KR101602293B1 - Method and apparatus to estimate coastline using image processing - Google Patents

Abstract

A method for measuring a shoreline through image processing according to the present invention includes: a first step of generating a plurality of images by photographing a shoreline for a predetermined time; A second step of geometrically correcting each of the plurality of images into actual terrain coordinates; A third step of obtaining an average value of each pixel value included in the plurality of geometrically corrected images to generate one average image; A BR difference image having a brightness value obtained by subtracting a red-band pixel value from a B-band pixel value for each pixel included in the generated average image, ; A fifth step of extracting a plurality of pixels in which a value obtained by subtracting a pixel value of a B-band and a pixel value of an R-band from the BR difference image is a predetermined value or less; A sixth step of calculating a deviation of a pixel value for each pixel included in the plurality of geometrically corrected images in the second step and generating a deviation image having the calculated deviation value as a pixel value; A seventh step of extracting a plurality of pixels having the deviation value equal to or larger than a predetermined value in the generated deviation image; And an eighth step of determining an average coastline by finding overlapping areas of the plurality of pixels extracted in the fifth step and the plurality of pixels extracted in the seventh step.

Description

TECHNICAL FIELD [0001] The present invention relates to a method and apparatus for coastline measurement using image processing,

The present invention relates to a method and apparatus for measuring a shoreline, and more particularly, to a method and apparatus for measuring the position of a shoreline in an image photographed through image processing, based on an image taken through a photographing apparatus fixedly installed on the coast .

Globally, coastal erosion is widespread due to the occurrence of high waves caused by typhoons and rising sea level due to climate change in each country. In order to establish medium- and long-term countermeasures against this serious coastal erosion problem, it is necessary to quantify the extent of coastal change and coastal erosion through long-term and continuous monitoring of coastal erosion.

Conventionally, in order to measure the coastline change and the degree of coastal erosion, it is common practice for a person to directly observe a coastal area by using a RTK-GPS (Real Time Kinematic? Global Positioning System) I am using it. Since RTK-GPS generates errors within a few centimeters on the move, there is a disadvantage in that the price is high instead of being high in accuracy, and because it is necessary to measure the person moving directly on the coast using such an observation device, have.

Therefore, when the coastline change and the coastal erosion degree are measured as in the past, it is impossible to continuously monitor the coastal erosion. In particular, there is a limitation in effectively measuring the coastal erosion where a large change occurs in a short period of time such as typhoon.

The present invention proposes a method for observing the shoreline change continuously and in the long term by installing the stationary photographing equipment at a place near the coast.

It is to be understood that the description described as the background of the above-mentioned invention is merely for enhancing the understanding of the background of the present invention, and it is accepted that it corresponds to the prior art already known to those having ordinary skill in the art It should not be accepted.

An object of the present invention is to measure the position of a shoreline by image processing based on an image taken through a photographing apparatus fixedly installed on the coast and to measure the position of a shoreline without the burden of cost and manpower due to direct observation And to provide a method and apparatus for measuring a shoreline.

According to another aspect of the present invention, there is provided a method for measuring a shoreline through image processing, the method comprising: a first step of generating a plurality of images by photographing a shoreline for a predetermined time; A second step of geometrically correcting each of the plurality of images into actual terrain coordinates; A third step of obtaining an average value of each pixel value included in the plurality of geometrically corrected images to generate one average image; A BR difference image having a brightness value obtained by subtracting a red-band pixel value from a B-band pixel value for each pixel included in the generated average image, ; A fifth step of extracting a plurality of pixels in which a value obtained by subtracting a pixel value of a B-band and a pixel value of an R-band from the BR difference image is a predetermined value or less; A sixth step of calculating a deviation of a pixel value for each pixel included in the plurality of geometrically corrected images in the second step and generating a deviation image having the calculated deviation value as a pixel value; A seventh step of extracting a plurality of pixels having the deviation value equal to or larger than a predetermined value in the generated deviation image; And an eighth step of determining an average shoreline by obtaining overlapping areas of the plurality of pixels extracted in the fifth step and the plurality of pixels extracted in the seventh step.

The third step is to classify each of the plurality of images into R-band, G-band, and B-band, and calculate an average value of each pixel value of the R-band, G- Band average image, a G-band average image, and a B-band average image, and then generates the average image by superimposing the R-band average image, the G-band average image, and the B- .

Here, in the seventh step, a plurality of pixels having the deviation value equal to or greater than a predetermined value may be determined as the breaking-band region.

According to another aspect of the present invention, there is provided an apparatus for measuring a shoreline through image processing, comprising: a photographing unit for photographing a coastline for a predetermined period of time and generating a plurality of images; A geometry correcting unit for geometrically correcting each of the plurality of images to actual terrain coordinates; The method of claim 1, further comprising: calculating a mean value of each pixel value included in the plurality of geometrically corrected images to generate an average image, and calculating a B-band pixel value And the R-band pixel value is subtracted from the BR difference image, and the R-band pixel value is subtracted from the R-band pixel value. An average image processing unit for extracting a plurality of pixels having a predetermined value or less; Calculating a deviation of a pixel value for each pixel included in the plurality of geometrically corrected images, generating a deviation image having the calculated deviation value as a pixel value, A deviation image processing unit for extracting a plurality of pixels that are equal to or larger than a predetermined number of pixels; And a shoreline determining unit for determining an overlapping area of a plurality of pixels extracted by the average image processing unit and a plurality of pixels extracted by the deviation image processing unit to determine an average shoreline.

The average image processing unit may classify each of the plurality of images into R-band, G-band, and B-band, and may calculate an average value of each pixel value of the R-band, G- Band average image, a G-band average image, and a B-band average image, and then generates the average image by superimposing the R-band average image, the G-band average image, and the B- .

Here, the deviation image processing unit may determine a plurality of pixels having the deviation value equal to or larger than a predetermined value as the breaking-band region.

According to the present invention, by measuring the position of a shoreline through image processing based on an image photographed through a photographing apparatus fixedly installed on the coast, accurate coastline position can be obtained without the burden of cost and manpower due to direct observation It is possible to provide a measurable shoreline measurement method and apparatus.

1 is a block diagram of a shoreline measuring apparatus according to an embodiment of the present invention.
2 is a flowchart of a coastline measurement method according to an embodiment of the present invention.
FIGS. 3 to 11 illustrate examples of images generated in each step according to processing of photographed images according to an embodiment of the present invention.

It is noted that the technical terms used herein are used only to describe specific embodiments and are not intended to limit the invention. It is also to be understood that the technical terms used herein are to be interpreted in a sense generally understood by a person skilled in the art to which the present invention belongs, Should not be construed to mean, or be interpreted in an excessively reduced sense. Further, when a technical term used herein is an erroneous technical term that does not accurately express the spirit of the present invention, it should be understood that technical terms that can be understood by a person skilled in the art are replaced. In addition, the general terms used in the present invention should be interpreted according to a predefined or prior context, and should not be construed as being excessively reduced.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings, wherein like reference numerals refer to like or similar elements throughout, and redundant description thereof will be omitted. In the following description, well-known functions or constructions are not described in detail since they would obscure the invention in unnecessary detail. It is to be noted that the accompanying drawings are only for the purpose of facilitating understanding of the present invention, and should not be construed as limiting the scope of the present invention with reference to the accompanying drawings. The spirit of the present invention should be construed as extending to all modifications, equivalents, and alternatives in addition to the appended drawings.

1 is a block diagram of a shoreline measuring apparatus 100 according to an embodiment of the present invention.

The shoreline measuring apparatus 100 includes a photographing unit 110, an image processing unit 120 and a shoreline determining unit 130. The image processing unit 120 includes a geometry correcting unit 122, (124) and a deviation image processing unit (126).

The photographing unit 110 is fixed near the shoreline to take an image including the shoreline. The photographing unit 110 may be implemented by a known device such as a camcorder or a closed circuit television (CCTV) capable of generating a digital image. The image processing unit 120 and the shoreline determining unit 130, which will be described later, It is also possible to transmit an image photographed via wired / wireless communication at a remote location.

The photographing unit 110 generates a plurality of (N) images through continuous photographing in a time series, and it is possible to perform photographing continuously for a predetermined time (15 to 20 minutes) so as to have a statistical meaning. For example, when a frame rate is photographed for 20 minutes at 30 frames / sec, a total of 27,000 images can be generated.

Hereinafter, an image refers to a frame unit unless otherwise stated, and one color image is divided into three bands constituting a separate layer, that is, R-band (R-band) Red-band, G-band and B-band, and color images can be completed by overlapping R-band, G-band and B-band.

The image processing unit 120 may include a geometric correction unit 122, an average image processing unit 124, and a deviation image processing unit 126. [ The geometric correction unit 122, the average image processing unit 124, and the deviation image processing unit 126 may be hardware (or software) .

3, the shoreline measuring apparatus 100 measures a shoreline through a series of image processing on the shot image shown in Fig. 3 Will be described with reference to the images shown in Figs. 4 to 11, respectively.

The geometry corrector 122 geometrically corrects each of the N images to the actual terrain coordinates. The photographed image is formed in two dimensions unlike the actual image, and may be distorted unlike the shape in the actual spatial coordinates. The distortion existing in the image includes a refractive distortion caused by a camera lens and a geometric distortion due to coordinate transformation. The geometry correction unit 122 corrects the geometric distortion of the image by using a known method such as DLT (Direct Linear Transformation) Distortion can be corrected. The image generated according to the geometric correction by the geometric correction unit 122 is as shown in FIG. In FIG. 4, the parts shown in black in the upper left and lower right corners correspond to the areas deleted by the geometric correction, and not in the actual terrain coordinates.

The average image processing unit 124 estimates the area of the shoreline based on the average image of the plurality of geometrically corrected images.

First, the average image processing unit 124 obtains an average value of each pixel value included in the plurality of geometrically corrected images, and generates one average image. At this time, the average image processing unit 124 obtains an average value of the pixel values of the R-band, G-band, and B-band images to generate an R-band average image, a G- Then, an R-band average image, a G-band average image, and a B-band average image are superimposed to generate an average image.

As is known, each image includes a plurality of pixels, for example, if the resolution is 1024 * 768, one image can be implemented with a combination of 786,432 pixels. In addition, each pixel implements one color by a combination of R, G, and B, for example, in the case of 256 colors, the R, G, and B pixel values of each pixel may have a value of 0 to 255. [ When the pixel value of the pixel corresponding to the nth row and mth column is represented by Pn * m (R_value, G_value, B_value), the corresponding pixel value of the R-band of the image is (R_value, 0, 0) (0, G_value, 0) and the B-band will be (0, 0, B_value).

To obtain an average image, the average image processing unit 124 obtains an average of pixel values of each pixel by R-band, G-band, and B-band of N images and obtains R-band average image and G- (R_avg, 0, 0), Pavg_n * m (0, G_avg, 0), and Pavg_n * m 0, 0, B_avg). Then, an average image is generated by superimposing the average images of the respective bands, and the pixel values of n rows and m columns of the average image can be expressed by Pavg_n * m (R_avg, G_avg, B_avg). The average image of the N images generated according to the above-described method is as shown in Fig.

When the average image processing unit 124 generates one average image as shown in Fig. 5 as the processing result of the N images, the R-band pixel value at the pixel value of the B-band for each pixel included in the average image And generates a BR difference image having a value obtained by subtracting the value from the brightness.

The value B_avg-R_avg obtained by subtracting the R-band pixel value from the B-band pixel value of the n-th row of the m-th column of the average image has a value of 0 to 255. When this is applied to each pixel, The pixel value of the pixel may be Pbr_n * m (B_avg - R_avg, B_avg - R_avg, B_avg - R_avg). That is, the BR difference image can be implemented as a monochrome image in which pixel values of each pixel are all the same in the RGB band, and only brightness differences between the respective pixels are different. The BR difference images thus generated are as shown in FIG.

As described above, the BR difference image is generated to utilize the characteristic of changing the color from the brown to the white line to the blue line at the boundary between land and sea. More specifically, it is common that the terrestrial sandy beaches are of the brown system, the sea of the sea has the blue system color, and the B-band pixel value and the R-band pixel value in the brown system and the blue system show different values. On the contrary, the part corresponding to the shoreline where land and sea meet corresponds to the white system in which the breaking wave breaking motion occurs. In the white system, the B-band pixel value and the R-band pixel value have the same value. 6, since the difference between the B-band pixel value and the R-band pixel value is close to 0, the pixel value is (0, 0, 0), that is, a color close to black, and the other portion may exhibit a constant brightness with a large difference between the B-band pixel value and the R-band pixel value.

The average image processing unit 124 extracts a plurality of pixels whose difference between the B-band pixel value and the R-band pixel value in the B-R difference image is predetermined or less, that is, close to zero. As described above, pixels in the shoreline region where the waves meet the land and represent white can be extracted.

Fig. 7 shows the coastline estimated finally, on the B-R difference image, as described later. As shown in FIG. 7, although it does not correspond to the shoreline like the areas of the x-axis 100 to 150 and y-axis 170 to 300, the difference between the B-band pixel value and the R- May be present. In the case of such an area, it is independent of the coastline and can be removed by overlapping with the area extracted by the deviation image processing unit 126 as described later.

The deviation image processing unit 126 estimates the area of the shoreline based on the deviation images of the plurality of geometrically corrected images.

First, the deviation image processing unit 126 calculates the deviation of pixel values for each pixel included in the N images subjected to geometric correction. At this time, the deviation of each pixel can be calculated for each pixel value of R-band, G-band, and B-band of N images. For example, when the pixel values of the nth row and mth column of the i-th image among the N images are represented by Pi_n * m (R_value, G_value, B_value), the deviation of the R-band of each of the N images is first obtained, The deviation can be calculated as the following formula.

 In this manner, deviation images of R-band, G-band, and B-band of each of the N images are respectively generated and then superimposed to generate one deviation image. The generated deviation image is as shown in Fig.

In this way, the reason for generating the deviation image is that there is not a large change in the sea and sandy areas in the continuous images, but the wave and the wave breaks up actively in the vicinity of the shoreline, Because. In other words, since the change is larger in the vicinity of the shoreline than in the other regions, a large deviation value can be generated.

The deviation image processing unit 126 extracts a plurality of pixels whose deviation values are equal to or larger than a predetermined value on the generated deviation images. More specifically, when the pixel values of the n-th row and m-th column of the deviation image are represented by Pdev_n * m = (R_dev, G_dev, B_dev), pixels satisfying R_dev ≥α, G_dev ≥α, and B_dev ≥α are extracted. In this case, the pixel values of the extracted pixels are all high in the R-band, G-band, and B-band, that is, close to 255, so they may be close to white as shown in FIG.

The deviation image processing unit 126 may determine the extracted plurality of pixels as the breaking-band region, and display only the determined breaking-band region in white as shown in FIG. That is, since the region indicated by white in FIG. 9 is a region where a lot of changes occur in the time-series image, the wave is assumed to be a breaking-band region that meets with the land and breaks.

The shoreline determination unit 130 includes a plurality of pixels extracted by the average image processing unit 124, that is, a plurality of pixels having a value obtained by subtracting a pixel value of a B-band and a pixel value of an R- A plurality of pixels extracted by the deviation image processing unit 126, that is, a superimposition area of pixels of a breaking-band area having a deviation value of a predetermined value or more in the deviation image is obtained. The obtained overlapping area is as shown in Fig.

As a result of obtaining the overlap region, the region that is not related to the coastline in Fig. 7 is removed, and only the region similar to the actual coastline is left.

The coastline determining unit 130 determines the obtained overlapping area as an average coastline. FIG. 11 shows the average coastline determined in accordance with the above-described method, which is superimposed on the average image of FIG. 5, and it can be confirmed that the shoreline that can be estimated visually matches with the coastline determined by the coastline determining unit 130.

As a result of using the shoreline measuring apparatus 100 according to the present invention, the average image and the deviation image are generated based on the photographed image, the position of the shoreline is measured through image processing of a series of processes, It is possible to measure the position of the accurate shoreline without the burden of cost and manpower.

2 is a flowchart of a coastline measurement method according to an embodiment of the present invention.

The coastline measurement method shown in FIG. 1 is performed by the coastline measurement apparatus 100 described above with reference to FIG. 1, and the description of the technical features already described will be omitted.

The coastline measuring apparatus 100 photographs the coastline for a predetermined time to generate a plurality of images (S10). The shoreline measuring apparatus 100 generates a plurality of (N) images through continuous shooting in a time series, and the shooting can be continuously performed for a predetermined time (15 to 20 minutes) so as to have a statistical meaning . The photographed N images are as shown in Fig.

The coastline measuring apparatus 100 geometrically corrects each of the plurality of photographed images to the actual terrain coordinates (S20). The photographed image may be distorted unlike the shape in the actual spatial coordinates because it is two-dimensionally formed unlike the actual image. The shoreline measuring apparatus 100 may be constructed by a known method such as DLT (Direct Linear Transformation) The geometric distortion of the image can be corrected. The image generated according to the geometric correction is as shown in Fig.

The shoreline measuring apparatus 100 calculates an average value of each pixel value included in a plurality of geometrically corrected images to generate an average image (S31). At this time, each of a plurality of images is classified into R-band, G-band, and B-band, and an average value of pixel values of R-band, G- After generating the G-band average image and the B-band average image, the R-band average image, the G-band average image, and the B-band average image may be superimposed to generate an average image. The average image thus generated is as shown in FIG.

Thereafter, the shoreline measuring apparatus 100 calculates a value obtained by subtracting the R-band (red-band) pixel value from the B-band pixel value for each pixel included in the generated average image A BR difference image having brightness is generated (S32). 6, since the difference between the B-band pixel value and the R-band pixel value is close to 0, the pixel value is (0, 0 , 0), that is, a color close to black, and the other portion may exhibit a certain amount of brightness with a large difference between the B-band pixel value and the R-band pixel value.

The coastline measuring apparatus 100 extracts a plurality of pixels whose values are less than or equal to a predetermined value, that is, a value obtained by subtracting a pixel value of the B-band from a pixel value of the R-band in the BR difference image (S33). Accordingly, the pixels of the shoreline region in which the waves meet the land and represent white can be extracted.

On the other hand, the shoreline measuring apparatus 100 calculates the deviation of the pixel values for each pixel included in the plurality of geometrically corrected images, and generates one deviation image having the calculated deviation value as the pixel value (S41). At this time, the deviation of each pixel is calculated for each pixel value of the R-band, G-band and B-band of the N images, and then the deviation image of the R-band, G- Can be superimposed to generate one deviation image. The generated deviation image is as shown in Fig.

Thereafter, the shoreline measuring apparatus 100 extracts a plurality of pixels having the deviation value equal to or larger than a predetermined value in the generated deviation image (S42). In this case, the pixel values of the extracted pixels are all high in the R-band, G-band, and B-band, that is, close to 255, so they may be close to white as shown in FIG.

The shoreline measuring apparatus 100 may determine the extracted plurality of pixels as a breaking-band region (S43) and display only the determined breaking-band region in white as shown in FIG.

When the average image analysis and the deviation image analysis are completed as described above, the shoreline measuring apparatus 100 obtains the overlapping areas of the plurality of pixels extracted in step S33 and the plurality of pixels extracted in step S43. The overlapping area obtained here is as shown in Fig. As a result of obtaining the overlap region, the region that is not related to the coastline in Fig. 7 is removed, and only the region similar to the actual coastline is left.

The coastline measuring apparatus 100 determines the obtained overlapping area as an average coastline (S50). Fig. 11 shows the average coastline determined according to the above, superimposed on the average image in Fig. 5, and it can be confirmed that the shoreline determined by the naked eye and the coastline determined by the coastline measurement apparatus are almost identical.

Meanwhile, the shoreline measurement method according to an embodiment of the present invention described above can be implemented through a computer program that can be executed through various arithmetic processing units. The computer program may include program instructions, data files, data structures, etc., alone or in combination, and examples of program instructions may include machine code such as those generated by a compiler, as well as instructions that may be executed by a computer using an interpreter, You can include the high-level language code. In addition, the computer program may be those specially designed and constructed for the present invention or may be those known to those skilled in the art of software.

Further, the computer program for executing the above method may be recorded in a computer-readable recording medium. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

100: Coastline measuring device
110:
120:
122: Geometrical Correction
124: average image processing section
126: Deviation image processing section
130: coastline determining section

Claims (6)

A coastline measurement method using image processing,
A first step of photographing a coastline for a predetermined time to generate a plurality of images;
A second step of geometrically correcting each of the plurality of images into actual terrain coordinates;
An average value is calculated for each color value of R-band, G-band, and B-band in each pixel included in the geometrically corrected images, A third step of generating one average image;
A fourth step of generating a BR difference image having a value obtained by subtracting a color value of an R-band from a color value of a B-band with brightness for each pixel included in the generated average image;
A fifth step of extracting a plurality of pixels in which a value obtained by subtracting a color value of a B-band from a color value of an R-band is equal to or less than a predetermined value in the BR difference image;
The deviation of the color values of the R-band, the G-band, and the B-band is calculated for each pixel included in the plurality of geometrically corrected images in the second step, A sixth step of generating one deviation image;
A seventh step of extracting a plurality of pixels having the deviation value equal to or larger than a predetermined value in the generated deviation image; And
An eighth step of determining an average coastline by obtaining overlapping areas of the plurality of pixels extracted in the fifth step and the plurality of pixels extracted in the seventh step.
The method according to claim 1,
In the third step,
Band image, an R-band average image, and an R-band average image by dividing each of the plurality of images into R-band, G-band, and B- A G-band average image, and a B-band average image, and then the R-band average image, the G-band average image, and the B-band average image are superimposed to generate the average image.
The method according to claim 1,
In the seventh step,
Wherein the plurality of pixels having the deviation value equal to or greater than a predetermined value are determined as the breaking-band region.
A shoreline measuring apparatus for image processing, comprising:
A photographing unit for photographing a coastline for a predetermined time to generate a plurality of images;
A geometry correcting unit for geometrically correcting each of the plurality of images to actual terrain coordinates;
An average value is calculated for each color value of R-band, G-band, and B-band in each pixel included in the geometrically corrected images, And generates a BR difference image having a brightness obtained by subtracting the color value of the R-band from the color value of the B-band with respect to each pixel included in the generated average image An average image processing unit for extracting a plurality of pixels whose values obtained by subtracting a color value of the B-band and a color value of the R-band from the BR difference image are equal to or less than a predetermined value;
Calculating deviations with respect to color values of R-band, G-band, and B-band for each pixel included in the plurality of geometrically corrected images, calculating one deviation image having the calculated deviation value as a color value A deviation image processing unit for generating a plurality of pixels having the deviation value equal to or greater than a predetermined value in the generated deviation image; And
And a shade determining unit for determining an overlapping area between a plurality of pixels extracted by the average image processing unit and a plurality of pixels extracted by the deviation image processing unit to determine an average shoreline.
5. The method of claim 4,
Wherein the average image processing unit comprises:
Band image, an R-band average image, and an R-band average image by dividing each of the plurality of images into R-band, G-band, and B- Band average image, a B-band average image, and then superimposes the R-band average image, the G-band average image, and the B-band average image to generate the average image.
5. The method of claim 4,
The deviation image processing unit,
And the plurality of pixels having the deviation value equal to or larger than a predetermined value are determined as the breaking-band region.

Images