JPH0777451A - Water level measuring device - Google Patents

Abstract

PURPOSE:To provide a water level measuring device measuring the water level of a river or the like via image processing, not required to be installed in the stream of the river, simple to install and maintain, and not broken at the time of a heavy rain. CONSTITUTION:The same water surface is photographed from different directions by the first and second television cameras arranged on a straight line at the uniform distance from the water surface to be measured. A field angle difference is generated between the image pickup patterns of the first and second images by the water level of the water surface, thus the water level is calculated. Feature points (waves or a floating object) on the water surface are acquired by a correlator to extract the field angle difference between the first and second images.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a water level measuring device for measuring the water level of a river or the like by image processing.

[0002]

2. Description of the Related Art Conventionally, there is a float type water level measuring device of this type. This is because the float floated on the surface of the water moves up and down as the surface of the water moves up and down. Therefore, there are problems in accuracy, maintenance, etc. due to the flow of the river and the adhesion of dust and the like. Therefore, using a TV camera,
A means to measure the water level by image processing was considered. A conventional example is described in JP-A-3-170029. Referring to FIG. 10, the water surface in the observation well is photographed by the TV camera, and the water surface image is output. Based on the size of the water surface image in the screen image of a certain area obtained by the TV camera, from the observation point of the TV camera to the water surface. To find the distance. Another conventional example is Japanese Patent Laid-Open No.
No. 3-66410. Referring to FIG. 11, the target object with the memory is arranged so as to project from the water surface of the river. This target is photographed by a TV camera and the water level is calculated from the video signal.

[0003]

However, in these conventional water level measuring devices, since the device is installed in the flow of the river, the construction cost is high, it is easily damaged during heavy rain, etc., and the maintenance is also difficult. There was a problem. The present invention has been made in view of such circumstances, and an object of the present invention is to solve the above-mentioned drawbacks by non-contact telemetry.

[0004]

In order to solve the above problems, in the water level measuring device of the present invention, as shown in FIG. 3, the water level measuring device is arranged on a straight line that is equidistant from the water surface to be measured.
The same water surface is shot with two TV cameras. The water level is calculated by causing a difference in angle of view between the image pickup patterns of the two TV camera images due to the water level on the water surface. The angle of view difference extraction between the two TV camera images is performed by capturing the characteristic points (waves, floating substances, etc.) on the water surface by image processing. In particular,~
It was configured with. The first to image the water surface to be measured from different directions
And a second image pickup means (television camera), a storage means for storing the first and second images taken by the first and second image pickup means, and a degree of similarity of part of each of the first and second images. And detecting a similar region (feature point) in the first image and a similar region (feature point) in the second image that are most similar to each other, at least similarity in the first image Calculation means for calculating the water level of the water surface to be measured using one of the position of the region and the position of the similar region in the second image.

[0005]

According to the water level measuring device configured as described above,
When the first and second imaging means are arranged on a straight line that is equidistant from the water surface to be measured, the water level is obtained by the relational expression shown in FIG. The number of horizontal pixels of the first image pickup means is S, the horizontal angle of view of the first image pickup means is 2φ, the interval between the first image pickup means and the second image pickup means is 2L, and the horizontal angle of the first image pickup means is θ. Then, the vertical bisector O of the first image pickup means and the second image pickup means
When the point Pn similar region (feature point) on −M is at the n-th pixel from the left of the captured image of the first imaging unit, the relational expression OPn = L · tan λ = L · tan (θ + φ-2φ · n / S ) ... (1) is established. Further, OPs is obtained in advance at the reference position. The water level is obtained from the water level at that time, OPs, and OPn obtained from Equation 1 (the relationship is shown in FIG. 8).

[0006]

[First Embodiment]

(Explanation of Equipment Installation Example) FIG. 5 is an external view showing an example in which the present invention is installed. As shown in FIG.
A TV camera is installed on a tower, which is installed on the side of the road separating the bank from the surface of the river. In addition to the TV camera, it is equipped with a floodlight so that the water level can be measured even at night. The installation angles of the two TV cameras are adjusted so that the same water surface is photographed. Also, the image output of the TV camera is transmitted to the processing device installed in another place,
The water level is obtained. That is, remote image monitoring of the water surface is realized.

(Description of Configuration) FIG. 1 is a block diagram showing a first embodiment of the present invention. The first TV camera 1 and the second TV camera 2 arranged on a straight line that is equidistant from the water surface photograph the water surface of the river. The image signals are converted into digital signals by the first A / D converter 3 and the second A / D converter 4, respectively. The digital signals are stored in the first image memory 5 and the second image memory 6, respectively. The stored image data is divided by the first and second image dividing means 7 and 8. Correlation is performed by the correlator 9 in order to compare the similarities of the respective divided data. A feature point is extracted by the extraction circuit 10 from the correlated data. The pixel position of the extracted feature point is calculated by the pixel number calculation means 11. The water level calculation means 12 calculates the water level based on the pixel position and the reference data preset by the reference value setting means 13.

It is also possible to share the A / D converter by shifting the photographing times of the first television camera 1 and the second television camera 2. Time difference is 1 frame (1/30
However, there is no difference in the image of the water surface to be measured. There is also a configuration in which one TV camera is used in a time division manner. In other words, it is composed of optical means for guiding images of almost the same water surface to the television camera from different directions at substantially the same time.

(Explanation of Processing) FIG. 2 is a flow chart for explaining the processing of the embodiment. Image signals are input from the first TV camera 1 and the second TV camera 2 (a). The screen is divided into small regions (9 pixels × 9 pixels in this embodiment) along the horizontal center line (b).
The correlation coefficient between the corresponding small areas of the images of the first TV camera 1 and the second TV camera 2 is calculated (c). A small area having the maximum calculated correlation coefficient is extracted (d). The number of pixels from the left in the image of the first television camera 1 of the corresponding maximum correlation coefficient is calculated (e). The distance from the first television camera 1 to the water surface is calculated by the equation 1 (f). The vertical distance from the first television camera 1 to the surface of the water is calculated by Equation 1 based on the distance and the reference data preset by the reference value setting means 13 (g). The water level to be obtained is calculated from the vertical distance and the water level at the reference position measured in advance (h). The vertical distance determined by (g) is shown in FIG.
As shown in (b), it can also be obtained by hn = OPn · cosσ. Here, σ indicates the elevation angle of the television camera.

(Explanation of Experimental Example) FIG. 6 is a block diagram showing the arrangement of the first embodiment of the present invention, FIG. 7 is a diagram showing an image example, and FIG.
FIG. 6 is a vertical arrow view for explaining calculation. With the configuration of FIG. 6, water is filled in the water tank, and the water tank is photographed by the first TV camera 1 and the second TV camera 2. By changing the water level of the tank, the first TV camera 1 and the second TV camera 2
The displacement amount is calculated from the angle of view of the captured image. FIG. 6A is a horizontal arrow view, and FIG. 6B is a vertical arrow view. The lens of the first TV camera 1 and the second TV camera 2 is set to f = 50 m.
m, θ = 79 °, φ = 5 °, L = 100c in FIG.
m. In addition, a float was floated on the water surface as a floating material that was a feature point. 7A shows an image when the water level is 80 cm, FIG. 7B shows an image when the water level is 60 cm, and FIG. 7C shows an image when the water level is 35 cm. Show. Here, the X mark on the screen indicates a position that is a feature point.

In FIG. 7A, the position of the characteristic point of the first television camera is the 367th pixel. From equation 1, OP367 =
427 cm is calculated. In FIG. 7B, the position of the characteristic point of the first television camera is the 256th pixel. From Expression 1, OP256 = 512 cm is calculated. In FIG. 7C, the position of the feature point of the first television camera is the 153rd pixel. From Expression 1, OP153 = 632 cm is calculated.
Here, a means for calculating another water level with the water level of 60 cm as a reference position will be described with reference to FIG. The installation height (point O) of the TV camera from the river bottom is 174c from FIG. 6 (b).
m. Therefore, hc = 114 cm. Figure 7
The calculation of the water level in (a) will be described. Ha corresponding to P367
From the formula of 114/513 = ha / 427, ha = 9
5 cm is required. The amount of displacement from the reference position is 114
-95 = 19 cm. The water level is 60 + 19 = 79c
m, and the resolution is 1 cm. Calculation of the water level in FIG. 7C will be described. Hc corresponding to P153 is 114
From the formula of / 513 = ha / 632, hc = 140 cm is obtained. The amount of displacement from the reference position is 114-140
= -26 cm. The water level is 60-26 = 34 cm, and the resolution is 1 cm.

(Other Experimental Example) In this experimental example, the lenses of the first television camera 1 and the second television camera 2 are f = 10.
0 mm, θ = 84.3 ° in FIG. 4, φ = 2.5 °,
L = 300 cm. With this configuration, the measurement range is
6.95 m to 17.89 m from the TV camera installation surface, and the resolution is 0.9 cm to 8.8 cm. The reference positions at this time were OPs = 30 m and hs = 10 m. that's all,
Although the digital processing has been described, the camera, the correlation processing and the like are analog type (for example, Japanese Patent Publication No. 60-343 by the same applicant.
It is needless to say that it is possible even with the publication No. 17 “TV image analog correlation device”).

[Second Embodiment] FIG. 9 is a conceptual diagram showing a second embodiment of the present invention. The second embodiment is an example in which the object is not equidistant from the first TV camera 1 and the second TV camera 2. As shown in FIG. 9, it is assumed that the first TV camera 1 detects an object (a feature point obtained by the correlation process) at the a pixel position and the second TV camera 2 detects the b pixel position. The first TV camera 1 and the second TV camera 2 each have a horizontal angle of view 2φ, an installation horizontal angle θ, and a
Θa is obtained from the value of, and θb is obtained from the value of b. According to the principle of triangulation, θa, θb, 2L (first TV camera 1
And the distance between the second TV cameras 2), the distance D to the object is obtained.

[Third Embodiment] In the first embodiment, as a means for obtaining a feature point, the feature area is divided into small regions of 9 pixels × 9 pixels along the horizontal center line. Compute a correlation coefficient between corresponding small areas of the image. A small area having the largest correlation coefficient is extracted. On the other hand, in the third embodiment, as a means for obtaining the feature points, an arbitrary portion of the screen is marked with X.
Divide into a middle area of pixels × X pixels. Compute the correlation coefficient between corresponding middle regions of the image. The middle area having the maximum calculated correlation coefficient is extracted. The extracted middle region is Y (X>
Y) pixel × Y pixel small area (for example, 9 pixels × 9 pixels)
Split into. Compute a correlation coefficient between corresponding small areas of the image. A small area having the largest correlation coefficient is extracted. Further, by moving the pixel area of the small area in units of one pixel and extracting the maximum pixel position, the distance can be calculated with the accuracy equivalent to one pixel.

Although the correlator is used as the similarity detection means, the invention is not limited to this. For example, two patterns are superposed and a difference is obtained at each position. Square the difference and add together in the small area. The smaller this value is, the more similar the two patterns are. In addition, although an example in which the small region has 9 pixels × 9 pixels is shown, the number of pixels is determined by the distance from the television camera to the water surface to be measured, the size of the wave serving as the characteristic point, and the like. In addition, the television camera may be used for river monitoring. In that case, the observer can remotely set the number of pixels in the small area when monitoring the television screen and measuring the water level.

[0016]

As described above, according to the present invention, the same water surface is photographed by two television cameras arranged on a straight line equidistant from the water surface to be measured. 2 depending on this water level
The water level is calculated based on the difference in the angle of view between the image pickup patterns of the television camera images. The angle of view difference extraction between the two TV camera images is performed by capturing the characteristic points (waves, floating substances, etc.) on the water surface by image processing. Therefore, it is not necessary to install the device in the flow of the river, the construction and the maintenance are easy, and it is not damaged by heavy rain.

[Brief description of drawings]

FIG. 1 is a block diagram showing a first embodiment of the present invention.

FIG. 2 is a flow chart showing a first embodiment of the present invention.

FIG. 3 is a diagram for explaining the measurement principle of the present invention.

FIG. 4 is a diagram illustrating a relational expression of the measurement principle of the present invention.

FIG. 5 is an external view showing an example in which the present invention is installed.

FIG. 6 is a configuration diagram showing an arrangement of a first embodiment of the present invention.

FIG. 7 is a diagram showing an image example according to the first embodiment of the present invention.

FIG. 8 is a vertical arrow view for explaining calculation in the first embodiment of the present invention.

FIG. 9 is a diagram for explaining the measurement principle of the second embodiment of the present invention.

FIG. 10 is a configuration diagram showing a conventional water level measuring device.

FIG. 11 is a configuration diagram showing another conventional water level measuring device.

[Explanation of symbols]

1 ... First imaging means (first television camera), 2 ... Second
Image pickup means (first TV camera), 3 ... First A / D
Converter, 4 ... second A / D converter, 5 ... storage means (first
Image memory), 6 ... storage means (second image memory),
7 ... Detecting means (first image dividing means), 8 ... Detecting means (second image dividing means), 9 ... Detecting means (correlator), 1
0 ... Detection means (extraction circuit), 11 ... Calculation means (pixel number calculation means), 12 ... Calculation means (water level calculation means), 13 ... Reference value setting means.

Claims (1)

[Claims] 1. A first and second imaging means (1, 2) for photographing the water surface to be measured from different directions, and first and second imaging means for photographing by the first and second imaging means. The storage means (5, 6) for storing the image is compared with the similarity of a part of each of the first and second images, and the most similar region in the first image and the second similarity region in the second image are compared. A detection means (7, 8, 9, 10) for detecting a similar region in the image, and at least one of the position of the similar region in the first image and the position of the similar region in the second image. A water level measuring device provided with a calculation means (11, 12) for calculating the water level of the water surface to be measured by using it.