JP6911697B2 - Water level judgment program, water level judgment method, and water level judgment device - Google Patents

Description

The disclosed technique relates to a water surface determination program, a water surface determination method, and a water surface determination device.

A running water area detection system that detects a flowing water area only from a video signal with a river or the like as a subject is known. In this running water region detection system, first, an image including a running water region and a non-flowing region taken by a fixed camera is added over a plurality of frames. As a result, an image in which the moving flowing water portion is blurred or smooth can be obtained. That is, the edge component (high frequency component) in the flowing water region is suppressed. Then, the running water region detection system emphasizes the edge component of the added image. Then, the flowing water region detection system classifies each pixel in the generated image into whether or not it is an edge pixel, and discriminates between the flowing water region and the non-flowing region based on the obtained binary image. Specifically, the region containing a large amount of edge components is discriminated as the land portion corresponding to the non-flowing region, and the other regions are discriminated as the flowing water portion corresponding to the flowing water region. As a result, the flowing water region can be detected only from the video signal of a river or the like as a subject without installing any object in the water.

Japanese Unexamined Patent Publication No. 2007-256254

However, even if the frame images included in the image of a river or the like are added over a plurality of frames, for example, if a flat area is included in a non-flowing area different from the running water area, the flowing water area is erroneously detected. It may end up. For example, when a smooth embankment with no unevenness exists on the land, the edge component is not emphasized in the embankment region, so that the embankment may be detected as a running water region.

On one side, the disclosed technology is aimed at accurately identifying areas of the water surface.

A first embodiment of the disclosed technique represents, in one embodiment, the edge strength of each pixel of an average image obtained by averaging the pixel values of the corresponding pixels in each of a plurality of frame images of an image representing the water surface. Get the edge image of. In addition, a second edge image representing the edge strength of each pixel of the specific frame image in the video is acquired. Further, a region in which the difference between the edge strength of each pixel of the first edge image and the edge strength of each pixel of the second edge image is equal to or greater than a predetermined threshold value is determined to be a region where the water surface is reflected. ..

The disclosed technique, in one embodiment, comprises the brightness of each pixel of an average image obtained by averaging the pixel values of the corresponding pixels in each of a plurality of frame images of an image representing the water surface and the image. Acquires the difference from the brightness of each pixel of a specific frame image of. Then, a region in which the difference between the brightness of each pixel of the average image and the brightness of each pixel of the specific frame image in the video is equal to or greater than a predetermined threshold value is determined to be the region where the water surface is reflected.

As one aspect, it has the effect of being able to accurately identify the region of the water surface.

It is a schematic block diagram of the water surface determination system which concerns on 1st Embodiment. It is a figure which shows an example of the image which a river is reflected. It is a figure which shows an example of the table stored in the image storage part. It is explanatory drawing for demonstrating the relationship between an average image and a specific frame image. It is explanatory drawing for demonstrating the edge image and the waterline which represents the area of a water surface. It is explanatory drawing for demonstrating the identification of the draft line candidate. It is explanatory drawing for demonstrating the difference histogram. It is explanatory drawing for demonstrating the determination of the waterline. It is a block diagram which shows the schematic structure of the computer which functions as the water surface determination apparatus which concerns on 1st Embodiment. It is a flowchart which shows an example of the water surface determination processing routine of 1st Embodiment. It is a schematic block diagram of the water surface determination system which concerns on 2nd Embodiment. It is a block diagram which shows the schematic structure of the computer which functions as the water surface determination apparatus which concerns on 2nd Embodiment. It is a flowchart which shows an example of the water surface determination processing routine of 2nd Embodiment.

Hereinafter, an example of an embodiment of the disclosed technology will be described in detail with reference to the drawings.

<First Embodiment>

As shown in FIG. 1, the water surface determination system 10 according to the first embodiment includes a camera 12, a water surface determination device 14, and an output device 30.

The camera 12 captures an image of a river as an image showing the surface of the water. In the present embodiment, for example, as shown in FIG. 2, the water surface W of the river is reflected below the frame image I of the river image in the vertical direction. Further, it is assumed that the image of the river of the present embodiment shows an area different from the water surface W of the river (for example, the area of the embankment T, the area of the grass G, and the area of the building B).

As shown in FIG. 1, the water surface determination device 14 includes an image storage unit 16, an average image acquisition unit 18, a brightness conversion unit 20, an edge acquisition unit 22, a histogram acquisition unit 23, and a waterline candidate acquisition unit 24. A difference acquisition unit 25, a score acquisition unit 26, and a water surface determination unit 28 are provided.

The image storage unit 16 stores a plurality of frame images of the images captured by the camera 12. The plurality of frame images are stored in the form of a table, for example, as shown in FIG. In the table shown in FIG. 3, a number representing the identification information, a time, and a frame image corresponding to the time are stored in association with each other. In the table shown in FIG. 3, for example, the frame image I1 acquired at time t1 is stored as the data corresponding to the number 1.

The average image acquisition unit 18 acquires a plurality of frame images stored in the image storage unit 16 in a predetermined time interval. Then, the average image acquisition unit 18 averages a plurality of frame images in a predetermined time interval and acquires an average image. For example, the average image acquisition unit 18 acquires an average image obtained by adding and averaging the pixel values of the corresponding pixels in each of the tens to hundreds of frame images. The frame image in the video is, for example, 30 frames per second.

The lightness conversion unit 20 converts the average image acquired by the average image acquisition unit 18 into lightness, and acquires an average lightness image showing a lightness image corresponding to the average image. Further, the brightness conversion unit 20 converts the specific frame image used when the average image acquisition unit 18 acquires the average image into brightness, and acquires a specific brightness image indicating the brightness image corresponding to the specific frame image. .. The lightness image is a grayscale image and can be obtained by an existing method.

For example, the average image acquiring unit 18, when acquiring the average image to acquire an average image by averaging a plurality of frame images I _'a shown in FIG. Further, in the present embodiment, the specific frame image used when acquiring the average image is, for example, as shown in FIG. 4, among _{a plurality of frame images I'a used for acquiring the average image.} Refers to any of the frame image It. Incidentally, a specific frame image I _t is not limited if it is one of frame images of a plurality of frame images I _'a. Each of the plurality of frame images I _'a and the specific frame image I _t is, as long as it is obtained from the time interval of a degree that the water surface of the water level does not change. For example, may be obtained immediately before or after the frame images of a plurality of frame images I _'a as a specific frame image I _t.

The edge acquisition unit 22 acquires a first edge image representing the edge intensity of each pixel of the average brightness image based on the average brightness image acquired by the brightness conversion unit 20. Further, the edge acquisition unit 22 acquires a second edge image representing the edge intensity of each pixel of the specific brightness image based on the specific brightness image acquired by the brightness conversion unit 20. The edge acquisition unit 22 acquires an edge image by applying a spatial filter to the brightness image, for example. The edge strength in the present embodiment represents the degree of edges included in the image, the vertical edge strength represents the degree of vertical edges in the image, and the horizontal edge strength represents the horizontal edge strength in the image. Represents the degree of edge in the direction. Further, the edge strength of the present embodiment also includes a value (for example, a value of 1 or 0) when the edge in the image is binarized.

FIG. 5 shows an explanatory diagram for explaining an edge image and a waterline representing a region of the water surface. _{FIG. 5 shows an average image Ia} showing the area W on the water surface, the embankment T, the grass G, and the building B reflected in the back of the river. It is assumed that the edge acquisition unit 22 obtains an edge image _{IE with respect to this average image I a. The edge image IE} shown in FIG. 5 is an image obtained by binarizing the edges extracted from the average image I _a. The white pixels in the edge image _IE represent pixels whose edge degree is equal to or higher than a predetermined threshold value. Here, consider a case where the position of the waterline representing the boundary of the water surface is determined based on the histogram H representing the frequency of white pixels in the horizontal direction with respect to each position in the vertical direction of _{the image IE.} In this case, for example, it is conceivable to determine that the region where the frequency of white pixels is low, that is, the region where the edge strength is low is the water surface. However, the edge intensity for regions of levee T appearing in the average image I _a is flat as shown in FIG. 5 is low. Therefore, as shown in the detection result R, the upper end tu of the embankment T is detected as a waterline.

In the average image, since the frame image at each time is averaged, the edge intensity becomes small in the region where the water surface is reflected even when the wave exists in the region of the water surface. On the other hand, in each of the frame images at each time, the edge intensity is large because the waves are present in the water surface region.

Therefore, in the present embodiment, the region where the water surface is reflected is determined based on the difference between the edge strength obtained from the average image and the edge strength obtained from the specific frame image. Specifically, a region in which the difference between the edge strength of the average image and the edge strength of the specific frame image is large is determined to be a region in which the water surface is reflected. This makes it possible to accurately identify the area where the water surface is reflected. Hereinafter, a specific description will be given.

The histogram acquisition unit 23 is based on the first edge image obtained by the edge acquisition unit 22, and the edge of each pixel in the horizontal direction corresponding to the position in the vertical direction with respect to each position in the vertical direction of the first edge image. Generate an edge strength histogram that represents the sum of the strengths. Further, the histogram acquisition unit 23 is based on the second edge image obtained by the edge acquisition unit 22, and the pixels in the horizontal direction corresponding to the positions in the vertical direction with respect to the positions in the vertical direction of the second edge image. Generates an edge strength histogram that represents the sum of the edge strengths of.

Specifically, as shown in FIG. 6, _{the vertical direction of the edge image IE} is the y direction, and the horizontal direction is the x direction. The histogram acquisition unit 23, a y direction of the y-coordinate of the edge image I _E on the horizontal axis, and the vertical axis the sum of the edge intensities of all x coordinates in the y-coordinate of the edge image I _E edge strength histogram H _E To generate. Specifically, for each y-coordinate of the edge image, the edge image is scanned in the x-axis direction at the y-coordinate to count the edge strength.

In the example shown in FIG. 6, the position of the upper end tu embankment in the image of the river is set beforehand, for the area above the upper end tu embankment in the frame image, the values of the edge intensity histograms H _E This is an example when it is set not to be generated. Therefore, the left than the upper end tu dikes edge strength histogram H _E is zero. Further, the edge image shown in FIG. 6 is an edge image showing both the edge strength of the vertical edge and the edge strength of the horizontal edge from the average image.

The waterline candidate acquisition unit 24 obtains the y-coordinate of the peak in the edge intensity histogram of the first edge image acquired by the histogram acquisition unit 23. For example, the waterline candidate acquisition unit 24 sets a position where the value of the edge intensity histogram obtained from the first edge image is equal to or more than a preset threshold value and the difference between the values before and after is equal to or more than a predetermined value. Detect as a peak. Then, the draft line candidate acquisition unit 24 sets a line segment parallel to the x-axis passing through the y-coordinate of the obtained peak as a draft line candidate.

For example, as shown in FIG. 6, the y-coordinate y1, y2, y3 corresponding to the peak P1, P2, P3 of the edge strength histogram _{H E,} draft candidate y1, y2, y3 in the frame image I Set.

The difference acquisition unit 25 includes the first edge image and the second edge image based on the edge intensity histogram of the first edge image and the edge intensity histogram of the second edge image acquired by the histogram acquisition unit 23. Get a difference histogram that represents the difference between. The difference histogram used in this embodiment is a histogram showing the difference between the edge strength of the average image and the edge strength of the specific frame image. Further, the difference histogram is a histogram representing the total difference of each pixel in the horizontal direction corresponding to the position in the vertical direction with respect to each position in the vertical direction of the frame image.

First, the difference acquisition unit 25 calculates the edge intensity histogram Ha _a (y) of _{the first edge image E a as shown in FIG.} Next, the difference acquisition unit 25 calculates the edge intensity histogram H _f (y) of _{the second edge image E f. The first edge image E a} and the second edge image E _f shown in FIG. 7 are edge images in which vertical edges are extracted. Further, as shown in FIG. 7, the edge image E _d representing the difference between the second edge image E _f and the first edge image E _a is that the difference between the edge strength at the water surface area is large Understand.

Then, the difference acquisition unit 25 _{calculates the difference histogram H d} (y) representing the difference between the _{edge intensity histogram H a} (y) and the edge intensity histogram H _{f (y).} As shown in FIG. 7, in the difference histogram H _d (y), it can be seen that the difference is large in the region where the water surface is reflected and small in the region other than the region where the water surface is reflected.

_{Further, the difference acquisition unit 25 normalizes the difference histogram H d} (y) according to the following equation (1), and acquires a _{normalized difference histogram H n} (y) representing the normalized difference histogram. In the normalization process according to the equation (1), the normalization is performed using the _{edge intensity histogram Ha a (y).} Thus, for example, be the difference values in the difference histogram H _{d (y)} is the same value, as the value of the average edge strength obtained from the image histogram H _{a (y)} is large, the value of the difference is smaller ..

_{H n (y) = H d} (y) / (H a (y) +1) (1)

The score acquisition unit 26 sets the vertical position corresponding to the peak of the normalized difference histogram obtained by the difference acquisition unit 25 as a candidate for the water surface waterline. Further, as shown in FIG. 7, the score acquisition unit 26 calculates a score according to the slope of the normalized difference histogram at the peak for each draft of the water surface.

Specifically, the score acquisition unit 26 calculates the differential value of the value _{of the normalized difference histogram H n} (y) at the y coordinate corresponding to each draft line candidate set by the draft line candidate acquisition unit 24. do. Each y coordinate of the waterline candidate _y i (i = 1, ..., N) when the score acquisition unit 26, according to the following equation (2), calculates a differential value _{F i} for each water line candidate i do. In addition, N represents the number of draft line candidates. In the present embodiment, in order to assume that the river is reflected in front of the frame image, in the case of the differential value F _{i <0} is set to F _{i =} 0.

_{F i = dH n (y i} ) / dy (2)

Then, the score acquisition unit 26, based on each of respective differential values calculated for each waterline candidate, calculates the score S _i in the waterline candidate i which is set by the draft candidate acquisition unit 24. _{The score S i} of the draft line candidate i (i = 1, ..., N) is expressed by, for example, the following equation (3).

（３）

(3)

The water surface determination unit 28 determines that the water surface candidate having the highest score is the water surface water line based on each of the scores of each water line candidate calculated by the score acquisition unit 26. For example, the water determination unit 28, as shown in FIG. 8, when the score S ₂ of the score S ₁ and waterline candidate F2 candidates F1 of the waterline is calculated, ranking the two this in descending order. Then, the water surface determination unit 28 determines that the draft line candidate F2, which is the first place, is a draft line, and identifies the draft line on the image according to the corresponding y-coordinate.

An image on which the waterline determined by the water surface determination unit 28 is superimposed is output to the output device 30. The output device 30 is realized by, for example, a display or the like.

The water surface determination device 14 can be realized by, for example, the computer 50 shown in FIG. The computer 50 includes a CPU 51, a memory 52 as a temporary storage area, and a non-volatile storage unit 53. Further, the computer 50 controls read / write (R /) to control reading and writing of data to the input / output interface (I / F) 54 to which the input / output devices such as the camera 12 and the output device 30 are connected, and the recording medium 59. W) A part 55 is provided. Further, the computer 50 includes a network I / F 56 connected to a network such as the Internet. The CPU 51, the memory 52, the storage unit 53, the input / output I / F 54, the R / W unit 55, and the network I / F 56 are connected to each other via the bus 57.

The storage unit 53 can be realized by a Hard Disk Drive (HDD), a Solid State Drive (SSD), a flash memory, or the like. The storage unit 53 as a storage medium stores a water surface determination program 60 for causing the computer 50 to function as the water surface determination device 14. The water surface determination program 60 includes an average image acquisition process 61, a brightness conversion process 62, an edge acquisition process 63, a histogram acquisition process 64, a waterline candidate acquisition process 65, and a difference acquisition process 66. Further, the water surface determination program 60 includes a score acquisition process 67 and a water surface determination process 68. Information constituting the image storage unit 16 is stored in the image storage area 69.

The CPU 51 reads the water surface determination program 60 from the storage unit 53, expands it into the memory 52, and sequentially executes the processes included in the water surface determination program 60. The CPU 51 operates as the average image acquisition unit 18 shown in FIG. 1 by executing the average image acquisition process 61. Further, the CPU 51 operates as the lightness conversion unit 20 shown in FIG. 1 by executing the lightness conversion process 62. Further, the CPU 51 operates as the edge acquisition unit 22 shown in FIG. 1 by executing the edge acquisition process 63. Further, the CPU 51 operates as the histogram acquisition unit 23 shown in FIG. 1 by executing the histogram acquisition process 64. Further, the CPU 51 operates as the waterline candidate acquisition unit 24 shown in FIG. 1 by executing the waterline candidate acquisition process 65. Further, the CPU 51 operates as the difference acquisition unit 25 shown in FIG. 1 by executing the difference acquisition process 66. Further, the CPU 51 operates as the score acquisition unit 26 shown in FIG. 1 by executing the score acquisition process 67. Further, the CPU 51 operates as the water surface determination unit 28 shown in FIG. 1 by executing the water surface determination process 68. Further, the CPU 51 reads information from the image storage area 69 and expands the image storage unit 16 into the memory 52. As a result, the computer 50 that has executed the water surface determination program 60 functions as the water surface determination device 14. The CPU 51 that executes the water surface determination program 60, which is software, is hardware.

The function realized by the water surface determination program 60 can also be realized by, for example, a semiconductor integrated circuit, more specifically, an Application Specific Integrated Circuit (ASIC) or the like.

Next, the operation of the water surface determination device 14 according to the first embodiment will be described. The water surface determination device 14 executes the water surface determination processing routine shown in FIG. 10 when each frame image of the river image starts to be stored in the image storage unit 16.

In step S100, the average image acquisition unit 18 acquires a plurality of frame images in a predetermined time interval stored in the image storage unit 16.

In step S102, the average image acquisition unit 18 averages the plurality of frame images acquired in step S100 and acquires an average image.

In step S104, the brightness conversion unit 20 converts the average image acquired in step S102 into brightness, and acquires an average brightness image showing the brightness image corresponding to the average image. Further, the brightness conversion unit 20 converts the specific frame image used when acquiring the average image in step S102 into brightness, and acquires a specific brightness image showing the brightness image corresponding to the specific frame image.

In step S106, the edge acquisition unit 22 applies a spatial filter to the average brightness image acquired in step S104 to acquire a first edge image representing the edge intensity of the average brightness image. Further, the edge acquisition unit 22 applies a spatial filter to the specific brightness image acquired in step S104 to acquire a second edge image representing the edge intensity of the specific brightness image.

In step S108, the histogram acquisition unit 23 generates an edge intensity histogram of the first edge image based on the first edge image acquired in step S106. Further, the histogram acquisition unit 23 generates an edge intensity histogram of the second edge image based on the second edge image acquired in step S106.

In step S110, the waterline candidate acquisition unit 24 obtains the y-coordinate of the peak in the edge intensity histogram of the first edge image acquired in step S108, and is parallel to the x-axis passing through the y-coordinate of the obtained peak. Set a line segment as a draft line candidate.

In step S112, the difference acquisition unit 25 acquires a difference histogram based on the edge intensity histogram of the first edge image and the edge intensity histogram of the second edge image acquired in step S108. Then, the difference acquisition unit 25 normalizes the difference histogram according to the above equation (1) and acquires the normalized difference histogram.

In step S114, the score acquisition unit 26, each water surface waterline candidates, as the slope of the normalized difference histogram at the peak, to calculate a differential value F _i of each waterline candidate according to the above formula (2). Then, the score acquisition unit 26, for each candidate waterline, calculates the score S _i in accordance with the above equation (3).

In step S116, the water surface determination unit 28 determines that the draft line candidate having the highest score is the water surface water line based on each of the scores of each draft line candidate calculated in step S114.

As described above, the water surface determination device 14 according to the first embodiment is an average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images in the image representing the water surface. A first edge image representing the edge strength of a pixel is acquired. In addition, the water surface determination device 14 acquires a second edge image representing the edge intensity of each pixel of a specific frame image in the video. Then, the water surface determination device 14 determines the region where the water surface is reflected according to the difference between the edge intensity of each pixel of the first edge image and the edge intensity of each pixel of the second edge image. This makes it possible to accurately identify the region of the water surface.

Since the embankment, etc. reflected in the river image does not change with the passage of time, even if there is no unevenness on the embankment, the edge strength of the one-frame image and the edge strength of the average image The difference between is small. Therefore, as in the present embodiment, by determining that the region where the difference between the edge strength of the average image and the edge strength of the specific frame image is large is the region where the water surface is reflected, there is no unevenness in the embankment portion. Even if there is, it is possible to accurately identify the area of the water surface.

<Second embodiment>

Next, the second embodiment will be described. The second embodiment differs from the first embodiment in that the brightness of the image is used instead of the edge strength of the image. The parts having the same configuration as that of the first embodiment are designated by the same reference numerals and the description thereof will be omitted.

In the second embodiment, the water surface covers a region where the difference between the brightness of the average image and the brightness of the specific frame image is large, based on the brightness of each pixel of the average image and the brightness of each pixel of the specific frame image. It is determined that the area is reflected.

As shown in FIG. 11, the water surface determination system 210 according to the second embodiment includes a camera 12, a water surface determination device 214, and an output device 30.

As shown in FIG. 11, the water surface determination device 214 includes an image storage unit 16, an average image acquisition unit 18, a brightness conversion unit 20, a histogram acquisition unit 223, a waterline candidate acquisition unit 24, and a difference acquisition unit 25. A score acquisition unit 26 and a water surface determination unit 28 are provided.

The histogram acquisition unit 223 represents the total brightness of each pixel in the horizontal direction corresponding to each position in the vertical direction of the average brightness image with respect to each position in the vertical direction based on the average brightness image obtained by the brightness conversion unit 20. Generate a lightness intensity histogram. The lightness intensity histogram of the average lightness image represents the lightness of each region of the average lightness image. Further, the histogram acquisition unit 223 is the sum of the brightness of each pixel in the horizontal direction corresponding to each position in the vertical direction of the specific brightness image with respect to each position in the vertical direction based on the specific brightness image obtained by the brightness conversion unit 20. Generates a brightness intensity histogram that represents. The brightness intensity histogram of the specific brightness image represents the brightness of each region of the specific brightness image.

The water surface determination device 14 of the second embodiment can be realized by, for example, the computer 50 shown in FIG. The storage unit 53 of the computer 50 can be realized by an HDD, an SSD, a flash memory, or the like. The storage unit 53 as a storage medium stores a water surface determination program 260 for causing the computer 50 to function as the water surface determination device 214. The water surface determination program 260 includes an average image acquisition process 61, a brightness conversion process 62, a histogram acquisition process 264, a waterline candidate acquisition process 65, a difference acquisition process 66, a score acquisition process 67, and a water surface determination process 68. Have. Information constituting the image storage unit 16 is stored in the image storage area 69.

The CPU 51 reads the water surface determination program 260 from the storage unit 53, expands it into the memory 52, and sequentially executes the processes included in the water surface determination program 260. The CPU 51 operates as the histogram acquisition unit 223 shown in FIG. 11 by executing the histogram acquisition process 264. As a result, the computer 50 that has executed the water surface determination program 260 functions as the water surface determination device 214. The CPU 51 that executes the software water surface determination program 260 is hardware.

The function realized by the water surface determination program 260 can also be realized by, for example, a semiconductor integrated circuit, more specifically, an ASIC or the like.

Next, the operation of the water surface determination device 214 according to the second embodiment will be described. The water surface determination device 214 executes the water surface determination processing routine shown in FIG. 13 when each frame image of the river image starts to be stored in the image storage unit 16.

Steps S100 to S104 and steps S110 to S116 are executed in the same manner as in the first embodiment.

In step S208, the histogram acquisition unit 223 generates a brightness intensity histogram of the average brightness image based on the average brightness image obtained in step S104. Further, the histogram acquisition unit 223 generates a brightness intensity histogram of the specific brightness image based on the specific brightness image obtained in step S104.

As described above, the water surface determination device 214 according to the second embodiment acquires an average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images in the image representing the water surface. do. Then, the area where the water surface is reflected is determined according to the difference between the brightness of each pixel of the average image and the brightness of each pixel of a specific frame image in the image. This makes it possible to accurately identify the region of the water surface. Further, since the water surface region can be specified without extracting the edge, the water surface region can be specified by a simple process.

In the above description, the mode in which each program is stored (installed) in the storage unit in advance has been described, but the present invention is not limited to this. The program according to the disclosed technology can also be provided in a form recorded on a recording medium such as a CD-ROM, a DVD-ROM, or a USB memory.

All documents, patent applications and technical standards described herein are to the same extent as if the individual documents, patent applications and technical standards were specifically and individually stated to be incorporated by reference. Incorporated by reference in the document.

Next, a modification of each embodiment will be described.

In each of the above embodiments, the case of determining the waterline representing the boundary of the area where the water surface is reflected has been described as an example, but the present invention is not limited to this. For example, the area where the water surface is reflected may be determined. In this case, for example, in the first embodiment, the difference between the edge strength of the average image and the edge strength of the specific frame image is calculated for each region in the image. Then, a region where the calculated difference is equal to or greater than a predetermined threshold value is determined to be a region where the water surface is reflected. Further, in the second embodiment, the difference between the brightness of the average image and the brightness of the specific frame image is calculated for each area of the image. Then, a region where the calculated difference is equal to or greater than a predetermined threshold value is determined to be a region where the water surface is reflected.

Further, in the first embodiment, in the example shown in FIG. 6, a case where both the vertical edge and the horizontal edge are extracted will be described as an example, and in the example shown in FIG. 7, the vertical edge will be described. The case where the edge is extracted has been described as an example, but the present invention is not limited to this. As the edge strength, an edge in any direction may be used. Further, as the edge image, a binarized image may be used.

Further, in each of the above embodiments, the case where the specific frame image is one frame has been described as an example, but the present invention is not limited to this, and a plurality of frame images may be used as the specific frame image. When a plurality of frame images are used as a specific frame image, for example, an image obtained by averaging the plurality of frame images can be used as the specific frame image. The number of specific frame images is sufficiently smaller than the plurality of frame images used for the average image.

The following additional notes will be further disclosed with respect to each of the above embodiments.

(Appendix 1)
A first edge image representing the edge strength of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface is acquired.
A second edge image representing the edge intensity of each pixel of a specific frame image in the video is acquired.
A region in which the difference between the edge strength of each pixel of the first edge image and the edge strength of each pixel of the second edge image is equal to or greater than a predetermined threshold value is determined to be a region where the water surface is reflected.
A water level determination program that allows a computer to perform processing.

(Appendix 2)
The brightness of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface, and the brightness of each pixel of the specific frame image in the image. A region in which the difference between the two is equal to or greater than a predetermined threshold is determined to be a region in which the water surface is reflected.
A water level determination program that allows a computer to perform processing.

(Appendix 3)
The water surface of the river is reflected below the frame image of the image in the vertical direction.
A difference histogram representing the difference between the average image and the specific frame image, and the difference histogram representing the sum of the differences of each pixel in the horizontal direction corresponding to each position in the vertical direction of the frame image. Generate and
The vertical position corresponding to the peak of the difference histogram is set as a candidate for the water surface waterline, and a score corresponding to the slope of the difference histogram at the peak is calculated for each candidate for the water surface waterline, and the score is used as the score. Based on this, the waterline of the water surface is determined.
The water surface determination program according to claim 1 or 2.

(Appendix 4)
A first edge image representing the edge strength of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface is acquired.
A second edge image representing the edge intensity of each pixel of a specific frame image in the video is acquired.
A region in which the difference between the edge strength of each pixel of the first edge image and the edge strength of each pixel of the second edge image is equal to or greater than a predetermined threshold value is determined to be a region where the water surface is reflected.
A water level determination method that causes a computer to perform processing.

(Appendix 5)
The brightness of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface, and the brightness of each pixel of the specific frame image in the image. A region in which the difference between the two is equal to or greater than a predetermined threshold is determined to be a region in which the water surface is reflected.
A water level determination method that causes a computer to perform processing.

(Appendix 6)
The water surface of the river is reflected below the frame image of the image in the vertical direction.
A difference histogram representing the difference between the average image and the specific frame image, and the difference histogram representing the sum of the differences of each pixel in the horizontal direction corresponding to each position in the vertical direction of the frame image. Generate and
The vertical position corresponding to the peak of the difference histogram is set as a candidate for the water surface waterline, and a score corresponding to the slope of the difference histogram at the peak is calculated for each candidate for the water surface waterline, and the score is used as the score. Based on this, the waterline of the water surface is determined.
The water surface determination method according to claim 1 or 2.

(Appendix 7)
A first edge image representing the edge strength of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface is acquired, and the first edge image representing the edge strength of each pixel is acquired. An edge acquisition unit that acquires a second edge image representing the edge strength of each pixel of a specific frame image of
A water surface determination in which a region where the difference between the edge strength of each pixel of the first edge image and the edge strength of each pixel of the second edge image is equal to or greater than a predetermined threshold value is determined to be a region where the water surface is reflected. Department and
A water level determination device.

(Appendix 8)
The brightness of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface, and the brightness of each pixel of the specific frame image in the image. A water surface determination unit that determines that a region in which the difference between the two is equal to or greater than a predetermined threshold is a region in which the water surface is reflected.
A water level determination device.

(Appendix 9)
The water surface of the river is reflected below the frame image of the image in the vertical direction.
A difference histogram representing the difference between the average image and the specific frame image, and the difference histogram representing the sum of the differences of each pixel in the horizontal direction corresponding to each position in the vertical direction of the frame image. The difference acquisition part to be generated and
A score acquisition unit that calculates a score according to the slope of the difference histogram at the peak for each candidate of the water surface water line with the vertical position corresponding to the peak of the difference histogram as a candidate for the water surface water line. With more
The water surface determination unit determines the waterline of the water surface based on the score.
The water surface determination device according to claim 1 or 2.

(Appendix 10)
A first edge image representing the edge strength of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface is acquired.
A second edge image representing the edge intensity of each pixel of a specific frame image in the video is acquired.
A region in which the difference between the edge strength of each pixel of the first edge image and the edge strength of each pixel of the second edge image is equal to or greater than a predetermined threshold value is determined to be a region where the water surface is reflected.
A storage medium that stores a water surface determination program for causing a computer to execute processing.

(Appendix 11)
The brightness of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface, and the brightness of each pixel of the specific frame image in the image. A region in which the difference between the two is equal to or greater than a predetermined threshold is determined to be a region in which the water surface is reflected.
A storage medium that stores a water surface determination program for causing a computer to execute processing.

10,210 Water surface determination system 12 Camera 14,214 Water surface determination device 16 Image storage unit 18 Average image acquisition unit 20 Brightness conversion unit 22 Edge acquisition unit 23,223 Schematic acquisition unit 24 Draft line candidate acquisition unit 25 Difference acquisition unit 26 Score acquisition unit 28 Water surface determination unit 30 Output device 50 Computer 51 CPU
53 Storage unit 59 Recording medium 60, 260 Water surface determination program 61 Average image acquisition process 62 Brightness conversion process 63 Edge acquisition process 64,264 Histogram acquisition process 65 Waterline candidate acquisition process 66 Difference acquisition process 67 Score acquisition process 68 Water surface determination process

Claims (7)

A first edge image representing the edge strength of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface is acquired.
A second edge image representing the edge intensity of each pixel of a specific frame image in the video is acquired.
A region in which the difference between the edge strength of each pixel of the first edge image and the edge strength of each pixel of the second edge image is equal to or greater than a predetermined threshold value is determined to be a region where the water surface is reflected.
A water level determination program that allows a computer to perform processing. The brightness of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface, and the brightness of each pixel of the specific frame image in the image. A region in which the difference between the two is equal to or greater than a predetermined threshold is determined to be a region in which the water surface is reflected.
A water level determination program that allows a computer to perform processing. The water surface of the river is reflected below the frame image of the image in the vertical direction.
A difference histogram representing the difference between the average image and the specific frame image, and the difference histogram representing the sum of the differences of each pixel in the horizontal direction corresponding to each position in the vertical direction of the frame image. Generate and
The vertical position corresponding to the peak of the difference histogram is set as a candidate for the water surface waterline, and a score corresponding to the slope of the difference histogram at the peak is calculated for each candidate for the water surface waterline, and the score is used as the score. Based on this, the waterline of the water surface is determined.
The water surface determination program according to claim 1 or 2. A first edge image representing the edge strength of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface is acquired.
A second edge image representing the edge intensity of each pixel of a specific frame image in the video is acquired.
A region in which the difference between the edge strength of each pixel of the first edge image and the edge strength of each pixel of the second edge image is equal to or greater than a predetermined threshold value is determined to be a region where the water surface is reflected.
A water level determination method that causes a computer to perform processing. The brightness of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface, and the brightness of each pixel of the specific frame image in the image. A region in which the difference between the two is equal to or greater than a predetermined threshold is determined to be a region in which the water surface is reflected.
A water level determination method that causes a computer to perform processing. A first edge image representing the edge strength of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface is acquired, and the first edge image representing the edge strength of each pixel is acquired. An edge acquisition unit that acquires a second edge image representing the edge strength of each pixel of a specific frame image of
A water surface determination in which a region where the difference between the edge strength of each pixel of the first edge image and the edge strength of each pixel of the second edge image is equal to or greater than a predetermined threshold value is determined to be a region where the water surface is reflected. Department and
A water level determination device. The brightness of each pixel of the average image obtained by averaging the pixel values of the corresponding pixels in each of the plurality of frame images of the image representing the water surface, and the brightness of each pixel of the specific frame image in the image. A water surface determination unit that determines that a region in which the difference between the two is equal to or greater than a predetermined threshold is a region in which the water surface is reflected.
A water level determination device.

Images