JP2006194705A - Surface profile measurement method under liquid and its system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a surface profile measurement method under liquid and its system in which compensation for determining the depth from a liquid surface to an object is considered by photographing position information for an image and the refractive index of the liquid. <P>SOLUTION: The surface profile measurement method under liquid comprises a data creation section 1 which performs digital mapping on a stereoscopic image using a digital mapping device and creates data on position information 2 of an image photographing device A placed on the left, position information 3 of an image photographing device B placed on the right, and a DEM value 4 and a computer 10 comprising a refractive index value input section 11, a calculation section 12 for calculating lateral liquid surface reaching coordinates, a lateral incident angle, and a lateral refraction angle, a two-dimension/three-dimension decision section 13, a two-dimensional true value calculation section 14, a decision section 15 for the three-dimensional incident angles of the two image photographing devices, a calculation section 16 for the case in which the three-dimensional incident angles are the same, a calculation section 17 for the case in which the three-dimensional incident angles are different, a creation section 18 for creating the surface profile of the object under liquid on the basis of the calculated two-dimensional and three-dimensional true values, and a storage section 19 for storing information of above-described sections. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to a method for measuring a surface shape under a liquid and a system thereof capable of measuring the shape of the surface of an object under the liquid using a stereo image.

  To measure the shape of the surface of an object, there are a method using a laser beam and a method using an ultrasonic wave, but a method using a stereo image is particularly popular as a method for easily grasping the shape of an object in a plane. It has been.

One of the most effective uses of stereo images is digital photogrammetry, and land topographic maps have been created by mapping stereo pairs of aerial photographs using a digital mapper (see the following patents). References 1 and 2). In addition to creating topographic maps, stereo pair images are used for distance measurement (see Patent Document 3 below), fundus three-dimensional data acquisition (see Patent Document 4 below), and the like.
Japanese Patent Laid-Open No. 8-210851 JP-A-9-069148 JP 2000-356514 A JP-A-8-000567

  However, in order to measure the surface shape of an object under the liquid such as the sea bottom using a stereo image taken from the air, it is necessary to correct the ratio of being viewed shallow by the refraction effect of the liquid.

  In view of the above situation, the present invention provides a surface shape measurement method and system under liquid that takes into account correction for determining the depth from the surface of the liquid to the object using image position information and the refractive index of the liquid. The purpose is to provide.

In order to achieve the above object, the present invention provides
[1] A method for measuring a surface shape under a liquid in which digital mapping is performed on a stereo image using a digital plotter, the step of digitizing the stereo image, and a reference point in a predetermined arrangement in an analysis range Setting a group and storing a plane coordinate value of each reference point; measuring and calculating an elevation of each pixel using the position and elevation value of the reference point group in the analysis range; and The method includes a step of calculating a correction coefficient that corrects a ratio of being viewed shallow by the refractive index and the image photographing position information, and a step of obtaining a true depth of an object under the liquid using the correction coefficient.

  [2] A surface shape measurement system under a liquid that performs digital mapping on a stereo image using a digital plotter, and performs digital mapping on the stereo image using a digital plotter on the left side. The position information of the image capturing device A located, the position information of the image capturing device B positioned on the right side, the data generator for generating each data of the DEM value, the calculation of the left and right water surface arrival coordinates, the left and right incident angles, and the left and right refraction angles Two-dimensional when the target point is on a straight line connecting the point directly below the two image capturing devices, a calculating unit for performing the imaging, a determining unit for the photographing position, an input unit for the refractive index value for inputting the refractive index, The difference between the incident angles of the two image capturing devices in the case of 3D, and the 2D / 3D discriminator when the other times are 3D, the arithmetic unit for calculating the 2D true value, A discriminator for discriminating the discriminator and An arithmetic unit that calculates a true value when the incident angles are determined to be equal, an arithmetic unit that calculates a true value when the incident angles are determined to be different, and the calculated two-dimensional and three-dimensional true values The computer includes a generation unit that generates the surface shape of the object under the liquid based on the value, and a storage unit that stores information on each unit described above.

  [3] The surface shape measurement system under the liquid as described in [2] above, further comprising an output unit that displays information on the measured surface shape of the object under the liquid on a screen.

  According to the present invention, it is possible to accurately and inexpensively and easily create surface shape data of an object under a liquid, such as a topographic map of the seabed, and measure the surface shape of the object under the liquid. .

  A surface shape measurement method under a liquid in which digital mapping is performed on a stereo image using a digital plotter, the step of digitizing the stereo image, and setting a reference point group in a predetermined arrangement in the analysis range Storing the plane coordinate value of each reference point, measuring and calculating the altitude of each pixel using the position and altitude value of the reference point group in the analysis range, and the refractive index of the liquid, A step of calculating a correction coefficient that corrects a ratio of the image captured by the image capturing position information, and a step of obtaining a true depth of an object under the liquid using the correction coefficient. Therefore, it is possible to accurately and inexpensively and easily create the surface shape data of the object under the liquid, and to accurately measure the surface shape of the object under the liquid.

  Hereinafter, embodiments of the present invention will be described in detail.

  Here, as an example, the creation of a water depth map using an aerial photograph with a target as a shallow sea area will be described as one of the most effective usage methods of the present invention.

  Hereinafter, a method for creating surface shape data under a liquid according to the present invention will be described in detail.

  Conventionally, a digital elevation model (DEM) has been created by performing stereoscopic viewing (stereo matching) using aerial photographs of two stereo pairs. Commercially available digital stereo plotting machines (Digital Photoworking Workstations: DPW) are widely used for creating elevation maps. In shallow water areas with high transparency of seawater such as coral reefs, the state of the seabed can be distinguished from aerial photographs, so it is expected that DEM can be created using stereo matching in the same way as the ground surface. However, since the DEM value calculated by DPW stereo matching shows a value shallower than the actual water depth due to the refraction effect of light on the water surface, correction thereof is indispensable. However, until now, there has not been a theoretical clarification of how much the depth of water looks shallow depending on the positional relationship between the main points and observation points of aerial photographs.

  In the present invention, this has been theoretically clarified and a computer program for correcting the light refraction effect has been developed. And the DEM value by DPW is corrected to a value close to the actual water depth by the water depth correction program. Actually, a DEM obtained from an aerial photograph (1 / 10,000 photographs taken in 1995 of Shiraho Coral Reef, Ishigaki Island, Okinawa Prefecture) was corrected by the depth correction program. Then, the accuracy was verified by comparing the DEM before and after correction with the actually measured water depth and calculating the average of the errors.

  First, the theory of light refraction effect and water depth correction will be explained separately for cases where the target point is on a straight line connecting the points directly under the two image capturing devices (two dimensions) and other cases (three dimensions). Then, it is considered how shallow it looks from the actual water depth, and the ratio of the shallow water is calculated. In addition, the accuracy of the theory is verified using DEM values obtained from actual aerial photographs.

[1] Refraction of light Underwater, a phenomenon occurs in which an object and the bottom of the water appear to float (shallow). This is due to light refraction.

  FIG. 1 is a diagram illustrating the principle.

  In order to think geometrically about how an underwater object (here, the bottom of the water) looks, you must consider how the refracted light forms an image.

  In FIG. 1, the light emitted from the water bottom A <b> 1 directly above reaches the eyes by going straight from A <b> 1 → P → R. However, the light directed slightly upward is refracted on the water surface Q and reaches S. However, the image shown in the retina of the eye (here, an aerial photograph) is imaged as if light travels straight on the extension line of the line segment SQ because this refraction is not considered, and the straight line A1R and the extension line of the line segment SQ It appears that there is an object (water bottom) at the intersection A2. This is the principle that looks shallow.

In general, the refractive index n of water relative to air is refractive index n = sin θn / sin θk≈1.33.
(Θn: incident angle, θk: refraction angle)
It is represented by

[2] Water depth correction theory As described above, when a DEM of the coral reef seabed is created by aerial photogrammetry using DPW, the result is shallower than the actual result due to the effect of light refraction. Therefore, in the present invention, the correction theory is established by theoretically clarifying that the ratio of the shallow-looking ratio changes depending on the positional relationship between the main point and the observation point of the aerial photograph, and created by DPW by the correction program based on this correction theory. It tries to make the DEM value closer to the actual value.

  FIG. 2 shows a bird's-eye view of an aerial photogrammetry as an example, and FIG. 3 shows a three-dimensional view.

  In this figure, two airplanes fly at a predetermined altitude, and are photographed so that the overlap portion between photographs is 60% at regular intervals according to the flight speed.

  The photographing altitude can be arbitrarily set. In the case of an aerial photograph, a photograph taken at an altitude of 3,000 m to 10,000 m is generally used.

  The aerial photograph acquired in this way is converted into digital data by a scanner. At this time, it is desirable to use a resolution of about 10 cm.

  FIG. 3 shows a state in which light rays from the right image capturing device coordinates and the left image capturing device coordinates pass through the right water surface arrival coordinates and the left water surface arrival coordinates, respectively, and catch an object on the seabed. At this time, the light is refracted on the water surface.

  2 and 3, X, Y, and Z coordinates are taken, the right image photographing device coordinates are (Xr, Yr, Zr), the left image photographing device coordinates are (Xl, Yl, Zl), and the right water surface. The arrival coordinates are (Xwr, Ywr, Zwr), the left water surface arrival coordinates are (Xwl, Ywl, Zwl), the DEM value by DPW is (Xd, Yd, Zd), and the true value (theoretical value) is (Xh, Yh, Zh) ).

[2-1] When the target point is on a straight line connecting the points directly below the two image capturing devices (two-dimensional, on the straight line a in FIG. 2)
In this case, the light trajectory can be considered on a two-dimensional plane. In this case, it becomes as shown in FIG.

  In FIG. 4, the height (elevation) of the water surface at the time of photo acquisition is Z, and the refractive index of water with respect to air is nw (= 1.33). Further, since the photo orientation is finished by DPW, it is assumed that the image capturing apparatus coordinates and the image capturing altitude are known and the DEM value has already been measured.

  First, the water surface arrival coordinates are obtained from the intersection of the straight line connecting the DEM value and the image capturing device coordinates and the water surface Z.

Xwr = [(Z−Zd) × (Xr−Xd)] / [(Zr−Zd)] + Xd
Xwl = [(Z-Zd) * (X1-Xd)] / [(Z1-Zd)] + Xd
Ywr = [(Z−Zd) × (Yr−Yd)] / [(Zr−Zd)] + Yd
Ywl = [(Z-Zd) * (Y1-Yd)] / [(Z1-Zd)] + Yd
Zwr = Z
Zwl = Z
Next, the square of the incident angle sin is obtained from the water surface arrival coordinates and the image capturing apparatus coordinates.

sin ² nr = [(Yr−Ywr) ² ] / [(Yr−Ywr) ² + (Zr−Zwr) ² ] sin ² nl = [(Yl−Ywl) ² ] / [(Yl−Ywl) ² + ( Zl−Zwl) ² ] (nr: incident angle from right image photographing device, nl: incident angle from left image photographing device)
Further, the square of the refraction angle sin is obtained from the refractive index nw = sin θn / sin θk.

sin ² kr = sin ² nr / nw ²
sin ² kl = sin ² nl / nw ²
(Kr: refraction angle with respect to the incident angle from the right image photographing device, kl: refraction angle with respect to the incident angle from the left image photographing device)
Finally, the true value (theoretical value) is obtained from the intersection of the straight line passing the true value and the right water surface arrival coordinate and the straight line passing the true value and the left water surface arrival coordinate. First, the square of tan is obtained from the square of sin of the refraction angle obtained previously.

tan ² kr = (sin ² kr) / (1−sin ² kr)
tan ² kl = (sin ² kl) / (1−sin ² kl)
From this, the slope of the two straight lines passing through the true value and the water surface arrival coordinates in FIG.

(However, positive / negative changes depending on the position coordinates of the DEM value).

  Therefore, the true values Yh and Zh are obtained as the intersection of the following two straight lines.

Xh is a point on a straight line connecting two image capturing devices on the X and Y planes.
Xh = [(Xr-Xl) / (Yr-Yl)] (Yh-Yl) + Xl
It becomes.

  In this way, true values (Xh, Yh, Zh) can be obtained.

  Here, as can be seen from FIG. 4, when the incident angles from the two image capturing apparatuses are different, the Y coordinate (Yd) of the DEM value and the Y coordinate (Yh) of the true value deviate. That is, it appears to be shifted laterally from the actual position. This lateral displacement width is the image capturing device height 3000 m, the water surface height Z = 0.92, the right image capturing device coordinates (0,500,3000), the left image capturing device coordinates (0, -500,3000), the DEM value. What was verified as water depth (Zd) -0.9 m is shown in FIG.

  Here, the lateral displacement width was determined by Yd−Yh.

  In FIG. 5, the Y coordinate width is set to −700 to 700 because it is assumed that two aerial photographs of 1 / 10,000 (22.8 cm × 22.8 cm, 60% overlap).

  Further, in FIG. 5, there is no lateral shift at the positions of the image capturing apparatuses whose Y coordinates are −500 and 500, as shown in FIG. Because.

  As can be seen from this figure, the maximum lateral deviation width is 0.004892 m, and the error can be said to be negligibly small. From this, it can be approximated as Yh≈Yd.

[2-2] When the target point is not on a straight line connecting directly under the two image capturing devices (three-dimensional)
In the above, we considered how the refracted light forms an image when geometrically considering how the bottom of the water looks. Although the same can be considered in three dimensions, unlike in two dimensions, if the incident angles to the water surface from the two image capturing devices are different, as shown in FIG. Do not cross at one point. In other words, the DPW seems to have forcibly formed an image by stereo matching.

(A) When incident angles from the two image capturing devices are the same (when connecting images, on the straight line b in FIG. 2)
When the incident angles on the water surface from the two image capturing devices are the same, an image is formed at a point on a perpendicular passing through the midpoint of a two-dimensional straight line connecting the two devices. This point can be obtained by the same method as the intersection of two straight lines in two dimensions. In this case, the water surface arrival coordinates are
Xwr = [{(Z−Zd) × (Xr−Xd)} / (Zr−Zd)] + Xd
Xwl = [{(Z-Zd) * (X1-Xd)} / (Z1-Zd)] + Xd
Ywr = [{(Z−Zd) × (Yr−Yd)} / (Zr−Zd)] + Yd
Ywl = [{(Z−Zd) × (Zl−Zd)} / (Zl−Zd)] + Yd
Zwr = Z
Zwl = Z
3 and the square of the incident angle sin is from FIG.
sin ² nr = {(Xr−Xwr) ² + (Yr−Ywr) ² } / {(Xr−Xwr) ² + (Yr−Ywr) ² + (Zr−Zwr) ² }
... (1)
sin ² nl = {(Xl−Xwl) ² + (Yl−Ywl) ² } / {(Xl−Xwl) ² + (Yl−Ywl) ² + (Zl−Zwl) ² }
... (2)
It becomes.

  Thereby, the inclination can be obtained and the true value can be obtained as in the two-dimensional case. In this case, since it is the middle point, there is no shift in the Y direction / X direction. As a result, Xh = Xd, Yh = Yd, and Zh is

And it is required.

(B) When the incident angles from the two image capturing devices are different (when images are not formed)
Next, consider a case where the incident angles from the two image capturing apparatuses are different. In this case, since it is assumed that the light beam after refraction does not form an image, the following approximation was made.

First, in FIG. 3, the following relational expression is established from the refraction angle, the water surface arrival coordinates, and the true value. sin ² kr = {(Xwr−Xh) ² + (Ywr−Yh) ² } / {(Xwr−Xh) ² + (Ywr−Yh) ² + (Zwr−Zh) ² }
... (3)
sin ² kl = {(Xwl−Xh) ² + (Ywl−Yh) ² } / {(Xwl−Xh) ² + (Ywl−Yh) ² + (Zwl−Zh) ² }
(4)
From the above formulas (1) and (2), the refractive index nw = sin θn / sin θk,
sin ² kr = sin ² nr / nw ² (5)
sin ² kl = sin ² nl / nw ² (6)
Therefore, the following equations are established from the above-described equations (3), (5), (4), and (6).

{(Xwr−Xh) ² + (Ywr−Yh) ² } / {(Xwr−Xh) ² + (Ywr−Yh) ² + (Zwr−Zh) ² } = sin ² nr / nw ²
{(Xwl-Xh) ² + (Ywl-Yh) ² } / {(Xwl-Xh) ² + (Ywl-Yh) ² + (Zwl-Zh) ² } = sin ² nl / nw ²
Here, in the above equation, it is assumed that the lateral deviation is smaller in three dimensions than the result of the lateral deviation verification in two dimensions, and approximated as Yh≈Yd,
{(Xwr−Xh) ² + (Ywr−Yd) ² } / {(Xwr−Xh) ² + (Ywr−Yd) ² + (Zwr−Zh) ² } = sin ² nr / nw ²
{(Xwl-Xh) ² + (Ywl-Yd) ² } / {(Xwl-Xh) ² + (Ywl-Yd) ² + (Zwl-Zh) ² } = sin ² nl / nw ²
Thus, simultaneous equations hold. As a result, true values Xh and Zh are obtained.

  Based on these theories, water depth correction can be performed within a range stereo-matched based on DEM values obtained from two aerial photographs from two image capturing apparatuses.

  FIG. 7 shows an image taking device altitude of 3000 m, an average water surface height Z = 0.92 m, right image taking device coordinates (0,500,3000), left image taking device coordinates (0, −500,3000). The DEM value depth from the average sea level is -1.0 m, and the water depth distribution is shown in the X and Y planes.

  From this figure, it can be seen that the distance from the center (X = Y = 0) of the stereo image appears shallower than the actual water depth.

  It was found that at the center of the stereo image, the DEM value looks 0.6040 times shallower than the true value (theoretical value) in water depth, and 0.5582 times shallower where it appears shallowest. Further, in this case, it appears to be 0.5802 times shallower on average.

  FIG. 8 is a flowchart of the water depth correction program of the present invention.

  In this figure, first, an image photographing device position and a DEM value are input (step S1). Next, the water surface arrival coordinates, the incident angle and the refraction angle are calculated (step S2). Next, a refractive index is input (step S3). Next, it is determined whether it is two-dimensional or three-dimensional (step S4). As a result of the determination, in the case of two dimensions (step S5), the true value of the water depth is calculated (step S6). In the case of three dimensions (step S7), the incident angle from the image capturing device is determined (step S8). When the incident angles are the same (step S9) and when the incident angles are different (step S10), The true value of the water depth is determined (step S11, step S12). Thus, in any case, the true value of the water depth is determined (step S12). Based on the true value of the water depth in Step S6, Step S11, and Step S12, the surface shape of the object under the liquid is generated (determination of the true value, Step S13).

  Hereinafter, the verification test using the DEM data of the present invention will be described.

  9 and 10 show two 1 / 10,000 aerial photographs (60% overlap) taken in 1995 by Shiraho Ishigaki, Okinawa.

  Hereinafter, the accuracy of the present invention was verified using the DEM data obtained from these two aerial photographs.

  Here, the theoretical depth value obtained by correcting the DEM value by the program of the present invention and the actual measured depth value of Shiraho were compared and verified. Under these conditions, the DEM value before the water depth correction and the theoretical value after the water depth correction were compared with the measured values, respectively.

  FIG. 11 shows a comparison result between the DEM value before correction according to the present invention and the actual measurement value, and FIG. 12 shows a comparison result between the theoretical value after correction according to the present invention and the actual measurement value.

  In these figures, a straight line y (actual value) = x (calculated value) is drawn, and the closer to this straight line, the smaller the error from the actual value.

  As can be seen from FIG. 11, the DEM value (before correction) appears to be shallower than the actually measured value, and the average value of the error is 0.622528 m. On the other hand, it can be seen from FIG. 12 that the theoretical values after the correction according to the present invention are concentrated around the y = x straight line, and the average error value is 0.297963 m, which improves the accuracy.

  This is probably because the DPW DEM value, which was shallower than the actually measured value, was theoretically corrected to correct the value close to the actually measured value, that is, the water depth deeply.

  Also, all the points were not on y = x because DPW stereo matching was not successful, the state of the image capturing device (tilt of the image capturing device at the time of capturing, fluctuations in the height of the image capturing device, etc.) ) Seems to be an error caused by not taking into account.

  The results in the present invention are summarized as follows.

  (1) The theory that the bottom of water and objects look shallow in water was considered in consideration of the refractive index.

  (2) Based on the theory, a water depth correction program was completed.

  (3) As a result of verifying the accuracy of the program using the data of Shiraho, Ishigaki Island, Okinawa Prefecture, the average error before correction was 0.62258 m, but the accuracy of the water depth correction by the present invention is 0.29796 m. Improved. Further, assuming that the DEM water depth is -1.0 m and making a water depth scatter diagram (Fig. 7) in the range of the stereo image of the aerial photograph, the DEM data is taken as the water depth from the measured data, and the average is 0.58. I found that it looked twice as shallow.

  FIG. 13 is a system block for generating the surface shape of the object under the liquid based on the true water depth value of the object under the liquid.

  In this figure, 1 is a data generation unit that performs digital mapping on a stereo image using DPW to generate data, 2 is position information of the image capturing device A located on the left side, and 3 is located on the right side. Position information of the image capturing apparatus B, 4 is a DEM value, and 5 is a refractive index value. Reference numeral 10 denotes a computer. The computer 10 receives position information 2 of the image capturing apparatus A located on the left side, position information 3 of the image capturing apparatus B located on the right side, and a DEM value 4 from the data generation unit 1. Is done. The computer 10 includes a refractive index value input unit 11, left and right water surface arrival coordinates, left and right incident angles, left and right refraction angle calculation unit 12, two-dimensional / three-dimensional determination unit 13, two-dimensional true value calculation unit 14, 3 Of the angle of incidence of two image capturing devices in two dimensions, a calculating unit 16 when the three-dimensional incident angle is the same, a calculating unit 17 when the three-dimensional incident angle is different, and generation of the surface shape of the object under the liquid The unit 18 includes a storage unit 19 that stores information from each unit, provides information stored in each unit, and an output unit 20 that outputs generation information of the surface shape of the object under the liquid. Here, the flow for obtaining the true value is as shown in FIG. 8, and as an output example from the output unit, a water depth scatter diagram (see FIG. 7) can also be obtained.

In the present invention,
(1) First, a stereo image is digitized.

  (2) Next, a reference point group is set in a predetermined arrangement in the analysis range, and the plane coordinate value of each reference point is stored.

  (3) The altitude of each pixel is measured and calculated using the position and altitude value of the reference point group in the analysis range.

  Up to this point, it can be created based on digital mapping using a digital plotter.

  (4) Next, a correction coefficient for correcting the ratio of the image that is viewed shallow is calculated based on the refractive index of the liquid and the image capturing position information.

  (5) The true depth of the object under the liquid is obtained using the correction coefficient.

  The steps (4) and (5) are important processes of the present invention.

Thus, the generation of the surface shape generation under the liquid will be described with reference to FIG.
In the data generation unit 1, digital mapping is performed on a stereo image using a digital plotter (refer to the prior art), position information 2 of the image capturing apparatus A located on the left side, and position of the image capturing apparatus B located on the right side Each data of information 3 and DEM value 4 is generated. Next, these data together with the refractive index value 5 are taken into the computer 10 and the calculation unit 12 calculates the left and right water surface arrival coordinates, the left and right incident angles, and the left and right refraction angles. The value is stored in the storage unit 19. Next, the two-dimensional / three-dimensional determination unit 13 performs two-dimensional / three-dimensional determination, and stores the value in the storage unit 19. Next, the true value calculation unit 14 performs a two-dimensional true value calculation and stores the true value in the storage unit 19. In the three-dimensional case, the incident angle determination unit 15 determines whether the incident angles are the same or different, and stores the values in the storage unit 19. Next, the true value calculation unit 16 calculates the true value of the three-dimensional storage when the incident angles are the same, and stores the true value in the storage unit 19. Next, the true value calculation unit 17 calculates a three-dimensional true value when the incident angles are different, and stores the true value in the storage unit 19. Based on the two-dimensional and three-dimensional true values obtained in this way, the surface shape of the object under the liquid performs mapping, and the surface shape of the object under the liquid is generated.

  Note that the present invention is not limited to aerial photogrammetry, and can be applied universally as long as the image is captured by an image capturing apparatus.

  Further, the present invention is not limited to the above-described embodiments, and various modifications can be made based on the spirit of the present invention, and these are not excluded from the scope of the present invention.

  The liquid surface shape data creation method of the present invention can be used as an effective tool for knowing the accurate surface shape of an object under the sea surface.

It is explanatory drawing of the principle of the appearance of the object under the liquid by the refraction of light. It is a bird's-eye view which made the aerial photogrammetry concerning this invention an example. FIG. 3 is a three-dimensional view of FIG. 2. It is a two-dimensional model figure in case an object point exists on the straight line which connects directly under 2 image photographing devices. It is a figure which shows the lateral shift characteristic of a DEM value and a true Y coordinate. It is a figure which shows the mode after the refraction | bending of the light ray in the water surface in case there exists an object point other than on the straight line which connects two image photographing devices directly (three dimensions). It is a figure which shows the true depth in case the altitude of an airplane is 3000 m and the apparent depth measured by digital photogrammetry is 1 m. It is a flowchart of the water depth correction program of the present invention. It is a figure which shows the aerial photograph (the 1) of Ishigaki-jima Shiraho of Okinawa used for the verification test of this invention. It is a figure which shows the aerial photograph (the 2) of Ishigakijima Shiraho of Okinawa used for the verification test of this invention. It is a figure which shows the comparison with the DEM value before correction | amendment by this invention, and a measured value. It is a figure which shows the comparison with the theoretical value after correction | amendment by this invention, and a measured value. It is a block diagram of the surface shape measurement system under the liquid which shows the Example of this invention.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Data generation part 2 Position information of the imaging device A located in the left side 3 Position information of the imaging device B located in the right side 4 DEM value 5 Refractive index value 10 Computer 11 Refractive index value input part 12 Left and right water surface arrival coordinates, Left / right incidence angle and left / right refraction angle calculation unit 13 Two-dimensional / three-dimensional discrimination unit 14 Two-dimensional true value calculation unit 15 Incidence angle discrimination unit of two image photographing devices in three dimensions 16 Three-dimensional incidence angle is Calculation unit 17 for the same case Calculation unit 18 for different three-dimensional incident angles 18 Surface shape generation unit for object under liquid 19 Storage unit 20 for storing information from each unit and providing information stored for each unit 20 Output unit that outputs generation information of surface shape of object under liquid

Claims (3)

A surface shape measurement method under a liquid that performs digital mapping on a stereo image using a digital plotter,
Digitizing the stereo image;
A step of setting a reference point group in a predetermined arrangement in the analysis range and storing a plane coordinate value of each reference point;
Measuring and calculating the altitude of each pixel using the position and altitude value of the reference point group in the analysis range; and
Calculating a correction coefficient that corrects a ratio of being viewed shallow by the refractive index of the liquid and the image shooting position information;
Obtaining the true depth of the object under the liquid using the correction factor;
A method for measuring a surface shape under a liquid, comprising: A surface shape measurement system under a liquid that performs digital mapping on a stereo image using a digital plotter,
(A) Digital mapping is performed on the stereo image using a digital plotter, and the position information of the image photographing device A located on the left side, the position information of the image photographing device B located on the right side, and the DEM value data A data generation unit for generating
(B) a calculation unit that calculates left and right water surface arrival coordinates, left and right incident angles, and left and right refraction angles, a shooting position determination unit, a refractive index value input unit that inputs a refractive index, and two image shooting apparatuses A two-dimensional / three-dimensional discriminating unit when the target point is on a straight line connecting the points immediately below, and a three-dimensional case other than that, and a calculating unit that calculates a two-dimensional true value; A determination unit that determines whether the incident angles of the two image capturing apparatuses are three-dimensional, a calculation unit that calculates a true value when the determination unit determines that the incident angles are equal, and the incident angle is A calculation unit that calculates a true value when it is determined that the two are different from each other, a generation unit that generates a surface shape of an object under the liquid based on the calculated two-dimensional and three-dimensional true values, A surface shape under a liquid, comprising a computer having a storage unit for storing information Measurement system.   3. The surface shape measurement system under liquid according to claim 2, further comprising an output unit that displays information on the measured surface shape of the object under the liquid on a screen.