KR20130107424A - Method, system and control system for detecting and forecasting fog based on ubiquitous sensor network and semantic technology - Google Patents

Method, system and control system for detecting and forecasting fog based on ubiquitous sensor network and semantic technology Download PDF

Info

Publication number
KR20130107424A
KR20130107424A KR1020120029174A KR20120029174A KR20130107424A KR 20130107424 A KR20130107424 A KR 20130107424A KR 1020120029174 A KR1020120029174 A KR 1020120029174A KR 20120029174 A KR20120029174 A KR 20120029174A KR 20130107424 A KR20130107424 A KR 20130107424A
Authority
KR
South Korea
Prior art keywords
fog
weather
situation
information
detection
Prior art date
Application number
KR1020120029174A
Other languages
Korean (ko)
Inventor
이상훈
Original Assignee
한국건설기술연구원
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 한국건설기술연구원 filed Critical 한국건설기술연구원
Priority to KR1020120029174A priority Critical patent/KR20130107424A/en
Publication of KR20130107424A publication Critical patent/KR20130107424A/en

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01W—METEOROLOGY
    • G01W1/00—Meteorology
    • G01W1/02—Instruments for indicating weather conditions by measuring two or more variables, e.g. humidity, pressure, temperature, cloud cover or wind speed
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/90—Details of database functions independent of the retrieved data types
    • G06F16/95—Retrieval from the web
    • G06F16/951—Indexing; Web crawling techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/10—Services
    • G06Q50/26—Government or public services
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W88/00—Devices specially adapted for wireless communication networks, e.g. terminals, base stations or access point devices
    • H04W88/02—Terminal devices

Abstract

PURPOSE: A real time fog situation sensing and predicting method, a system thereof, and a control system thereof are provided to prevent an accident caused by fog. CONSTITUTION: Weather data, composed of weather-related data including fog sensing and prediction target areas, a visible distance of the target areas, and weather information collected from an AWS in the target areas, is stored and collected in a weather server at a collection step. A fog state is predicted by classifying the state into a low risk level to a high risk level by inputting cloud existence, rainfall, wind strength, relative humidity, and cloud cover into fog sensing algorithm data sources at a fog state prediction step. A current weather state and a predicted fog state in the fog sensing and prediction target areas are transmitted to a portable terminal at a transmission step. [Reference numerals] (AA) Whether fog exists in the area or not (whether or not the weather center notify that it is cloudy); (BB) Rainfall trace (check whether it rains or snows before and after the present); (CC) Wind strength (7m/s); (DD) Relative humidity = 90%; (EE) Amount of clouds, whether the amount of clouds is small or not (there is some information about AWS); (FF) Relative humidity = 95%; (GG) Check visibility and whether dew-point temperature is included or not (when it is morning or midday)

Description

Method, System and Control System for Detecting and Forecasting Fog in Real Time {Method, System and Control System for Detecting and Forecasting Fog based on Ubiquitous Sensor Network and Semantic Technology}

The present invention relates to a real-time fog situation detection and prediction method, a system and a control system using the same, and more specifically, through the historical data obtained in real time and accident history data from past fog extracted from social network services or news, etc. The present invention relates to a real-time fog situation detection and prediction method, system, and a control system using the same, which can prevent fog accidents by detecting a fog situation in real time and predicting the possibility of fog occurrence.

Fog occurs under a variety of conditions, and there are limits to modeling or detecting fog using specific numerical models. It is also true that the removal or reduction of fog is limited after recognizing fog. At present, research on the recognition and reduction of fog in Korea is based on only sensor of municipal government developed outside of Korea as an early stage, and it is not able to systematically cope with prediction by using prediction. In particular, the disaster prevention manual for fog is not available, and it is not linked automatically with Weather Information System (WIS), so it relies on passive management.

The fog recognition technology includes a method of observing fog using a visibility system, a method of predicting the fog generation potential using an automatic weather station (AWS) and a weather information system, a method of using the visibility information obtained through a camera There is a method to detect fog using GPS, radiosonde, and a method of predicting the fog potential using atmospheric water vapor, using the method, MODIS and GOES-9 satellite observations. Among the various fog recognition methods, a method of observing the current correction distance using a visibility system or a camera is popularized, but it relies on foreign sensor and program.

As a background technology of the present invention, Patent Registration No. 0946749 "Fog detection method and system based on image recognition and image learning method" (Patent Document 1).

In the background art, an image camera, an image input unit, a fog information analyzer as shown in FIG. 8; And a database for analyzing the acquired image data to extract fog generation status information of the road, the fog detection method comprising the steps of: (a) Acquiring image data of environmental information and fog generation information; (b) the image data is classified and imaged by the image input device, and the image analysis means calculates the number of frames and the image state and performs a numerical calculation on the image data, Generating fog occurrence status information for each region by averaging the fog situation information for each region of the fog information, and transmitting the fog occurrence status information to the fog information analyzer and the database, respectively; (c) generating fog generation status information of the entire road including the connection section between the regions, the fog generation status information being transmitted to the fog information analyzer, being statistically processed through the fog situation analysis means; And (d) fog occurrence status information for the region and fog occurrence status information for the entire road, which are the data calculated by the fog situation analyzing means, from the fog information providing means to the text information, the audio information, the geographical information and the combined image information step; And (e) any or all of the text information, the audio information, the geographical information, and the combined image information corresponding to the fog generation status information for the region and the fog generation status information for the road region is input to one of the CDMA system and the WiBro system And transmitting the data to the image input device, the database, or the client wirelessly through a data communication means.

However, in the background art, since the fog occurrence state information of the road is extracted by analyzing the image data acquired by the image camera, it is difficult to comprehensively judge the weather data such as the relative humidity and wind and the accuracy is remarkably decreased due to the narrow fog There was a problem.

Patent registration No. 0946749 "Fog detection method and system based on image recognition and image learning method"

The present invention is to solve the above problems, by collecting meteorological data obtained through the Meteorological Agency homepage, AWS and visibility system and fog accident history data extracted from SNS and news, etc. in real time to detect the fog situation and the possibility of fog occurrence The purpose is to provide a real-time fog situation detection and prediction method that can prevent the accident caused by fog by predicting and transmitting it to the manager or driver of a road, bridge or tunnel.

The real-time fog situation detection and prediction method according to a preferred embodiment of the present invention, the weather-related data of the meteorological office of the wide area including the fog detection and prediction target area, the fog distance collected through the visibility system and the visibility distance and prediction of the prediction target area A collection step of storing and collecting weather data including weather information collected by AWS installed in a target region in a weather server; In order to predict the fog situation using the weather data collected by the weather server, input the cloud presence, rainfall trail, wind intensity, relative humidity and cloudiness, A fog situation predicting step for predicting the fog situation in low risk to high risk stage; And transmitting a current weather situation and a predicted fog situation of the region to be detected and predicted to a fog to a manager or a portable terminal for a general user.

The collecting step may further collect the past history data related to the fog situation including the fog situation and fog information expressed in the SNS and news including Facebook, Twitter, blog for the target area of fog detection and prediction. It features.

The collection of the historical data 20 is characterized in that to collect information by using a Web crawler program (Web Crawler) program.

The fog situation prediction step is characterized by predicting the fog situation by building the final fog situation data consisting of time, place, fog situation by combining the fog situation prediction information and historical data predicted by the weather data.

In the transmitting step, the fog situation current information and fog situation prediction information is provided through a fog situation information search query of the weather server storage.

In the transmission step, the location information is displayed as a geographic information system-based region, and the fog situation prediction information is expressed in the form of a contour line to transmit a difference in risk.

The weather information collected by the AWS is characterized by consisting of weather information of temperature, wind, humidity, air pressure.

The fog situation prediction step is a cloud presence determination step of determining the presence of clouds based on the weather data, if there is no cloud if there is no risk of fog generation; Determining whether there is a cloud in the step of determining whether or not the cloud is present, determining whether the cloud is present or not, and determining that there is no risk of fog if there is no rain; If there is a trace of rainfall in the rainfall trail determination step, the intensity of the determined wind is checked. If the rainfall is above the reference value, the wind strength determination step is determined as the first step without the risk of fog generation. Determining a relative humidity of the wind when the wind intensity does not exceed the reference value in the wind intensity determination step; A first step of determining a degree of cloudiness if the relative humidity is higher than a reference value in the first relative humidity determination step and a second step of determining a risk of fog occurrence in the case of low cloudiness; A second relative humidity determination step of determining a higher reference value than the first relative humidity determination step by checking the relative humidity if the amount is a fine amount in the cloud checking step, and a third step of determining a low risk of mist generation when the reference value is less than the reference value; A third relative humidity determination step of determining a higher reference value than the second relative humidity determination step if the relative humidity is higher than a reference value in the second relative humidity determination step, and a fourth occurrence of a mist generation risk when the relative humidity is lower than the reference value; And a fourth step of determining relative humidity in the third step of determining relative humidity when the relative humidity is higher than a reference value, which is determined as a fourth step of high risk of fog generation.

The real-time fog situation detection and prediction method according to another suitable embodiment of the present invention, the weather-related data of the meteorological office of the region including the fog detection and prediction target region, the visibility distance of the fog detection and prediction target region collected through the visibility system and A collection step of collecting, by a weather server, weather data including weather information collected from AWS installed in a predicted region; In order to predict the fog situation using the weather data collected by the weather server, input the cloud presence, rainfall trail, wind intensity, relative humidity and cloudiness, A fog situation predicting step for predicting the fog situation in low risk to high risk stage; A transmission step of transmitting a current weather situation and a predicted fog situation of a fog detection and prediction target region to a portable terminal of an administrator; And instructing the portable terminal to perform a predetermined action on the control unit of the road facility of the fog detection and prediction target area when the information received by the administrator after the transmitting step is satisfied with the fog generation. do.

The road facility is characterized in that the fog net.

The road facility control system using a real-time fog situation detection and prediction method according to an embodiment of the present invention comprises a meteorological office-related data and weather information collected from the fog detection collection means of a wide area including a fog detection and prediction target area. Storage (S1) and storage (S1) for receiving and storing historical data related to the fog situation, including weather data and Facebook and Twitter, blogs and fog information represented in the news and accidents caused by the fog; A weather server (S) comprising data communication means (S2) for collecting and processing weather data and historical data stored therein and transmitting in real time through wireless communication; A portable terminal that receives weather data and historical data collected by the weather server in real time through wireless communication; The road facility 40 is provided with a control unit 41 for wireless communication with the portable terminal, characterized in that it is possible to control the road facilities in real time through the portable terminal.

The fog information collecting means is a visibility and AWS installed in the area to detect and predict fog.

The real-time fog situation detection and prediction system according to a preferred embodiment of the present invention, the weather-related data of the regional meteorological office including the fog detection and prediction target region, the fog distance collected through the visibility system and the visibility distance and prediction of the prediction target region It provides an interface to the weather data, which consists of weather information collected from AWS installed in the target area, and the past history data related to the fog situation including the fog situation and fog accident information expressed in SNS and news including Facebook, Twitter, and blog. Fog situation semantic storage to convert the ontology model through the storage; A semantic reasoner for inference of each ontology model of weather data and historical data; And an ontology model combiner for combining ontology models of weather data and historical data, respectively.

The real-time fog situation detection and prediction method of the present invention collects weather data acquired through the Meteorological Agency homepage, AWS and visibility system and fog accident history data extracted from SNS and news to detect the fog situation in real time and predict the possibility of fog occurrence Sending them to managers or drivers of roads, bridges or tunnels has a very useful effect of preventing accidents from fog.

BRIEF DESCRIPTION OF THE DRAWINGS The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate exemplary embodiments of the invention and, together with the description, serve to explain the principles of the invention, Shall not be construed as limiting.
BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of a real-time mist detection and prediction method according to the present invention.
2 is a diagram illustrating a decision tree model for predicting a fog situation according to the present invention.
3 is an example of a national weather information screen output to the portable terminal 30 in conjunction with the weather server (S).
4 is an example of a detailed weather information screen output to the portable terminal 30 in conjunction with the weather server (S).
5 is an example of a fog prediction information screen output to the portable terminal 30 in conjunction with the weather server (S).
6 is a diagram illustrating a road facility control system using a real-time mist detection and prediction method.
7 is a diagram illustrating a user interface of a portable terminal of a manager of a road facility control system using the real-time mist detection and prediction method of the present invention.
8 is a block diagram illustrating an image camera and an image input apparatus according to a conventional embodiment.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described in detail below with reference to the embodiments shown in the accompanying drawings, but the present invention is not limited thereto.

 Hereinafter, the technical structure of the present invention will be described in detail with reference to the preferred embodiments.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of a real-time mist detection and prediction method according to the present invention.

The real-time fog situation detection and prediction method of the present invention is a meteorological office weather-related data of the wide area including the fog detection and prediction target area, the visibility distance of the fog detection and prediction target region collected through the visibility system and the fog detection and prediction target area A collection step of collecting weather data collected by the installed AWS from a weather server; In order to predict the fog situation using the weather data collected by the weather server, input the cloud presence, rainfall trail, wind intensity, relative humidity and cloudiness, A fog situation predicting step for predicting the fog situation in low risk to high risk stage; And a transmission step of transmitting the current weather condition and the predicted fog situation of the fog detection and forecasting area to the manager or the general user portable terminal.

The collecting step is a step of collecting weather meteorological data, visibility distance, and meteorological data 10 collected from AWS for weather detection and forecasting areas in the weather server S.

The meteorological data 10 collected by the weather server S includes the weather situation information of the regional meteorological office including the fog detection and prediction area, the visibility distance obtained through the visibility installed in the fog detection and the prediction area, and the fog detection. And meteorological information such as temperature, wind, humidity, and air pressure collected through an Atomatic Weather System (AWS) installed in a predicted region.

For example, if the Yeongjong Bridge becomes an area to detect and predict fog, the weather forecast information of the Korea Meteorological Agency for Incheon, which includes the Yeongjong Bridge, the visibility distance obtained through the visibility system installed on the Yeongjong Bridge, and the AWS installed on the Yeongjong Bridge The collected weather information constitutes weather data 10 stored in the weather server S.

Atomatic Weather System (AWS) is a device that automatically observes the fog detection and the atmospheric conditions of a predicted area and transmits it in a specialized form according to a predetermined communication method or stores it in a storage medium.

The weather data 10 collected as described above may be represented, stored, and utilized as a contextual information model used by the weather server S. The contextual information model is a simple data model: a key-value model, a logic-based model, a mark-up schema model, an object-oriented model, Ontology) model. However, most of the context information model is too simple to express the various structures and relationships of context information, and complex programming is required for functions such as context information abstraction. Therefore, an ontology model with high expressiveness and flexibility is most desirable for the modeling of context information that needs to express information of various objects that are not dependent on a specific domain.

Ontology can be thought of as a kind of dictionary consisting of formal and explicit specifications, words and relations for shared conceptualization, in which words related to a particular domain are hierarchically And it is possible to extend knowledge-sharing between knowledge-based applications and application programs, and to reuse them. Ontology is the most central concept of semantic web application. Ontology language is a language that defines schema and syntax structure to represent it. Currently, there are DSML + OIL, OWL, Ontolingun and so on.

In the fog situation prediction step, in order to predict the fog situation through the weather data (10) collected on the weather server (S), the presence of clouds in the fog detection area and the predicted region, rainfall traces, wind intensity, relative humidity and cloud cover, and low cloud volume The decision is made to input the real-time fog detection algorithm as a data source and to classify the current fog situation into low to high risk stages.

That is, the weather data 10 stored in the weather server S is fused to predict the current fog situation step by step by the decision tree.

2 is a diagram showing a decision tree model for predicting a fog situation.

As shown in FIG. 2, based on a decision tree, a low-risk to high-risk five-stage prediction is predicted, and a binary tree is formed for each variable.

First, the presence or absence of clouds is determined through the weather forecast (weather, cloudy, rain, etc.) of the weather station. If there is no cloud, it is judged as No Risk (No Risk) that there is no danger of fog generation. If clouds are present, determine the rainfall trail through the weather status of the Korea Meteorological Administration (current rainfall (rain / snow)). If there is no trace of rainfall, it is judged as No Risk (No Risk).

Since there is a possibility of occurrence of fog depending on the amount of water vapor, the amount of water vapor in the present air is predicted on the basis of the presence or absence of the cloud and the rainfall trail, and the fog situation is classified stepwise by other variables depending on the amount of water vapor.

If there is a trace of rain, check the intensity of the wind through the AWS installed in the fog detection and prediction area.

If the intensity of the wind exceeds 7 m / s, the risk of fogging is small. Therefore, it is judged as No Risk that there is no danger of fogging. If the wind intensity does not exceed 7 m / s, , The relative humidity is checked through the AWS installed in the fog detection and prediction area.

If the relative humidity is less than 90%, it is determined as No Risk, and if the relative humidity is more than 90%, the probability of fogging is high. Therefore, the AWS and the weather information Check the amount of clouds (cloud amount).

In the low cloud area, there is a cloud. Risk not estimated, which is a risk that can not be measured by integrating the values of accumulated variables with rainfall traces, wind intensity exceeding 7m / s and relative humidity of more than 90% ).

However, if there is a lot of cloudiness, it is judged by comparing the cumulative value of the variable with the relative humidity. Since the relative humidity is determined based on the relative humidity of 90% in the past, the relative humidity is slightly increased to predict the possibility of fog generation.

If the relative humidity is less than 95% based on the relative humidity of 95%, it is judged as a low-level risk, which is low risk of mist generation.

When the relative humidity exceeds 95%, it is determined based on the relative humidity of 97%. If the relative humidity is less than 97%, it is determined as the medium-level risk, which is the risk of fog generation. If the relative humidity is more than 97%, it is judged to be a high-level risk of fogging.

As described above, the fog situation prediction information is calculated by analyzing the weather information of the corresponding branch of the weather station and the collected data of AWS.

In addition, in the collecting step, in addition to collecting the weather data 10, the past history data 20 related to the fog situation, such as fog information and accident information caused by the fog represented on Facebook, Twitter, blog, news, etc. Can be collected.

The collection of the historical data 20 collects fog information and accidents caused by fog expressed on Facebook, Twitter, blog, news, etc., using a web crawler program. Fog situations and related accident occurrence information such as 'Jincheon', 'Haze', 'Okcheon Intersection' and 'Four-Central Stone' are made and collected through keyword search.

Then, various vocabularies expressing fog as synonyms are stored and managed, and information gathered in predefined data dictionary and defined history ontology model (position-time-road information-fog information-accident information) To create an instance.

Based on the historical data 20 collected as described above in the fog situation prediction step, the fog situation prediction information predicted by the weather data 10 and the historical data 20 are combined to form a final time, place, fog situation, etc. Fog situation data can be built to predict fog conditions.

Create a meteorological ontology model of the meteorological data 10 and the actual data values of the fog prediction algorithm, create an instance according to the meteorological ontology model, and automatically combine the collected historical ontology through the RDF / OWL inference machine Situational contextual history is stored in the form of semantics, which predicts the possibility of accidents caused by fog.

In the transmission step, the weather situation and the predicted fog situation of the fog detection and prediction target region are transmitted to the manager or the portable terminal 30 for the general user.

The portable terminal 30 may be a variety of portable terminals such as a PDA, a smart phone, a smart pad, a UMPC, and the like, and an administrator and a general user connect to receive various data.

The portable terminal 30 expresses national weather information, detailed weather information, fog prediction information, and fog occurrence analysis results.

3 is an example of the national weather information screen output to the portable terminal 30 in conjunction with the weather server (S), and outputs the weather information of the National Weather Service on the screen of the portable terminal 30 to output the weather information of the whole country.

4 is an example of a detailed weather information screen output to the portable terminal 30 in conjunction with the weather server (S).

Field name Korean name Field name Korean name kma_id Unique ID 온도 Current temperature stn_id Branch ID dewpoint Dew point temperature stn_name Branch name s_temperature Feeling temperature date Observation date and time rainfall_a Daily precipitation (mm) status Current diary humidity Humidity(%) visibility Visibility (km) wind_direction Wind direction a_cloud Cloudiness (1/10) wind_speed Wind speed (m / sec) b_cloud Heavy Load ap_sealevel Sea surface pressure

Table 1 shows an example of the meteorological station weather information table.

As shown in FIG. 4, detailed weather information is output by outputting data collected by the Korea Meteorological Administration on the screen of the portable terminal 30 based on the table of Table 1.

The Meteorological Agency stores the weather conditions at intervals of 1 hour, exchanges data, and outputs detailed weather information by outputting the weather server to the screen of the portable terminal 30.

5 is an example of a fog prediction information screen output to the portable terminal 30 in conjunction with the weather server (S).

Field name Korean name Field name Korean name aws_id Unique ID visibility Visibility stn_id Branch ID rainfall_flag Precipitation time_n Observation time rainfall_amount Precipitation wind_direction Wind direction status Weather wind_speed Wind velocity ceiling Uncle 온도 Temperatures am_cloud Cloudiness humidity Relative humidity fog_factor Fog Index a_pressure Atmospheric pressure

Table 2 is an example of the weather information table collected in AWS.

As shown in FIG. 5, the possibility of fog generation is analyzed based on the weather information collected by the Korea Meteorological Administration and AWS, and the analysis results are displayed on a chart.

The manager uses the fog information received through the portable terminal 30 as condition information for starting any operation, and the system continuously monitors whether the current condition information satisfies a specific condition, and in advance when the condition is satisfied. You can take a specified service or action. For example, when weather conditions are collected through weather data 10 collected on the weather server S, the fog net can be operated when the fog condition is established. You can inform.

In the transmission step, the fog situation information search query of the weather server (S) storage, the fog situation information extracted through the current weather data (10) and the short-term future fog prediction situation extracted through the historical data (20) Provided through the user's portable terminal 30, the location information can be displayed in the geographic information system-based area, and the short-term future fog can be expressed in the form of contours so that the user can clearly recognize the difference in risk. .

6 is a diagram illustrating a road facility control system using a real-time mist detection and prediction method.

The road facility control system using the time fog situation detection and prediction method of the present invention comprises a weather server (S) consisting of a storage (S1) and data communication means (S2), a portable terminal (30) and a road facility (40). .

The storage (S1) stores meteorological data consisting of meteorological data collected from meteorological stations in the metropolitan meteorological field including the fog detection and forecasting areas and weather information collected from the fog detecting means, and SNS including Facebook, Twitter, It is a place to receive and store past history data related to fog situation such as accident information due to situation and mist.

The information collecting means may be a visibility system and an AWS installed in the mist detection and forecasting area, and collect the weather information through the system.

The data communication means (S2) collects and processes the weather data and the history data stored in the storage (S1) and transmits them in real time through wireless communication.

The mobile terminal 30 receives the weather data and the historical data collected by the weather server S in real time through wireless communication, and whether the current fog is a traffic accident risk type or the future fog situation prediction through wireless communication. Based on the received data, the user directly accesses the control unit 41 that can control the road facility 40 through the portable terminal 30 to directly control road facilities such as a fog net.

7 is a diagram illustrating a user interface of a portable terminal of a manager of a road facility control system using the real-time mist detection and prediction method of the present invention.

As shown in Fig. 7, it is possible to operate a hot line of a mist net (a small net), or to directly operate the up / down descending net to block or reduce the inflow of mist.

The real-time fog situation detection and prediction system of the present invention is a fog situation semantic storage (ST) and weather data and history data to convert the weather data 10 and past history data 20 into an ontology model through an interface The semantic inference for inference of ontology model is composed of the ontology model combiner for combining each ontology model of weather data and historical data.

Meteorological data consisting of meteorological data from the regional meteorological offices including fog detection and forecasting areas, fog collected and collected through visibility systems, and weather information collected from AWS installed in the forecasting areas. And converts into ontology models through the interface of the past history data (20) related to the fog situation, such as the fog situation and the accident information caused by the fog, and the social media including Facebook, Twitter, and blog, and the weather data and history. Enter data information to build a fog-aware semantic repository.

When the context aware semantic repository is constructed, it is transformed to fit the ontology model by analyzing the meteorological data, transformed into the historical data ontology model collected by SNS, etc., and the two semantic repositories are linked through the adjustment of the weather data ontology and the history data ontology model Build a final fog situation awareness service semantic repository (ST) composed of "time-site-fog situation".

In the combination of two different ontology models, the semantic inference and ontology model combiner are combined according to pre-defined attributes and axiom rules.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art in light of the above teachings. will be. The invention is not limited by these variations and modifications, but is limited only by the claims appended hereto.

S: Weather Server
ST: Semantic repository
10: sensor data
20: history data
30: portable terminal

Claims (13)

Weather server including weather data collected from the meteorological office of the region including fog detection and forecasting area, fog detection and visibility distance collected through visibility system, and weather information collected from AWS installed in the forecasting area. A collection step of storing in the collection;
In order to predict the fog situation by using weather data collected on the weather server, the fog detection and the presence of cloud at the target area, rainfall trace, wind intensity, relative humidity and cloud volume, and low cloud volume are input as real-time fog detection algorithm data source. A fog situation prediction step of predicting the fog situation by dividing the stage from low risk to high risk;
And a transmission step of transmitting a current weather situation and a predicted fog situation of a region to be detected and predicted by a fog to a manager or a portable terminal for a general user. The method of claim 1,
The collecting step may further collect the past history data related to the fog situation including the fog situation and fog information expressed in the SNS and news including Facebook, Twitter, blog for the target area of fog detection and prediction. A real-time fog situation detection and prediction method. The method of claim 2,
Collection of historical data (20) is a real-time fog situation detection and prediction method, characterized in that collecting information using a Web Crawler (Web Crawler) program. The method of claim 2,
The fog situation prediction step is a real-time fog situation detection, characterized in that by constructing the final fog situation data consisting of time, place, fog conditions by combining the fog situation prediction information predicted by the weather data and the historical data And prediction method. The method according to any one of claims 2 to 4,
A real-time fog situation detection and prediction method, characterized in that to provide the fog information current information and fog situation prediction information through the query of the fog situation information search of the weather server storage in the transmission step. 6. The method of claim 5,
Real-time fog situation detection and prediction method characterized in that the position information is displayed as a geographic information system-based region in the transmission step, and the fog situation prediction information is transmitted in the form of contours to transmit the difference in risk. The method of claim 1,
The weather information collected from AWS consists of weather information of temperature, wind, humidity, barometric pressure real-time fog situation detection and prediction method. The method according to any one of claims 1 to 4,
The fog situation prediction step is a cloud existence determination step of determining whether there is no cloud based on the weather data to determine whether there is no risk of fog if there is no cloud;
Determining whether there is a cloud in the step of determining whether or not the cloud is present, determining whether the cloud is present or not, and determining that there is no risk of fog if there is no rain;
If there is a trace of rainfall in the rainfall trail determination step, the intensity of the determined wind is checked. If the rainfall is above the reference value, the wind strength determination step is determined as the first step without the risk of fog generation.
Determining a relative humidity of the wind when the wind intensity does not exceed the reference value in the wind intensity determination step;
A first step of determining a degree of cloudiness if the relative humidity is higher than a reference value in the first relative humidity determination step and a second step of determining a risk of fog occurrence in the case of low cloudiness;
A second relative humidity determination step of determining a higher reference value than the first relative humidity determination step by checking the relative humidity if the amount is a fine amount in the cloud checking step, and a third step of determining a low risk of mist generation when the reference value is less than the reference value;
A third relative humidity determination step of determining a higher reference value than the second relative humidity determination step if the relative humidity is equal to or higher than a reference value in the second relative humidity determination step and a fourth occurrence of a mist generation risk when the relative humidity is lower than the reference value; And
In the third relative humidity determination step, if the relative humidity is higher than the reference value, the fourth relative humidity determination step of judging the fourth stage of high risk of fog generation; real-time fog situation detection and prediction method comprising the. Weather server including weather data collected from the meteorological office of the region including fog detection and forecasting area, fog detection and visibility distance collected through visibility system, and weather information collected from AWS installed in the forecasting area. Collecting step of collecting;
In order to predict the fog situation by using weather data collected on the weather server, the fog detection and the presence of cloud at the target area, rainfall trace, wind intensity, relative humidity and cloud volume, and low cloud volume are input as real-time fog detection algorithm data source. A fog situation prediction step of predicting the fog situation by dividing the stage from low risk to high risk;
A transmission step of transmitting a current weather situation and a predicted fog situation of a fog detection and prediction target region to a portable terminal of an administrator;
After the step of transmitting the information received by the administrator to the portable terminal is a condition that the fog is generated, instructing the portable terminal to perform a predetermined action to the control unit of the road facilities of the fog detection and prediction target area; Detection and prediction method. The method of claim 9,
Real-time fog situation detection and prediction method characterized in that the road facility is a fog net. It is composed of weather data composed of weather-related data collected from the Meteorological Agency in the region including fog detection and forecast areas, and weather information collected from the fog detection means, and fog situations and fog expressed in SNS and news including Facebook, Twitter, and blog. Storage (S1) for receiving and storing past history data related to the fog situation including the accident information, and data communication means for collecting and processing weather data and history data stored in the storage (S1) through wireless communication in real time ( A weather server S composed of S2);
A portable terminal that receives weather data and historical data collected by the weather server in real time through wireless communication;
Road facility 40 is provided with a control unit 41 for wireless communication with the portable terminal; road using a real-time fog situation detection and prediction method, characterized in that to control the road facility in real time through the portable terminal Facility control system. 12. The method of claim 11,
The fog information collecting means is a road facility control system using a real-time fog situation detection and prediction method, characterized in that the visibility and AWS installed in the area to detect and predict fog. Weather data and Facebook consisting of meteorological data from the regional meteorological offices including fog detection and forecasting areas, fog collected and collected through visibility systems, and weather information collected from AWS installed in the forecast area Fog status semantic storage for converting and storing historical data related to the fog situation including the fog situation and the fog information represented on the SNS and news including Twitter and blogs into an ontology model through an interface;
A semantic reasoner for inference of each ontology model of weather data and historical data; And
Ontology model combiner for combining each ontology model of weather data and historical data; real-time fog situation detection and prediction system consisting of.
KR1020120029174A 2012-03-22 2012-03-22 Method, system and control system for detecting and forecasting fog based on ubiquitous sensor network and semantic technology KR20130107424A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1020120029174A KR20130107424A (en) 2012-03-22 2012-03-22 Method, system and control system for detecting and forecasting fog based on ubiquitous sensor network and semantic technology

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1020120029174A KR20130107424A (en) 2012-03-22 2012-03-22 Method, system and control system for detecting and forecasting fog based on ubiquitous sensor network and semantic technology

Publications (1)

Publication Number Publication Date
KR20130107424A true KR20130107424A (en) 2013-10-02

Family

ID=49630730

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020120029174A KR20130107424A (en) 2012-03-22 2012-03-22 Method, system and control system for detecting and forecasting fog based on ubiquitous sensor network and semantic technology

Country Status (1)

Country Link
KR (1) KR20130107424A (en)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20150134810A (en) * 2014-05-23 2015-12-02 주식회사 솔트룩스 Social sensor system for sensor web
KR20160029397A (en) * 2014-09-05 2016-03-15 경희대학교 산학협력단 Method and apparatus for value added service based on evaluation of the sights
KR20180055205A (en) * 2016-11-16 2018-05-25 이승재 System and method for fog detection
KR101982470B1 (en) * 2018-03-21 2019-05-27 한국전력공사 Marine activity risk forecasting system
KR102040562B1 (en) 2019-09-16 2019-11-06 주식회사 아라종합기술 Method to estimate visibility distance using image information
KR20210044127A (en) * 2019-10-14 2021-04-22 주식회사 에드오션 Visual range measurement and alarm system based on video analysis and method thereof
KR102352881B1 (en) * 2020-09-02 2022-01-18 주식회사 환경과안전 Methods for setting up dangerous air space for unmanned aerial vehicles
KR20220071637A (en) 2020-11-24 2022-05-31 주식회사 이루리 High Beam System
EP3951439A4 (en) * 2019-03-28 2022-12-14 Xiamen Kirincore IOT Technology Ltd. Advection fog forecasting system and forecasting method
KR102580758B1 (en) 2022-03-29 2023-09-20 (주)이륜코리아 visible light maximization collecting apparatus using the mist inside laser beam scattering phenomenon

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20150134810A (en) * 2014-05-23 2015-12-02 주식회사 솔트룩스 Social sensor system for sensor web
KR20160029397A (en) * 2014-09-05 2016-03-15 경희대학교 산학협력단 Method and apparatus for value added service based on evaluation of the sights
KR20180055205A (en) * 2016-11-16 2018-05-25 이승재 System and method for fog detection
KR101982470B1 (en) * 2018-03-21 2019-05-27 한국전력공사 Marine activity risk forecasting system
EP3951439A4 (en) * 2019-03-28 2022-12-14 Xiamen Kirincore IOT Technology Ltd. Advection fog forecasting system and forecasting method
KR102040562B1 (en) 2019-09-16 2019-11-06 주식회사 아라종합기술 Method to estimate visibility distance using image information
KR20210044127A (en) * 2019-10-14 2021-04-22 주식회사 에드오션 Visual range measurement and alarm system based on video analysis and method thereof
KR102352881B1 (en) * 2020-09-02 2022-01-18 주식회사 환경과안전 Methods for setting up dangerous air space for unmanned aerial vehicles
KR20220071637A (en) 2020-11-24 2022-05-31 주식회사 이루리 High Beam System
KR102580758B1 (en) 2022-03-29 2023-09-20 (주)이륜코리아 visible light maximization collecting apparatus using the mist inside laser beam scattering phenomenon

Similar Documents

Publication Publication Date Title
KR20130107424A (en) Method, system and control system for detecting and forecasting fog based on ubiquitous sensor network and semantic technology
KR101732819B1 (en) Disaster predicting platform system based on big data and methd thereof
CN109145954A (en) A kind of net based on multisource spatio-temporal data about vehicle safety evaluation method and system
US8751490B1 (en) Automatically determining reputations of physical locations
CN114866608B (en) Intelligent water affair data processing platform
KR101890497B1 (en) Analysis Method for Risk of Traffic Accident and Storage Medium Having the Same
Sammarco et al. Using geosocial search for urban air pollution monitoring
KR20190077682A (en) System and Method for Analysing Urban Safety Index using Public Big Data and SNS Data
US20150281653A1 (en) System and method for selecting sensors in surveillance applications
CN113570867B (en) Urban traffic state prediction method, device, equipment and readable storage medium
CN111524353A (en) Method for traffic text data for speed prediction and travel planning
KR102124935B1 (en) Disaster Monitoring System, Method Using Crowd Sourcing, and Computer Program therefor
CN111401218A (en) Smart city monitoring method and system
Mahmud et al. State of mobile crowdsourcing applications: A review
de Assis et al. Geographical prioritization of social network messages in near real-time using sensor data streams: an application to floods.
CN117079148A (en) Urban functional area identification method, device, equipment and medium
Singh et al. Flood detection framework fusing the physical sensing & social sensing
Re et al. Assessment of crowdsourced social media data and numerical modelling as complementary tools for urban flood mitigation
Sánchez-Ávila et al. Detection of barriers to mobility in the smart city using Twitter
US20150348407A1 (en) Recommendation Engine Based on a Representation of the Local Environment Augmented by Citizen Sensor Reports
Ramírez et al. Accident risk detection in urban trees using machine learning and fuzzy logic
CN110928922B (en) Public policy analysis model deployment method and system based on big data mining
Huang et al. Estimating urban noise along road network from street view imagery
CN108563676B (en) Integrated retrieval system of meteorological data
Feng et al. Social media as a rainfall indicator

Legal Events

Date Code Title Description
A201 Request for examination
E902 Notification of reason for refusal
E601 Decision to refuse application