KR101357600B1 - System for observing wether and thereof method - Google Patents

Abstract

The present invention relates to a meteorological observation system and method, comprising: first and second cameras used to measure cloud-related clouds and cloud shape and altitude; A synchronized image acquisition unit for acquiring cloud image information by receiving a cloud image signal synchronized with a predetermined time or need from the first and second cameras; A multilevel image constructing unit for acquiring cloud image information from the first and second cameras and collecting three-dimensional and multi-layer cloud image information to extract and construct cloud information generated and moving in various layers; A cloud observation information extracting unit for extracting cloud observation information of the quantity, shape, height, and moving direction of the cloud from the multilevel cloud image information collected through the multilevel image forming unit; The first and second cameras as well as the weather observation system including a synchronized image acquisition unit, a multi-level image configuration unit, an operation control unit for controlling the operation of the cloud observation information extraction unit and a weather observation method performed using the same Technical configuration is disclosed.

Description

Meteorological observation system and its method {SYSTEM FOR OBSERVING WETHER AND THEREOF METHOD}

The present invention relates to a meteorological observation system and a method thereof, and more particularly, to a meteorological observation system and method for enabling accurate weather prediction by observing and comparing accurate cloud information in real time.

ㅇ Domestic situation

-Currently, a trained observer observes the clouds with the naked eye every three hours, and should be able to grasp the changes in the state of the clouds and the changes in air pressure at other times as well.

In addition, the night observation is used to observe the cloud after darkening the eyes.

ㅇ Overseas

-There is a device to measure the amount of clouds during the daytime sunny day, but there is also a problem that a person must install and adjust.

Recently, many studies on cloud observation and prediction have been conducted as one of the countermeasures against uncertainties in climate prediction.
Clouds cover more than half of the earth at any moment and have a major impact on weather, climate and radiation balance, and the global water cycle. Therefore, accurate cloud observation data is very important not only for accurate weather forecast but also for precipitation and temperature forecasts.
In the current weather observation, a trained observer observes the clouds with the naked eye every three hours. In this system, the observer can grasp the changes in the cloud's state and air pressure at other times as well as the observation time. There was an inconvenience to be.
In addition, it is inefficient because the night observation is used to observe the cloud after dark eyes.
On the other hand, there is a device for measuring the amount of clouds in the daytime of sunny days outside, there is also inconvenience that a person must install and adjust directly.
Accordingly, researches on cloud observation and prediction have recently been conducted as one of the measures against uncertainties in climate prediction.

The present invention has been made in view of the above problems, and its purpose is to observe the current state of the cloud in real time to enable accurate interpretation and prediction of meteorological phenomena such as torrential torrential and guerrilla torrential weather and climate. The present invention provides a meteorological observation system and method capable of providing a more accurate weather information and forecasting weather, while improving the technology of forecasting and forecasting significantly.
The purpose of the present invention is to predict and forecast weather and climate by using two cameras to observe the current situation of clouds in real time and to enable accurate interpretation and prediction of meteorological phenomena such as torrential rain or guerrilla rain. In addition, the present invention provides a weather observation system and method capable of providing more accurate weather information and forecasting weather forecasts.

According to an aspect of the present invention,

It consists of an intelligent expert system that can be learned and trained by an expert, and then design and manufacture an automatic observation system of the sky state that can be automatically observed by a program.

Configure the synchronized image acquisition unit to acquire the cloud image information of the nation at the same time according to the predetermined time and needs,

It consists of a system that can acquire 3D, multi-layered cloud image information to extract cloud information generated and moving in multiple layers, and constructs a multilevel image acquisition unit that can remove water from the camera when precipitation occurs. ,

The cloud observation information extraction unit is configured to process the collected multilevel cloud image information to measure cloud quantity, shape, height, vector, etc.

It integrates the cloud information collected from various sky state observation equipments in various parts of the country and constructs an integrated server unit that can provide information to meteorological personnel if necessary.
The meteorological observation system of the present invention,
First and second cameras used for measuring cloud-related clouds and cloud shape and altitude;
A synchronized image acquisition unit for receiving cloud image signals synchronized with the first and second cameras to obtain cloud image information;
A multilevel image constructing unit for acquiring cloud image information from the first and second cameras and collecting three-dimensional and multi-layer cloud image information to extract and construct cloud information generated and moving in various layers;
A cloud observation information extracting unit for extracting cloud observation information of the quantity, shape, height, and moving direction of the cloud from the multilevel cloud image information collected through the multilevel image forming unit;
The first and second cameras, as well as the synchronized image acquisition unit, the multi-level image configuration unit, the cloud control information extraction unit comprises an operation control unit for controlling the operation of the technical features of the technical configuration.
The meteorological observation method of the present invention,
Pre-adjusting the directionality by performing position correction on the first and second cameras;
Capturing clouds with the first and second cameras on which the position correction has been performed to obtain stereo image information on the clouds;
Constructing a multilevel image from the stereo image information;
Extracting a cloud area by using cloud image information when the multilevel image configuration is completed;
When the cloud area is extracted, constructing a 3D cloud information map using stereo image information; The basic feature of the technical method is to perform cloud observation by analyzing the stereo image information, multi-level image and 3D cloud information, and extracting and storing cloud observation information.

ㅇ Ripple effect

-If accurate observation data are provided in real time through the development of 『Sky State Automated Observation System』, it contributes to the interpretation and prediction of Korean weather phenomena such as heavy rain and guerrilla rain, and predicts and forecasts the weather and climate of the Korean Peninsula. It will be able to improve the technology to provide more accurate weather information and weather forecasts.

-By accurately observing and automating accurate information in real time, it saves a lot of time and money for training experts who can visually observe, and additional expenses and labor costs incurred by working a large number of workers. Can be reduced.

-In addition, in the situation where there is no equipment that can be commercially applied to domestic and foreign countries, it can succeed in localization of commercialization in a short time, and it can not only go one step further to the advanced weathered country, but also acquire national intellectual property rights (patent registration). It is expected that it will be able to obtain and export to foreign countries without compromising foreign patents.

ㅇ Field of application and application

-Meteorological information is a fact that the need for economic and industrial aspects is known and everyone agrees. Therefore, it can contribute to the establishment of weather forecasting models using supercomputers, and it is possible to improve the technology of predicting and forecasting the weather and climate of the Korean Peninsula, thus providing more accurate weather information, which can be useful for various economic and industrial activities.

-In case of meteorological disasters, the damages are largely caused by erroneous observations and forecasts, inaccuracies in forecasting, and lack of forecasting technology. In this project, accurate cloud observation data are provided to forecasters in real-time through the development of the Automatic Sky Observation System. It is expected to be able to help more accurate forecasts, and to provide the DB of cloud observation information to other organizations that study meteorological observations and forecasts, thereby contributing to raising the level of meteorological forecasting and improving the quality of public service. It is expected to be.
The present invention can perform visibility and cloud observation through image processing using two cameras, and can measure a wide range of visibility criteria and viewing distances in various directions, as well as the shape and amount of clouds. , The speed of movement, height, etc. can be easily and more accurately measured, and by providing such accurate observation data, it can contribute to the interpretation and prediction of meteorological phenomena such as torrential rain or guerrilla heavy rain. Improve forecasting and forecasting skills and provide more accurate weather information and forecasts.
Since the present invention can observe and automate accurate information, the present invention not only saves a lot of time and expense, but also greatly increases the efficiency of weather observation.
The present invention can contribute to the establishment of weather prediction model using a supercomputer, and can improve the technology of predicting and forecasting the weather and climate of the Korean Peninsula, thereby providing more accurate weather information, which can be useful for various economic and industrial activities. It can also contribute to minimizing damage from meteorological disasters.

A preferred embodiment of the weather observation system and method thereof according to the present invention will be described with reference to the drawings.

ㅇ Development contents: 『Sky state automatic observation system』

-Design and manufacture of the sky automatic observation system: It consists of an intelligent expert system that can be learned and trained by an expert, and then a system capable of automatic observation by a program.

-Synchronized image acquisition unit: configured to acquire the cloud image information of the nation at the same time according to the specified time and needs.

-Multilevel Image Acquisition Unit: It consists of a system that can acquire three-dimensional and multi-layered cloud image information to extract moving cloud information generated from multiple layers, and can remove water that is formed in the camera when precipitation occurs. Configured as a system.

-Cloud Observation Information Extraction Unit: It is configured to measure cloud quantity, shape, height, vector by processing multilevel cloud image information collected.

-Integrated server unit: Integratedly processes the cloud information collected from the various sky state observation equipment in the country, and consists of a system that can provide information to the weather observation personnel if necessary.

ㅇ Sky processing system

① Synchronized image acquisition unit

-Designed and installed for camera calibration

-Acquisition of synchronized information according to fixed time and instructions

② Multilevel Image Acquisition Unit

-Can be obtained in the three-dimensional form of the 360 ㅀ panoramic pyramid type

-Composed of a system that can acquire multi-layer cloud image information

-Composed of a system that can acquire cloud image information even at night

③ cloud observation information extraction unit

-Can acquire cloud amount, cloud type, cloud height, direction of movement, etc.

④ Integrated Server

-Integrated processing of cloud information collected from observation equipments nationwide

-S / W construction that can provide information to meteorological observation personnel

ㅇ Observation information of sky state automatic observation system

① cloud volume (cloud quantity)

Measurement range: 10 minutes, 0,1,2,3,4,5,6,7,8,9,10

-Uncertainty: within ㅁ 1

② Cloud shape (cloud form)

Cirrus Cirrus

Cirrocumulus (Cc)

-Cirrostratus (Cs)

-Cumulus Altocumulus (Ac)

Skyscraper Altostratus (As)

Difficult cloud Nimbostratus (Ns)

Stratocumulus (Sc)

Stratus (St)

Cumulus Cumulus (Cu)

Cumulonimbus (Cb)

③ cloud height (cloud height)

-Upper cloud: 3 ~ 13km

-Middle cloud: 2 ~ 8km

-Lower cloud: surface ~ 2km

Uncertainty: ± 10%

④ vector of clouds

-Measuring range: 0 ~ 360 °

Uncertainty: ± 5 °

Clouds cover more than half of the Earth at any moment and have a great impact on weather, climate, radiation balance, and global water circulation. Therefore, accurate cloud observation data are not only accurate weather forecasts, but also very helpful for precipitation and temperature forecasts, so they are used as important data not only in Korea but also in foreign countries.

* On the other hand, Korea is currently suffering a lot of damage from meteorological disasters such as rapid weather changes and guerrilla heavy rains, and although much efforts and investment have been made for accurate weather prediction, there are still insufficient parts and are lagging behind advanced weather countries.

1. 3D panorama, multi level image acquisition system

PANNING, TILTING, ZOOMING, Focal Length Adjuster

2. Cloud Area Extraction

-Use information such as sun position, moon position, constellation and comet according to date and time

-Use the distribution characteristics and changes of the sky color during sunny, cloudy, sunrise and sunset

-Change of color and distribution characteristics according to the position of the sun at sunrise and sunset

-At night, using ultra low light camera and extracting cloud area considering moon and constellation

3. Calculation of clouds (cloud)

-Extracted cloud area / sky

4. Cloud form (cloud)

-3D shape model construction by cloud type

-2D cloud model building according to location

-Established as an expert support system (built by a system taught by experts and capable of program learning)

5. Calculate the height of the cloud (cloud height)

-Height information extraction using 3D stereoscopic (3D binocular vision, triangulation)

6. Calculation of cloud movement direction (cloud motion vector extraction)

-Using motion vector extraction algorithm

7. Full cloud information model DB construction

Review

One). There is an existing product that measures the cloud volume by taking one camera image in the form of a panorama, and makes a database by configuring it as a network.

2). One camera can measure clouds and clouds, but cannot measure the height of clouds. Possible using two or more stereo methods

In FIG. 1, when there is a cloud, the image of camera1 cannot distinguish whether it is located at the altitude of cloud A and the altitude of cloud B, but in another camera2, the triangulation method shows the position of cloud A and cloud B differently. Using the principle of, the altitude of the cloud can be measured from the camera. Therefore, if the image is acquired at regular intervals, the direction and speed of the cloud can also be measured.

Municipal Hyuncheon Observation System Using Image Processing

■ Outline

This system is a meteorological observation system that can measure a wide range of visibility distance in various directions, and can measure cloud shape, cloud volume, cloud direction, cloud movement speed, and cloud height.

■ Configuration

1) Camera whose PTZ control is directly controlled by S / W program

2) PT LASER is controlled by S / W program.

3) PTZ control unit

4) Moisture and dirt prevention and removal device

■ Operation Description of FIG. 2

1) Perform camera calibration using the combination of the camera control device with PTZ control.

2) When used as a visibility system, three-dimensional distance information map is extracted by acquiring stereo image information of an object on the ground.

3) From the constructed 3D map, select and store the location of the main observation point, and then determine the viewing distance by acquiring the image of the reference position later. Color information is used during the day, and black and white images or laser beams are illuminated at night to determine the viewing distance based on the image information input to the camera and store it in the DB.

4) When the cloud information is extracted, the sky image is acquired at regular time intervals, the cloud image region is extracted, and then the 3D information of the cloud is extracted.

5) After that, the cloud quantity, cloud type, cloud height, direction of movement, and speed of movement are extracted and stored in the DB.

6) Integrate the data observed from each observation device, and as the number of observation points increases, visibility and cloud related information can be obtained in real time.

* It is possible to use laser, but it is better to use it.

* Moisture and dirt prevention and removal devices will be added in future mechanical details.
A preferred embodiment of the weather observation system and method according to the present invention will be described with reference to the drawings.
As shown in FIG. 1, the weather observation system according to the present invention includes first and second cameras 10 and 20 used to measure cloud-related clouds, clouds, and altitude; A synchronized image acquisition unit 40 which receives the cloud image signal synchronized with a predetermined time or need from the first and second cameras 10 and 20 to obtain cloud image information; The multilevel image constructing unit 50 acquires cloud image information from the first and second cameras 10 and 20, and collects and builds cloud information generated and moved in various layers by collecting three-dimensional and multi-layer cloud image information 50. )Wow; A cloud observation information extracting unit (60) for extracting cloud observation information such as the amount, shape, height, and moving direction of the cloud from the multi-level cloud image information collected by the multi-level image forming unit (50); An operation control unit 30 for controlling operations of the first and second cameras 10 and 20, as well as the synchronized image acquisition unit 40, the multilevel image construction unit 50, and the cloud observation information extraction unit 60. )
In this case, it is preferable that the first and second cameras 10 and 20 use stereoscopic type cameras for stereoscopic image acquisition, respectively, and PTZ (Pan / Tilt / To control the position according to the Zoom value.
In addition, by further comprising a laser 70 for generating a laser beam and enabling PT (Pan / Tilt) position control, image acquisition through the first and second cameras 10 and 20 can be obtained even when the weather is dark or at night. It is desirable to configure it so that it can be easily and accurately performed and acquire color information at any time.
Through the multi-level image configuration unit 50 is processed to obtain three-dimensional and multi-layer cloud information, such as pyramid-shaped panorama.
Meanwhile, the present invention may further include a cloud region extracting unit 80 for extracting a cloud region from the multilevel cloud image information obtained by the multilevel image constructing unit 50.
In this case, the cloud region extracting unit 80 may extract the cloud region from the cloud image information by using information such as the position of the sun, the position of the moon, the constellation, and the comet according to the date or time, or when it is sunny or cloudy, The cloud region can be extracted from the cloud image information by using the distribution characteristics and changes of the sky colors at sunrise and sunset, or by using the distribution characteristics and changes of color according to the position of the sun at sunrise or sunset. The area can be extracted.
In the present invention, the stereoscopic image information of an object on the ground is obtained using the first and second cameras 10 and 20 for use as a visibility system together with a cloud observation system, and through this, three-dimensional distance information. A three-dimensional distance information map constructing unit 90 for extracting and constructing a map; A correction reference position selection unit (100) for selecting and storing positions of major observation points for preparing correction criteria from the three-dimensional distance information map constructed in the three-dimensional distance information map constructing unit (90); When the reference position is selected, the first and second cameras (10, 20) can be configured to further include a correction distance extraction unit 110 for obtaining the image of the reference position and extracting the correction distance. 4 shows an example of performing cloud observation and visibility observation in parallel.
Furthermore, in the present invention, as shown in the example of Figure 3, integrated to process the cloud information collected from various weather observation systems in the country as well as to provide meteorological observation information to the weather observation personnel as needed. In the present invention, the operation control unit 30 may include the first and second cameras 10 and 20, as well as the synchronized image acquisition unit 40 and the multilevel image construction unit 50. , Cloud observation information extraction unit 60, laser 70, cloud region extraction unit 80, three-dimensional distance information map configuration unit 90, visibility reference position selection unit 100, visibility distance extraction unit 110 It will control the operation of.
The present invention provides a meteorological observation system that can not only measure a wide range of visibility criteria and visibility distance in various directions through the above-described technical configuration, but also measure the shape, cloud quantity, moving speed, height, etc. of clouds. To provide.
Hereinafter, the meteorological observation method together with the operation of the meteorological observation system according to the present invention having the above-described configuration will be described with reference to FIGS. 1 to 12.
First, a description will be given of the case of using the cloud observation with two first and second cameras 10 and 20,
Referring to the flowchart of FIG. 5, the orientation is pre-adjusted by performing position correction on the first and second cameras 10 and 20 provided to enable PTZ control (S1).
The cloud is imaged using the first and second cameras 10 and 20 on which the position correction is performed to obtain stereo image information on the cloud (S2).
At this time, by using the first camera 10 and the second camera 20, it is possible to obtain a video image from each of them as shown in the example of FIG. 7, with the video image obtained from the first camera 10 Although the cloud A and cloud B can not distinguish the altitude at which altitude is located, the altitude of the cloud A and cloud B can be measured using a principle such as triangulation through the image image obtained from the second camera 20, Through the stereo system through the first and second cameras 10 and 20, the cloud shape and amount can be more accurately extracted.
Here, the first and second cameras 10 and 20 acquire a video image of the cloud at a predetermined time interval while controlling the PTZ values for each of them.
When the cloud image information is acquired as the stereo image information, a multi-level image is constructed and stored from the cloud image information, such as a panorama conversion of the still image segmented as shown in FIGS. 10 to 12 (S3).
At this time, it is possible to compose various multi-level images for the multi-layered clouds that occur in multiple layers and move.
When the multilevel image configuration such as the panorama conversion process through the cloud image information is completed, the cloud region is extracted by dividing the cloud image by region using the cloud image information (S4).
At this time, the image image of the cloud is divided into regions by a unique pattern that is considered homogeneous among the image images of the cloud. In particular, when the weather is sunny or cloudy, it is classified according to sunrise, sunset, or time. Cloud area can be extracted.
When the cloud region is extracted, a 3D cloud information map is constructed using stereo image information to extract 3D information of the cloud (S5).
In this case, the 3D information map of the cloud information may be configured using a method such as matching the stereo image information acquired through the first and second cameras 10 and 20 with each other.
By analyzing the stereo image information, multi-level image and three-dimensional information, cloud observation information such as cloud cloud, cloud shape, cloud height, moving direction, moving speed is extracted and stored (S6).
At this time, cloud cloud shape information and cloud region extraction information can accurately determine cloud form (cloud form) such as cirrus clouds, cirrus clouds, cirrus clouds, cumulus clouds, tall clouds, cumulus clouds, stratocumulus, stratosphere, cumulus clouds, cumulonimbus clouds, and cloud information from video information. The cloud volume can be measured by extracting the area.
The movement direction and velocity of clouds can be extracted using multilevel images and 3D information maps, and can be measured by tracking and vector analysis of positions by comparing before and after the movement of clouds as in the example of FIG. have.
In addition, a description will be given of the case where the two first and second cameras 10 and 20 are used for visual observation.
Referring to the flowchart of FIG. 6, the orientation is pre-adjusted by performing position correction on the first and second cameras 10 and 20 provided to enable PTZ control (S11).
The stereo image information of the object on the ground is obtained with the first and second cameras 10 and 20 on which the position correction is performed (S12).
When a stereo image of the object on the ground is obtained, a multilevel image such as pyramid-type panorama processing is constructed and stored from the stereo image information obtained while controlling the PTZ value (S13).
After constructing the multilevel image, a 3D distance information map is constructed (S14).
In this case, the 3D distance information map may be constructed by matching stereo image information with each other, measuring distance using a laser beam of the laser 70, or using GIS information.
When the 3D distance information map is constructed, the location of the main observation point to be used for the correction criteria is selected and stored (S15).
Then, the positions of the main observation points to be used for the correction criteria are selected, and the images of the selected correction reference positions are obtained by using the first and second cameras 10 and 20 and the correction distance is determined from the obtained image information. And store (S16).
Therefore, through the present invention, it is possible to easily and accurately obtain visibility and cloud-related information applicable to meteorological observations, and can be used for more accurate cloud observation by selecting and determining key observation points through visibility observation. It can be used as information for performing more accurate weather forecasting.

1 is an overall block diagram showing a meteorological observation system according to the present invention.
Figure 2 is an exemplary camera capable of direct control according to the PTZ value in the present invention.
3 is an exemplary diagram of comprehensive information processing for explaining the present invention;
Figure 4 is an exemplary view that can be used in parallel with the visibility system and cloud observation in the present invention.
Figure 5 is a flow chart when used as a cloud observation in the present invention.
6 is a flow chart when used as a visibility system in the present invention.
7 is a view illustrating a cloud observation using two cameras of the present invention.
8 to 12 are views for explaining the present invention.
Description of the Related Art [0002]
10, 20: camera 30: motion control unit
40: synchronized image acquisition unit 50: multi-level image composition unit
60: cloud observation information extraction unit 70: laser
80: cloud area extraction unit 90: 3D distance information map component
100: location selection government 110: municipal street extraction unit

delete

delete

delete

delete

delete

delete

delete

Claims (6)

First and second cameras used for measuring cloud-related clouds and cloud shape and altitude; A synchronized image acquisition unit for receiving cloud image signals synchronized with the first and second cameras to obtain cloud image information; A multilevel image constructing unit for acquiring cloud image information from the first and second cameras and collecting three-dimensional and multi-layer cloud image information to extract and construct cloud information generated and moving in various layers; A cloud observation information extracting unit for extracting cloud observation information of the quantity, shape, height, and moving direction of the cloud from the multilevel cloud image information collected through the multilevel image forming unit; The first and second cameras, as well as a weather observation system comprising a motion control unit for controlling the operation of the synchronized image acquisition unit, multi-level image configuration unit, cloud observation information extraction unit. The method of claim 1, And a laser that generates a laser beam and performs PT position control by an operation controller so as to more easily and accurately perform image acquisition through the first and second cameras. The method of claim 1, A 3D distance information map component for acquiring stereo image information of an object on the ground using the first and second cameras, and extracting and constructing a 3D distance information map through the first and second cameras; A correction reference position selection unit for selecting and storing positions of major observation points for preparing correction criteria from the 3D distance information map constructed by the 3D distance information map constructing unit; When the visibility reference position is selected, further comprising a visibility distance extracting unit for acquiring the image of the visibility reference position and extracting the visibility distance through the first and second camera, characterized in that the operation is controlled by the operation controller Weather observation system. The method of claim 1, And the first and second cameras each use a stereoscopic camera for stereoscopic image acquisition. Pre-adjusting the directionality by performing position correction on the first and second cameras; Capturing clouds with the first and second cameras on which the position correction has been performed to obtain stereo image information on the clouds; Constructing a multilevel image from the stereo image information; Extracting a cloud area by using cloud image information when the multilevel image configuration is completed; When the cloud area is extracted, constructing a 3D cloud information map using stereo image information; The cloud observation method characterized in that the cloud observation through the step of extracting and storing the cloud observation information by analyzing the stereo image information, multi-level image and three-dimensional cloud information to use for weather observation. Pre-adjusting the directionality by performing position correction on the first and second cameras equipped to enable PTZ control; Acquiring stereo image information of an object on the ground with the first and second cameras on which the position correction has been performed; When a stereo image of the object on the ground is obtained, constructing a multilevel image from the obtained stereo image information while controlling the PTZ value; Constructing a 3D distance information map after constructing the multilevel image; If the three-dimensional distance information map is configured, selecting a location of a main observation point to be used as a correction criterion therefrom; The position of the main observation point to be used for the correction criteria is selected, and the corrective observation is performed by acquiring the images of the selected correction reference positions using the first and second cameras and determining the correction distance from the obtained image information. A weather observation method, characterized in that used for weather observation.

Images