KR100428586B1 - Method for measuring snowfall - Google Patents

Abstract

PURPOSE: A method for telemetering snowfall is provided to improve resolution of a snowfall image and precision of measurement ,and to permit a snowfall image to be photographed in the dark. CONSTITUTION: The method for telemetering snowfall comprises the steps of: (1) dividing a reference scale into multiple vertical zones and disposing a camera at the position corresponding to each divided zone, wherein various preset values are initialized; (2) capturing a snowfall image of the reference scale from the camera disposed at the bottom; (3) extracting only the image of the reference scale and removing undesired images; (4) removing noises from the image of the reference scale; (5) searching sequentially from the pixel line located at the top on a screen to extract a snowfall border line; (6) capturing a snowfall image with the next camera, when the snowfall border line is not present, and repeating steps (3)-(5); and (7) calculating and determining the snowfall by using a real snowfall transformation table, when the snowfall border line is extracted.

Description

Snowfall telemetry method {METHOD FOR MEASURING SNOWFALL}

The present invention relates to a snow quantity telemetry system, and in particular, by splitting the snow snow criterion into a plurality of areas, by recording the snow image for each corresponding area with a different infrared camera and observing each, respectively, the resolution of the snow amount is improved to obtain a more precise observation. In addition, it is possible to remotely monitor the acquired image and snowfall data through a communication network without being influenced by the external environment, and thus to relate to a snowfall remote measurement method that can collect snowfall in a specific area quickly and cope more effectively. .

In recent years, national heavy snowfall and the surprise southern snowfall have caused enormous national losses. However, no immediate snowfall or snowfall warnings were made and no immediate and accurate snowfall measurements were performed in each region, so limited snow removal equipment was not efficiently carried out. Therefore, the necessity of snow measurement and management has been increased as part of effective weather observation for preventing such weather disasters and managing water resources.

In general, snowfall refers to the height of solid precipitation (snow, snow, hail, etc.) on the ground. Snow is defined as snow when the snow covers more than half of the ground. No matter how much snow it falls, if it doesn't cover more than half of the ground before it melts, there's no snowfall, only weather.

The unit of snow is cm, measured to the first digit after the decimal point. A common measure of snowfall is to measure the amount of snow accumulated using a ruler on a snowboard.

However, due to advances in science and technology, various sensing technologies have been developed, and more and more cases have been applied to meteorological observations.

However, the general snowfall measuring instrument uses analog and logic circuits, and the combination of simple analog and simple logic makes it difficult to reliably cope with complex situations. It is being repaired, and the snow design is made by using a general ultrasonic sensor or a temperature sensor instead of the snow observation sensor, so the reliability is lowered and the snow strength is operated only to a certain degree, so it is measured during heavy snow disaster. And lack of correction.

1 is a snowfall measuring apparatus according to the prior art, which has a support 11 installed in a vertical direction with respect to gravity, a crosspiece 12 installed in a horizontal direction at an upper end of the support 11, and the crossbar. A laser distance sensor 13 installed at an end of the laser beam 12 and measuring a distance from a reflection surface by firing a laser to the lower end, and controlling the operation of the laser distance sensor 13 and measuring the measured data; And a control unit 14 for temporarily storing the selected final data, a wireless antenna 15 for transmitting the final data selected by the control unit 14, a solar cell 17 for supplying power to the apparatus, The receiver 11 is installed at a position spaced apart from the support 11, receives data transmitted through the wireless antenna 15, processes the data through comparison with other data, and transmits the data to a destination. It is composed of a modem 23 for transmitting data to the data collector 20, a destination.

Of course, AC commercial power may be used as a driving power source without providing the solar cell 17. In addition, the data collector 20 is provided with a receiving antenna 21 for receiving data transmitted through the wireless antenna 15, the power source uses an AC commercial power.

In addition, in case the wireless communication is unstable, the RS-485 communication cable 16 is connected between the control unit 14 and the data collector 20 so that data can be transmitted by wire.

That is, the strut 11 is erected at a point to measure the amount of snow, and the laser distance sensor 13 is installed at the end of the crosspiece 12 located at the upper end of the strut 11 which is standing. The laser distance sensor 13 may measure the distance L1 from the laser sensor 13 to the ground by reading the reflected wave reflected by the ground after firing the laser toward the ground. In addition, when snow is accumulated on the ground, since the laser is reflected by the snow, the distance L2 from the laser distance sensor 13 to the upper surface of the stacked eyes is measured. Here, the height H of the snow accumulated on the ground is measured. Can be calculated

In other words, since the distance L1 between the laser distance sensor 13 and the ground is already recognized when the laser distance sensor 13 is installed, only the distance l between the laser distance sensor 13 and the upper surface of the eye is measured. Then, the difference between the two distances can be used to calculate the amount of snow accumulated.

However, such a measurement method is also very difficult to measure during a lot of snow fall because the laser emitted from the laser distance sensor 13 is reflected by the flying snow, accurate measurement was possible only after the snow stopped. Therefore, it is very difficult to measure the amount of snow in real time, there was a problem that it is very difficult to prompt information transmission and heavy snow handling due to heavy snow and heavy snow.

Accordingly, an object of the present invention is to divide the snow criterion into a plurality of areas, and to observe snowfall images of each corresponding area with different infrared cameras, thereby improving the resolution of the snow amount, enabling more accurate observation, and night photographing. It is to provide a possible snow telemetry method.

Another object of the present invention is to remotely monitor the snow data, if the snow or snow strength is generated above a certain standard, as well as the field image and snow quantity data automatically generates an alarm signal and transmits to the weather monitoring server through the communication network, time The present invention aims to provide a remote snow measurement method that can cope with the snow intensity and snow quantity step by step.

1 is a view showing a snowfall measuring apparatus according to the prior art,

2 is a conceptual diagram showing a snow observation system according to the present invention,

3 is a detailed configuration diagram showing a snow observation system according to the present invention,

4 to 5b are flowcharts illustrating the operation of FIG. 3 according to an embodiment of the present invention.

6 to 8 are screen diagrams for explaining the method of processing the snow image according to the present invention,

9A to 9C are diagrams for explaining the image noise filtering method according to the present invention;

FIG. 10 is a web screen illustrating a main control screen of the weather monitoring server of FIG. 3.

* Explanation of symbols for the main parts of the drawings

100: On-site design system 110: Snow standard

111 to 115: a plurality of divided regions 116: reference point

117,118: upper and lower baseline 120: snow border

130: measuring camera 140: surveillance camera

150: snowfall analysis controller 200: weather monitoring server

300: client

In the technical method of the present invention for achieving the above object, in the snow quantity measurement method: by dividing the snow reference standard into a plurality of vertical areas and to place the measuring camera at a position corresponding to each divided area, and initializes various set values First step; A second step of capturing the snow image of the snow criterion from the measurement camera at the bottom of the plurality of measurement cameras; A third step of extracting only the snow criterion image from the captured images and removing the remaining unnecessary images; A fourth step of removing the noise included in the snow criterion image after obtaining the snow criterion image; A fifth step of extracting a snow boundary by sequentially searching from the top pixel line on the screen to the bottom after removing the noise; A sixth step of acquiring a snow image with a next measuring camera and repeating the process from the third step when the snow boundary is absent; And a seventh step of calculating and determining the snow amount by comparing the divided area with the actual snow amount conversion table corresponding to the screen coordinates of the area when the snow boundary is extracted.

Hereinafter, the present invention will be described in more detail with reference to the accompanying drawings.

FIG. 2 is a conceptual diagram illustrating a snow observation system according to the present invention, and includes a snow cover standard 110, a measurement camera 130, a snow quantity analysis controller 150, a weather monitoring server 200, and a client 300. .

The snow criterion 110 is embedded in the road to indicate the amount of snow with a scale, but has a measurement area (111 to 115) divided into a predetermined length, the measurement camera 130 is divided into each of the snow criterion 110 The plurality of cameras 131 to 135 are installed to correspond to the areas 111 to 115 so as to capture and acquire an image of each of the divided areas 111 to 115. The snow quantity analysis controller 150 includes a measurement camera ( 130 is controlled to receive the image obtained from the measuring camera 130 at a predetermined time interval after the image processing is configured to analyze the snowfall and collect snowfall video and snowfall data, the weather monitoring server 200 Remotely control the snowfall analysis controller 150, and receive the snowfall data transmitted from the snowfall analysis controller 150 through a communication network to build a database to monitor local snowfall and a specific area through the Internet Consists to provide snow cover according to information, the client 300 is a weather monitor connected to the server 200, the administrator or the general user of the information communication terminal to confirm or monitor the snow accumulation information for a particular area over the Internet.

The snow criterion 110 is installed in the furnace (露 場), the height of the vertical upright is about 1.5m, the scale is engraved on the surface at approximately 5mm intervals visible to the naked eye. The snow criterion 110 has a measurement area divided at approximately 30 cm intervals. For example, the first area 111 from the ground to 30 cm, the second area 112 from 30 cm to 60 cm, the third area 113 from 60 cm to 90 cm, the fourth area 114 from 90 cm to 120 cm, and 120 cm to 150 cm Is divided into a fifth region 115.

In addition, the measuring camera 130 is composed of at least one infrared measuring camera 130 for acquiring an image for each area in which the snow standard 110 is divided vertically, for example, the first measuring camera 131 ) Photographs the first region 111 of the snow criterion 110, the second measurement camera 132 photographs the second region 112 of the snow criterion 110, and the third measurement camera 133. ) Photographs the third region 113, the fourth measurement camera 134 photographs the fourth region 114, and finally the fifth measurement camera 135 mutually photographs the fifth region 115. The measurement cameras 131 to 135 may be photographed by including a part of the corresponding area of the snow cover standard 110 and a part of other areas adjacent to the area.

The plurality of measuring cameras 130 may be used to improve the resolution of the amount of snow, thereby making it possible to observe more precisely. Also, the angle of view of the measuring camera 130 may be narrowed to reduce errors generated from the angle of view.

In addition, the unit length mark of the snow criterion 110 may be enough to identify when the image taken by the measurement camera 130 with the naked eye, the measurement camera 130 is capable of infrared photography for night photography It is preferable to use an infrared camera.

3 is a detailed block diagram showing the snow observation system according to the present invention, referring to Figure 2 as follows.

The snowfall analysis controller 150 is a controller based on a personal computer, and the snowfall analysis controller 150 is connected to a plurality of measurement cameras 130 and a surveillance camera 140.

The measurement camera 130 is a camera for measuring the snow, the surveillance camera 140 is a camera for monitoring the surrounding environment of the snow snow criterion 110, the surveillance camera 140 of the snow quantity analysis controller 150 Through the monitoring angle adjusting unit 159, pan / tilt / zoom adjustment is possible.

The processing controller 151 of the snowfall analysis controller 150 first receives an image of a unit time and another sensor signal from the measuring camera 130 to determine snowfall amount through image processing and signal processing, and measured snowfall amount. Is stored in the storage means 153 together with other image data, if necessary, together with other sensor data.

Then, according to the set time to transmit the numerical data on the surveillance image and snowfall amount or precipitation through the communication means 155 to the remote weather monitoring server 200, the weather monitoring server 200 from the snowfall analysis controller 150 The received snow quantity data and the image transmitted through the communication means 230 is properly processed by the processor 210 is displayed on the monitor 240 and stored in the database 220.

The snowfall analysis controller 150 receives the user's command from the weather monitoring server 200 to perform the measurement camera 130, the monitoring camera 140, and other operation commands.

Therefore, the weather monitoring server 200 should basically be able to build and store the weather-related data transmitted from the snowfall analysis controller 150 to the database 220, and receives and manages data from the snowfall analysis controller 150 Be able to play a role.

In addition, the weather monitoring server 200 is designed to allow each client 300 to see in real time from the remote site the snowfall data and field image obtained by the web-based remote image transmission system that can be used in the Internet and communication network environment, In order to check the environment through an image from a remote location, the viewing angle of the surveillance camera 140 can be adjusted remotely.

As such, the snow observation system overcomes the disadvantages of the conventional method and the unidirectionality by using image recognition technology, which is a multimedia technology, based on the signals of the measuring camera 130 and the intelligent sensors 161 and 165. It is a system that enables more accurate grasp of snow conditions in bad weather, and allowed users to remotely monitor from anywhere through the Internet.

The snowfall analysis controller 150 converts the analog signals input from the sensors 161 and 165 into digital signals and supplies the interfaces 163 and 167 to the processing controller 151 and a 7-segment or LCD screen. The display means 157 consisting of the date and time, the amount of snow or precipitation data received from the processing control unit 151 and displayed.

The precipitation sensor 161 and the precipitation sensor 165 connected to the sensor interfaces 163 and 167 basically purchase and use a sensor certified by the Korea Meteorological Administration, and the precipitation sensor 165 is a sensor for indicating precipitation. Since the DC voltage is output in the form of a pulse for each predetermined precipitation unit, the output control unit 151 of the snow quantity analysis controller 150 measures the precipitation by counting the output through the counter interface 167.

In addition, the precipitation sensor 161 is a sensor for detecting rain or snow, the output of the sensor generates both a digital output indicating the presence of precipitation and an analog output indicating the degree of precipitation, snowfall analysis controller 150 ) Accepts these signals via analog or digital interfaces 163 and 167 to measure presence and degree of precipitation.

Therefore, the snowfall analysis controller 150 monitors snowfall or precipitation and if snow, snow intensity or precipitation occurs above a certain standard, automatically generates alarm signals as well as snowfall image and snowfall data, and then monitors the weather through the communication network. As the transmission to the 200, the weather monitoring server 200 can see the snow-based or precipitation-related data on the web-based, and can immediately check the emergency situation, it is possible to respond quickly and step by step.

In addition, the snowfall analysis controller 150 additionally connects sensors such as the precipitation sensor 161 and the precipitation sensor 165 as well as an ultrasonic sensor, a wind speed sensor, and a wind direction sensor (not shown) to view on-site weather conditions. It is natural to be able to monitor accurately.

Hereinafter, a method of measuring the snow amount will be described with reference to the drawings of FIGS. 4 to 9.

First, the snow criterion 110 is divided into a plurality of vertical regions as shown in FIG. 6, and then divided into five regions 111 to 115, and the measurement camera is positioned at a position corresponding to each of the divided regions 111 to 115. In addition to disposing the portions 131 to 135, the processing control unit 151 of the snow quantity analysis controller 150 uses the upper reference line 117 and the lower reference line 118 as shown in FIG. 7 for each of the divided regions 111 to 115. After setting the reference points 116 above the upper reference line 117, the storage means 153 generates an actual snow amount conversion table for each of the divided areas 111 to 115 of the snow dropper 110. Initialization operation for snowy observation, such as storing in the (S1).

In addition, the reference point of a specific partition belongs to the upper reference line and the lower reference line of the upper partition, the upper reference line of the specific partition coincides with the lower reference line of the upper partition, and the lower reference line of the specific partition is the lower reference line of the lower partition. Matches the upper baseline.

Subsequently, the snow quantity analysis controller 150 controls the lowest measurement camera 131 among the plurality of measurement cameras 130 to capture a snow image of the snow reference standard 110 (S2).

The image obtained from the measuring camera 131 is taken by the processing control unit 151 as shown in FIG. 8 only the portion 119 where the snow criterion 110 is located and removing the remaining unnecessary images (S3), Removing the noise included in the image of the snow criterion 110 after obtaining the image of the snow criterion 110 at (S4), and after removing the noise, sequentially search from the top pixel line on the screen to the bottom In step S5, the boundary line 120 is extracted.

Before extracting the snowy boundary line 120 (S5) to perform the image preprocessing process (S4) as shown in Figure 5a, the preprocessing operation (S4) is to modify the deformation to easily extract the snowy boundary line by processing the input image It's work. In addition, the snow boundary uses an image segmentation technique (Region Growing, etc.) as a task of distinguishing the snow boundary of the snow cover standard 110 to extract and measure feature elements.

That is, in the method of observing snowfall using image processing, the first measurement region 131 of the snowfall reference 110 is photographed by controlling the lowest measurement camera 131 among the measurement cameras 130 (S2). The image captured by the measuring camera 131 is converted into digital image data using a capture board and stored in the storage means 153 of the snow quantity analysis controller 150.

Subsequently, the processing control unit 151 of the snow quantity analysis controller 150 cuts only the portion 119 of the snow criterion 110 to reduce the processing load of the digital image data obtained from the measurement camera 131. After taking it out, unnecessary images are removed (S3).

The acquired image is subjected to a preprocessing operation S4 as shown in FIG. 5A for easier recognition before recognizing the snow boundary of the snow criterion 110. In the preprocessing operation, the color image is converted into a gray image ( S4-1), the filtering unit such as a median filter is used to remove noise such as a point generated in the image (S4-2).

Meanwhile, FIG. 9A illustrates a process of filtering using a median filter. In the initial image, the neighboring points of the points to be filtered are grouped into one window, the values of the points are sorted, and the median value is filtered. This process is repeated to show a completely filtered image.

However, a large amount of snow accumulated in the vicinity of a camera lens or lens cannot be removed by a general median filter, and in order to remove noise such as large snow and rain, a plurality of continuous eyes such as FIG. 9B or 9C are removed. A frame median filter for removing noise by comparing the images of the frames is used (S4-3).

9B illustrates a method of extracting a previous image and a subsequent image based on an initial input image and grouping the previous and subsequent images into a window in the same manner as in FIG. 9A when a large noise that is difficult to remove by the median filter occurs. FIG. 9C illustrates a method of filtering initial image noise to be filtered to read values from before and after frames, align respective points, and select an intermediate value.

That is, the frame median filter receives images of consecutive frames and moves when photographing a fixed object with a filter for generating a single filtered image by selecting a median value of points located at the same position in each frame. It is a filter to remove when an object appears as noise.

In addition, since the snow image is continuously observed for 24 hours, the snow standard 110 and the contrast of the eye continuously change. It is necessary to normalize the changing contrast, and normalization is also done in preprocessing. This normalization is performed by creating a histogram of gray images, using contrast stretching techniques and histogram smoothing techniques.

The preprocessed image is converted into digital data by performing binarization as shown in FIG. 5B in order to extract the snow boundary line 120 (S5) (S5-1). The binarization is performed by automatically adjusting the threshold value according to the overall contrast of the image. The binarization criteria 110 are searched from the position of the reference point 116 of the snow criterion 110 to the lower reference line 118 in the binarized image. The snowy boundary of 120 is extracted (S5-2).

The method of extracting the snow boundary of the snow criterion 110 is performed by an image segmentation technique (Region Growing, etc.), and extracts the snow boundary line 120 by searching the screen from the reference point 116 of the snow criterion 110. To start.

When the snow cover line 120 is extracted (S6), it is determined whether the extracted snow cover line 120 exists between the upper reference line 117 and the lower reference line 118 (S7).

If present between the upper and lower reference lines 117 and 118, the processing control unit 151 calculates the actual height of the snow boundary line 120 by comparing with the conversion table stored in the storage means 153 (S8). .

When the snowfall is normally observed, the video and the snowfall of the snowfall boundary are stored in the storage means 153 and the display means 157 displays the snowfall or the weather monitoring server 200 through the communication means 155. The snow video and its data are transmitted (S9).

If the snow boundary is present between the reference point 116 and the upper reference line 117 in the above, it is determined whether it is the uppermost measuring camera 135 (S10), and if it is not the uppermost measuring camera 135, the current measuring camera ( It is determined that the snow boundary line exists in the image of the higher measurement camera 132 than the image of 131 and receives the image of the next higher measurement camera 132 and continues the image recognition operation (S11).

If there are no more upper-level measurement cameras, the snow cover is considered to have been snowed beyond the observation range of the snow criterion 110 and the maximum snow measurement value is displayed (S12).

10 illustrates a screen in which the client 300 accesses the weather monitoring server 200 and receives snowfall information for a specific region. The client 300 does not require a separate program, and the web browser (Internet Explorer, Netscape, etc.) can be used if the computer is installed.

First, the client 300 receives a service after inputting a domain name or address of the weather monitoring server 200 in an address window while driving a web browser.

On the screen, the left image is the image of the measuring camera 130, the right image is the image of the surveillance camera 140.

In addition, the 'Fixed CAM' button at the bottom of the screen is a button that shows the image of the selected measurement camera 130 on the screen alone or requests snowfall and image data, and the 'Pan / Tilt CAM' at the bottom of the screen. The button is a button which can show the image of the selected pan / tilt surveillance camera 140 on the screen alone or perform the pan / tilt control of the surveillance camera 140, and the snow dropper 110 is installed at the bottom of the screen. It provides snowfall data and video data for each point, and the 'SETUP' button on the bottom right of the screen provides a function for setting the IP address and other settings of the weather monitoring server 200, and the 'SEARCH' button for weather monitoring. It serves to retrieve the snowfall information contained in the database 220 of the server 200.

Although a specific embodiment of the present invention has been described and illustrated, the present invention is not limited thereto, and the present invention can implement not only snowfall but also precipitation measurement or crop growth management by analyzing images and analyzing results using a plurality of measuring cameras. It can be varied by those skilled in the art, such as changing the number of cameras to be measured or analyzing snowfall by inputting an image by pan / tilt / zoom adjustment of the area of the snow criterion divided into a single camera instead of a plurality of cameras. It is obvious that it may be modified. Such modified embodiments should not be individually understood from the technical spirit or the prospect of the present invention, but should fall within the claims appended to the present invention.

Therefore, in the present invention, the snow accumulated in the snow criterion is imaged by an infrared camera to automatically analyze the snowfall, and then transmits the snowfall to a monitoring center without being influenced by the external environment, thereby remotely communicating the acquired image and snowfall data through a communication network. It is possible to monitor and collect snowfall in a specific area more quickly to cope with it more effectively, and divide snow cover standards into a plurality of areas, and observe snowfall images for each area with different infrared cameras. By improving the resolution, more accurate observations are possible, as well as night shots.

In addition, when snow or snow strength is exceeded by a certain standard by remotely monitoring snow data, alarm signal is automatically generated as well as field image and snow quantity data and transmitted to the weather monitoring server through a communication network, This has the advantage of being able to deal with snowfall in stages and appropriately.

Claims (12)

delete delete delete delete In the snowfall measurement method: A first step of dividing the snow criterion into a plurality of vertical regions and arranging a measurement camera at a position corresponding to each division region, and initializing various setting values; A second step of capturing the snow image of the snow criterion from the measurement camera at the bottom of the plurality of measurement cameras; A third step of extracting only the snow criterion image from the captured images and removing the remaining unnecessary images; A fourth step of removing the noise included in the snow criterion image after obtaining the snow criterion image; A fifth step of extracting a snow boundary by sequentially searching from the top pixel line on the screen to the bottom after removing the noise; A sixth step of acquiring a snow image using a next measurement camera and repeating the operation from the third step when the snow boundary is not present; And And a seventh step of calculating and determining the snow amount by comparing the divided area with the actual snow amount conversion table corresponding to the screen coordinates of the area when the snow boundary is extracted. The method according to claim 5, The first step is, Setting an upper reference line and a lower reference line for each divided area; Setting an area reference point at an upper end of the upper reference line; And And a step of preparing an actual snow amount conversion table for each partition of the snow criterion. The method according to claim 6, The reference point is a snowfall telemetry method characterized in that it belongs between the upper reference line and the lower reference line of the upper partition. The method according to claim 6, The upper reference line is identical to the lower reference line of the upper partition area, the lower reference line is a snow quantity remote measurement method, characterized in that the same. The method according to claim 5, The fourth step, Converting a color image of the acquired image into a gray image; Removing noise such as dots generated in the image; And And removing the noise by comparing with a previous image frame obtained sequentially to remove sporadically occurring noise such as snow and rain falling on the measuring camera lens. The method according to claim 5, The method of extracting the snow boundary of the fifth step, the snow quantity remote measurement method, characterized in that using the image segmentation technique. The method according to claim 5 or 6, The fifth step, Binarizing each pixel of the snow criterion image; Dividing the image and extracting snow boundary lines by searching from a reference point to a lower reference line via an upper reference line; Determining whether the snow cover is present between the upper reference line and the lower reference line; And And recognizing the image of the next higher measurement camera when the snowfall boundary line is extracted between the reference point and the upper reference line. The method according to claim 5 or 6, The snow cover of the seventh step, if there exists between the upper reference line and the lower reference line snow cover remote measurement method, characterized in that to calculate the snow cover using the snow cover conversion table.