters can be obtained from meteorological observation data, ground sensors, satellite and aerial photographs, historical disaster data, and detection of abnormal events from social media. By integrating this data and analyzing, in particular, the correlation between precipitation and river levels, and the relationship between topography and soil moisture content, disaster risk can be assessed.

[0067] In calculating the risk level, factors such as cumulative precipitation and precipitation intensity, the rate of rise in river water levels and the predicted maximum water level, soil moisture saturation and indicators of ground stability, past disaster occurrence patterns under similar conditions, and correlations with topographic characteristics can be considered.

[0068] If the analysis predicts, for example, that the risk of landslides in a certain area may increase within the next few hours, it becomes possible to provide evacuation information to residents in the affected area, provide risk information to local government disaster response headquarters, and provide information to emergency and rescue agencies.

[0069] In the virtual space, dangerous areas can be indicated using a tiered color-coding system, and information on evacuation shelters and routes can also be provided. Furthermore, it is possible to dynamically simulate and display changes in risk over time.

[0070] Application Example 3: Application to a Crime Occurrence Prediction System A third application example of the present invention is its application to the pattern analysis and prediction of crime occurrences in urban areas.

[0071] Crime-related situational data can be obtained from sources such as police crime statistics, security camera analysis results, patrol records, resident reports, and urban environment data. By integrating this data, it is possible to analyze the correlation between crime occurrence and environmental factors, as well as temporal and spatial patterns.

[0072] In calculating risk levels, factors such as the frequency and temporal distribution of each crime type, correlation with regional characteristics, influence of environmental factors, association with special events, and influence of seasonality and weather conditions can be considered.

[0073] If the analysis predicts, for example, an increase in the risk of a particular crime in a commercial area of ​​a district late at night on weekends, it may be possible to propose police patrol deployment plans, provide information to local crime prevention organizations, and issue warnings to users of the area.

[0074] In a virtual space, high-risk crime locations can be visualized by time of day, and filtering by crime type is also possible. Furthermore, changes in crime trends based on past data can be viewed over time.

[0075] Application Example 4: Application to anomaly monitoring systems for infrastructure equipment. A fourth application example of the present invention is its application to anomaly monitoring of critical infrastructure equipment such as power grids, water supply facilities, and communication infrastructure.

[0076] Status data regarding infrastructure equipment can be obtained from real-time data from various sensors (temperature, vibration, voltage, flow rate, etc.), equipment inspection records, failure history data, and malfunction reports from users. This data includes information such as location, time, measured values, equipment type, and operating status.

[0077] In calculating the risk level, factors such as the degree of deviation of sensor values ​​from the normal range, the rate of change of values, changes in the correlation between multiple sensors, the years of service of the equipment and past failure patterns, and the relationship with weather conditions and usage load can be considered.

[0078] If the analysis reveals, for example, that the insulation resistance of a specific substation is gradually decreasing and that the risk of failure is expected to increase within the next few days, it may be possible to issue an emergency inspection order to the maintenance team, prepare for the supply of power to an alternative route, and provide advance notice to potentially affected businesses and facilities.

[0079] In a virtual environment, the entire infrastructure network can be visualized, and high-risk equipment and sections can be color-coded. Furthermore, it's possible to simulate the scope and ripple effects of a failure and develop preventative measures.

[0080] In particular, by analyzing the interdependencies of multiple different infrastructures (electricity, water, telecommunications, etc.), it becomes possible to detect complex risks, such as "the possibility that a malfunction in power equipment could spread to the telecommunications infrastructure, thereby affecting the water management system."

[0081] This system will enable a shift from conventional periodic inspections and reactive maintenance to predictive preventive maintenance, which is expected to contribute to the stable operation of critical infrastructure. Furthermore, it will allow for the development of optimal maintenance plans based on the condition of the equipment, leading to the optimization of maintenance costs.

[0082] Application Example 5: Application to a User-Participatory Data Collection System Another application example of the present invention is a system that incorporates user-participatory data collection and a reward mechanism.

[0083] This system allows ordinary users to take photos of real-world situations with their smartphones and provide the situation data through an app, in exchange for rewards such as cryptocurrency and proprietary tokens. The captured data is reviewed by AI and then reflected on a virtual map.

[0084] In terms of data provision methods, it is possible to provide not only spot photography but also route driving videos similar to those from a dashcam. Furthermore, it is possible to attach a small camera to support hands-free recording.

[0085] For example, the following methods can be used to award rewards.

[0086] - Rewards for responding to location requests from other users - Variable rewards based on data quality and usefulness - Special rewards for completing regular missions and events - Bonuses for providing data covering a wide area - Competitive incentives based on rankings Through these user-participatory mechanisms, it becomes possible to efficiently collect wide-ranging and up-to-date situational data. In particular, it is effective in understanding detailed situations that cannot be fully covered by public institutions and companies alone, as well as rapidly changing events. Furthermore, by analyzing the data accumulated by this system, it is possible to help in the early detection of infrastructure malfunctions and anomalies.

[0087] By allowing users to report minor anomalies that are often overlooked during regular inspections, as well as incidents that occur immediately after they happen, infrastructure managers can respond more quickly. For example, early detection of road cracks, damaged signs, and malfunctions in public facilities is expected to prevent major accidents and damage.

[0088] Furthermore, to enhance user engagement, a system can be implemented that allows users to utilize earned rewards in various ways. Examples include unlocking special features within the app, customizing avatars and homes on the map, and displaying a status that indicates the user's contribution as a data provider.

[0089] Furthermore, since it updates in real time, in the future it will be possible to display things like friends standing in tourist spots or waving, and real-time updates of traffic congestion, disaster situations, and weather information will also allow users to understand the situation.

[0090] Furthermore, the platform allows for realistic simulations of all data, utilizing algorithms that automatically match, compare, and synthesize the metadata of uploaded videos (GPS, gyro sensor data, time) with data from multiple other users and existing GIS data. It can also incorporate a system-wide "verification process" to eliminate "fake information" and ensure the data has official evidentiary value as survey data.

[0091] Without using surveying equipment, it is possible to use a correction calculation method to ensure "surveying accuracy" using only consumer-grade smartphones. This provides a platform that displays not only the "current state" but also "changes from the past" in chronological order, and allows for the calculation of these changes. Furthermore, in order to improve accuracy, other devices may be used in addition to consumer-grade smartphones.

[0092] Furthermore, it can be an infrastructure monitoring system characterized by the ability to complete planning and calculations such as safety management and situation simulations on the metaverse without going to the site, and to utilize the simulation results in actual work, as well as automatic correction of spatial distortion based on the photographer's device characteristics and movement path, and the assurance of the non-tampering of photographic information and accuracy as survey data.

[0093] Furthermore, the AI ​​could determine whether the posted location is critical infrastructure (such as bridges or main roads) or if it is "highly urgent (severely damaged)," and the smart contract could dynamically determine the token reward amount based on the urgency and importance score, as well as other data, not limited to infrastructure, according to its importance.

[0094] Furthermore, the illustrated embodiments are merely illustrative examples and are not intended to limit the technical scope of the present invention.

[0095] 1 Information provision system 2 Status data acquisition unit 3 Virtual space display unit 4 Storage unit 5 Analysis unit 6 Notification unit 7 Control unit 10 User terminal 11 Information processing system S1-S7 Processing steps S11-S18 Detailed analysis processing steps

Claims

1. An information provision system comprising: a situation data acquisition unit that acquires situation data obtained by photographing a situation in real space; and a virtual space display unit that, based on the situation data, displays information related to the situation at coordinates in the virtual space corresponding to the location in the real space where the situation occurred, wherein the situation data includes information relating to at least one of traffic accidents, crimes, infrastructure malfunctions, and natural disasters.

2. The information provision system according to claim 1, characterized in that the situation data includes time information relating to the time when the situation occurred, and the data is accumulated in a time series.

3. The information provision system according to claim 1 or 2, comprising an analysis unit that calculates the degree of risk in the real space based on the situation data, wherein the virtual space display unit displays the degree of risk in the virtual space.

4. The information provision system according to claim 3, characterized in that the analysis unit calculates the risk level for each time period based on the situation data.

5. The information provision system according to claim 3, characterized in that the analysis unit calculates the degree of risk in the real space by comparing accumulated past situation data with current situation data.

6. The information provision system according to claim 3, wherein the situation data includes meteorological information at the location where the situation occurred, and the analysis unit analyzes the correlation between the meteorological information and the degree of risk.

7. The information provision system according to claim 3, characterized in that the analysis unit analyzes the correlation between geographical characteristics and the risk level and predicts the risk level of other areas having similar geographical characteristics.

8. The information provision system according to claim 1 or 2, characterized in that the situation data acquisition unit acquires video data acquired from a camera mounted on a bicycle, commercial vehicle, delivery vehicle, airplane or drone, and 2D data and 3D data including video data, image data, and design drawing data taken by the user during business or sightseeing.

Images