JP6789868B2 - Image processing device - Google Patents

Description

The present invention is a photographed image of a pipeline facility inspection moving body that moves in a structure extending in the longitudinal direction, for example, a pipeline facility such as a sewage pipe, acquires an image, and grasps the state of the inner surface of the pipeline facility. Regarding the image processing device of.

In recent years, the problem of aging drainage pipes has become more serious. As aging progresses, abnormalities such as cracks and breakage occur. As a result, earth and sand enter the pipe from the outside of the pipe, creating a gap in the ground, which causes an accident in which the road above the sewage pipe collapses. Therefore, there is a need for a technique for efficiently grasping abnormalities such as cracks and breakage in drainage pipes. In response to such needs, for example, Patent Document 1 describes a technique of running a small self-propelled vehicle equipped with a TV camera in a sewer pipe and monitoring the state in the pipe.

Japanese Unexamined Patent Publication No. 9-226570

In Patent Document 1, an image is acquired by a TV camera of a self-propelled vehicle placed in a sewer pipe. The so-called direct-view image taken in the longitudinal direction of the drainage pipe is generally provided as a developed view of a cylindrical pipe so that it can be seen at a glance where an abnormality has occurred in the drainage pipe. is there. If the TV camera is in the center of the cylindrical cross section and the vanishing point of the drainage channel is in the center of the image, an accurate development can be obtained by image processing based on the geometric relationship.

However, when an image is acquired by a moving body, the TV camera is not always located at the center of the cross section of the circular tube. This is noticeable when the moving body is a device such as a flying body or a floating body that is not in contact with a wall surface. In that case, in the image processing assuming that the TV camera is in the center of the cylindrical shape and the vanishing point of the drainage pipe is in the center of the image as described above, the distortion of the developed view becomes large and the quality of the developed view is improved. There was a problem that it would decrease.

If the position of the TV camera on the plane of the cylindrical cross section is known, the position information can be taken into consideration when calculating the image processing for obtaining the developed view, and the quality deterioration can be prevented. Ideally, a distance sensor that measures the distance to the inner surface of the cylinder should be installed on a moving body equipped with a TV camera. However, when the moving body is a flying body, it is not desirable to install a distance sensor because the flight time decreases due to the increase in weight. Furthermore, in the case of a sewer pipe, since there are droplets and fog inside, noise is included in the measured value of the distance sensor, and an appropriate measurement result may not be obtained.

The present invention has been made in view of this subject, and an object of the present invention is to extend from the inner surface of a structure extending in the longitudinal direction to an imaging device even if the moving body is not provided with a distance sensor. An object of the present invention is to provide an image processing device having a high-quality development drawing by obtaining distance information based on a captured image.

In the image processing apparatus according to the present invention, the first image and the second image taken by the photographing apparatus from two different points in the longitudinal direction on the inner surface of the structure extending in the longitudinal direction, and the first image from the second image, A region extraction means for extracting the second and third regions, a matching means for searching for a region in which the extracted first, second, and third regions correspond in the first image, and a first image and a first image. A position deviation calculating means for calculating the deviation of the positions of the first, second and third regions in the second image, and an imaging device from the inner surface of the structure based on the deviation of the positions of the first, second and third regions. It is characterized by including an inner surface distance ratio calculating means for calculating the ratio of the distances to and a display device for displaying the ratio of the distances from the inner surface of the structure to the photographing device numerically or in a diagram.

According to the present invention, even if the moving body equipped with the photographing device does not include the distance sensor, the distance information from the inner surface of the structure extending in the longitudinal direction can be obtained based on the photographed image. Since a distance sensor is not required, if the moving body is a flying body, it contributes to weight reduction and does not cause a decrease in flight time. Further, by using the obtained information on the distance from the inner surface, it is possible to provide a high-quality development drawing with less distortion.

It is a block diagram of each means of image processing which concerns on Example 1 of this invention. It is a schematic diagram of a direct view image. It is a schematic diagram which shows the example of the position of the 1st, 2nd and 3rd regions extracted from a photographed image. It is a schematic diagram which shows the positional relationship of the imaging apparatus which photographs the inner surface of a structure extending in the longitudinal direction from two different points. It is a schematic diagram which shows that the 1st region extracted from a captured image is at a different position in a 1st image and a 2nd image. It is a schematic diagram which shows that the 2nd region extracted from a captured image is at a different position in a 1st image and a 2nd image. It is a schematic diagram which shows that the 3rd region extracted from a captured image is at a different position in a 1st image and a 2nd image. It is a schematic diagram which shows that the position of the 1st region and the 2nd region extracted from a photographed image is at a different position in a 1st image and a 2nd image, and each has a different deviation. It is a schematic diagram which shows the optical geometric relation in the 1st image and the 2nd image. This is an example of a display screen of a display device that numerically or illustrates the ratio of the distance from the inner surface of the structure to the photographing device. It is a block diagram of each means of image processing which concerns on Example 2 of this invention. This is an example of a display screen of a display device that numerically or illustrates the distance from the inner surface of the structure to the photographing device. It is a block diagram of each means of image processing which concerns on Example 3 of this invention. It is explanatory drawing of the procedure of making the development drawing which concerns on Example 3 of this invention. It is explanatory drawing of the connection procedure of the development view which concerns on Example 3 of this invention.

The image processing device according to the present invention is a device that processes an image obtained by a moving body that moves in a structure extending in the longitudinal direction, that is, a pipeline facility and is used for checking for cracks and breakage. Pipeline facilities include, for example, sewage pipes, gas pipes, tunnels, collecting pipes, water pipes whose contents have been removed and replaced with gas, steam pipes, oil pipes, overturned pipes, drainage culverts, manholes, chimneys, etc. It is not limited to gas. Further, the structure extending in the longitudinal direction does not have to have a circular cross section, and may have a rectangular cross section.

Hereinafter, as a specific example, an embodiment in the case of targeting a sewer pipe will be described with reference to the drawings.

FIG. 1 is a block diagram of each means of image processing according to the first embodiment of the present invention. The information given is the first image 12 and the second image 10. The first image 12 and the second image 10 are direct-view images taken in the longitudinal direction in the pipe, and a schematic diagram thereof is shown in FIG. For clarity, FIG. 2 also shows the sewage 40 and the seams 46 of the structure extending in the longitudinal direction.

The second image 10 is given to the region extraction means 14. The region extraction means 14 extracts the first region 16 of the second image, the second region 18 of the second image, and the third region 19 of the second image from the second image 10. It is desirable that the first region 16 of the second image and the second region 18 of the second image are symmetrical with respect to the center of the image and are located as far apart as possible. On the other hand, it is desirable that the third region 19 of the second image is on a line intersecting at right angles from the midpoint on the line connecting the first region 16 of the second image and the second region 18 of the second image. .. For example, as shown in FIG. 3, the vicinity of the right edge and the vicinity of the left edge of the second image are used as the first region 16 of the second image and the second region 18 of the second image, and the vicinity of the upper end of the center of the image is the second. It is desirable to use it as the third region 19 of the second image. If the liquid is flowing downward like a drain and the surface features change in a short time, it is desirable not to use the vicinity of the lower end of the image as the first, second and third regions. .. It is desirable that the first region 16 of the second image, the second region 18 of the second image, and the third region 19 of the second image are all located near the edge of the second image 10. However, it does not necessarily have to be the edge when the distortion of the lens is large, or when there is blurring or out of focus. The size of the first region 16 of the second image, the second region 18 of the second image, and the third region 19 of the second image is not limited, but we want to speed up the subsequent processing. In some cases, it is desirable to set it small.

In the following, the first region 16 of the second image is near the left edge of the image, the second region 18 of the second image is near the right edge of the image, and the third region 19 of the second image is near the upper edge of the image. The case where it is set as a small area of is described.

The first region 16 of the extracted second image is sent to the matching means 20. The matching means 20 is also provided with the first image 12. It is assumed that the first image 12 is captured in a wider range than the photographing range of the second image 10. That is, the shooting position of the first image 12 is located behind the shooting position of the second image. When an image is taken by a moving body, it is common to take a picture of the front screen, so that the first image 12 is taken at a time earlier than the second image 10. FIG. 4 shows the relationship between the shooting positions of the first image 12 and the second image 10. In this figure, two photographing devices are drawn for convenience, but in reality, the positional relationship between the front and the rear when one photographing device is moved by the moving body is shown. The photographing device 42 at the position where the first image is photographed is located on the left side, that is, behind the photographing device 44 at the position where the second image is photographed.

The matching means 20 searches the first image 12 for a region in which the first region 16 of the second image best matches. In this case, the full screen may be searched, but if the difference in shooting time between the first image 12 and the second image 10 is short, the vicinity of the first region 16 of the second image is present. It is good to search for the first image 12 at the coordinates of. FIG. 5 shows an example of the search result in a schematic diagram. The extracted first region corresponds to the region 48 in the first image. The correspondence information 22 of the first region obtained as a result of the matching means 20 is sent to the position difference calculating means 26.

The position difference calculation means 26 calculates how much the first region 16 of the second image is displaced in the first image 12. This deviation is defined as a deviation in the direction of the line connecting the first region 16 of the second image and the second region 18 of the second image. The information of this deviation is sent to the inner surface distance ratio calculating means 32 as the deviation 28 of the position of the first region.

The same processing as described above is also applied to the second region 18 of the second image. FIG. 6 shows an example of the search result in a schematic diagram. The second region 18 of the extracted second image corresponds to the region 50 in the first image. Similar to the case of the first region, the correspondence information 24 of the second region and the deviation 30 of the position of the second region are obtained, and the deviation 30 of the position of the second region is sent to the inner surface distance ratio calculating means 32. ..

For the third region 19 of the second image, the deviation is obtained in a direction different from that of the first region 16 of the second image and the second region 18 of the second image. Specifically, the deviation in the direction perpendicular to the line connecting the first region 16 of the second image and the second region 18 of the second image is obtained. The information of this deviation is sent to the inner surface distance ratio calculating means 32 as the deviation 31 of the position of the third region. FIG. 7 shows an example of the search result in a schematic diagram. The third region 19 of the extracted second image corresponds to the region 51 in the first image.

FIG. 8 is a schematic diagram showing that the positions of the first region and the second region are different in the first image and the second image, and each has a deviation. An example is shown in which the deviation 28 of the position of the first region is larger than the deviation 30 of the position of the second region.

The inner surface distance ratio calculating means 32 is a distance from the inner surface of the structure 38 extending in the longitudinal direction to the photographing device based on the values of the deviation 28 of the position of the first region and the deviation 30 of the position of the second region. Calculate the ratio of. FIG. 9 is a schematic view showing the optical geometric relationship between the first image and the second image. The first region 16 of the second image and the second region 18 of the second image are also reflected on the equivalent image plane 64 on which the first image is projected. The projection positions of the first region 16 of the second image and the second region 18 of the second image on the equivalent image plane 65 on which the second image is projected are different from each other. The deviation of the position of one region is 28 and the deviation of the position of the second region is 30.

The ratio of the deviation 28 of the position of the first region and the deviation 30 of the position of the second region is based on the geometrical positional relationship, and is a coefficient related to the angle of view of the photographing device, the position where the first image is taken and the second. It can be expressed by the equation (1) using the difference 62 of the position where the image is taken, the distance 58 from the inner surface on the left side, and the distance 60 from the inner surface on the right side. However, dL: deviation of the position of the first region 28, dR: deviation of the position of the second region 30; L: distance from the inner surface on the left side 58, R: distance from the inner surface on the right side 60, m: first image. The difference between the position where the image is taken and the position where the second image is taken 62, a: A coefficient related to the angle of view of the photographing device.

dL: dR ＝ a ・ m ＋ R: a ・ m ＋ L ・ ・ ・ Equation (1)

Here, if the difference 62 between the position where the first image is taken and the position where the second image is taken is sufficiently smaller than the distance 58 from the left inner surface and the distance 60 from the right inner surface, the angle of view is 90. Since a is at most 1.0 at the time of degree, a · m, which is a minute term on the right side of equation (1), can be ignored.

dL: dR ＝ R: L ・ ・ ・ Equation (2)

Is established. In the case of a moving image shooting device, it is usual that an image of 50 to 60 frames can be shot per second, and the difference 62 between the position where the first image is shot and the position where the second image is shot can be set sufficiently small. .. That is, by using this equation (2), the ratio of the distances from the left and right inner surfaces of the structure to the imaging device 34 is based on the values of the deviation 28 of the position of the first region and the deviation 30 of the position of the second region. Can be calculated. If the difference 62 between the position where the first image is taken and the position where the second image is taken and the angle of view a of the photographing device are known, the equation (1) can be used as it is, and the result is more accurate. It is possible to obtain.

Next, the ratio 35 of the distance from the inner surface in the vertical direction to the photographing device is obtained by using the deviation 31 of the position of the third region. The deviation 31 of the position of the third region is generally smaller than the deviation 28 of the position of the first region and the deviation 30 of the position of the second region. This is because the image taken by a normal photographing device is not a square but a horizontally long rectangle. That is, the angle of view in the vertical direction is smaller than that in the horizontal direction. When the ratio of the angle of view in the vertical direction and the horizontal direction is used as the correction coefficient α, it is almost the same as in Eq. (2).

dL: dR: α × (1-dU) ＝ R: L: U ・ ・ ・ Equation (3)

Is obtained. However, dU: the deviation of the position of the third region is 31, and U: the distance to the upper inner surface is 61. Since the value of the correction coefficient α is known in advance as the specifications of the imaging device, the ratio of U can be found by multiplying by (1-dU).

Using this result, in the case of a pipe having a circular cross section, the ratio of the distance to the lower inner surface can also be obtained by a simple subtraction. For example, if R: L: U = 0.35: 0.65: 0.3, the ratio of the distance to the lower inner surface can be calculated as 1.0-0.3 = 0.7.

Similarly, even if the cross section of the pipe is rectangular and the ratio of height to width is known, the ratio of the distance to the lower inner surface can also be obtained by subtraction if the ratio of U is known. However, since the ratio is the ratio to R and L, if the cross-sectional shape is not square, it can be obtained by the calculation shown below.

For example, suppose that R: L: U = 0.35: 0.65: 0.3 when the width of the rectangular tube is 5 m and the height is 3 m. In that case, the distance from the inner surface on the left side to the imaging device is 5 x 0.35 = 1.75 m, the distance from the inner surface on the right side to the imaging device is 5 x 0.65 = 3.75 m, and the distance from the inner surface on the upper side to the imaging device is 5 x 0.3 = 1.5 m. It becomes. Since the height of the rectangular tube is 3 m, the distance from the lower inner surface to the photographing device is 3-1.5 = 1.5 m, and the photographing device is located at half the height from the ceiling of the rectangular tube.

The ratio 34 of the distance from the left and right inner surfaces of the structure to the photographing device and the ratio 35 of the distance from the upper and lower inner surfaces of the structure to the photographing device obtained as described above are sent to the display device 36, and the result is displayed on the screen. Will be done.

FIG. 10 is an example of the screen of the display device 36. In this example, the distance ratio from the wall surface to the imaging device is shown as left: right = 0.35: 0.65, and the distance ratio from the ceiling to the imaging device is shown as 0.30. In FIG. 10, the sum of the left and right distance ratios is shown as a numerical value of 1.0, but this example is not particularly limited as long as the total is displayed as a numerical value such as 100 in which the relative relationship is easy to understand. ..

Below the numerical display is an illustrated example. The large circle represents the cross section in the case of a circular tube, and the black circle existing in the large circle indicates the position of the photographing device. If the distance ratio from the wall surface to the imaging device is left: right: top = 0.5: 0.5: 0.5, the black circle is located in the center of the large circle, and as mentioned above, left: right: top = 0.35: 0.65: 0.30. If there is, a black circle is located on the upper left side. By providing such an illustrated display, it is possible to grasp at a glance at a glance at which position on the cross section the image was taken.

FIG. 11 is a block diagram of each means of image processing according to the second embodiment of the present invention. The same process as in Example 1 is performed until the ratio 34 of the distance from the inner surface of the structure to the photographing device and the ratio 35 of the distance from the upper and lower inner surfaces of the structure to the photographing device are calculated. The ratio 34 of the distance from the inner surface of the structure to the photographing apparatus is given to the inner surface distance calculating means 70. The inner surface distance calculating means 70 is also given a cross-sectional dimension 68 of the inner surface of the structure extending in the longitudinal direction, and the distance 58 with the left inner surface, the distance 60 with the right inner surface, and the distance 61 with the upper inner surface are calculated. ..

For example, in the case of a circular pipe having a circular cross-sectional shape, if the diameter C is given as the cross-sectional dimension 68 of the inner surface of the structure extending in the longitudinal direction, the distances can be calculated by Eqs. (4) to (6), respectively. Here, L: distance from the left inner surface 58, R: distance from the right inner surface 60, U: distance from the upper inner surface 61, dL: deviation of the position of the first region 28, dR: position of the second region. Deviation 30, dU: Deviation 31 of the position of the third region.

L ＝ C × (dR ÷ (dL ＋ dR)) ・ ・ ・ Equation (4)

R ＝ C × (dL ÷ (dL ＋ dR)) ・ ・ ・ Equation (5)

U ＝ C × ((1-dU) ÷ (dL ＋ dR)) ・ ・ ・ Equation (6)

When the cross-sectional shape is rectangular, by multiplying the width W or height H of the rectangle instead of the diameter C, the distance 58 to the left inner surface, the distance 60 to the right inner surface, and the distance 61 to the upper inner surface are calculated. It can be calculated from 7) by equation (9).

L ＝ W × (dR ÷ (dL ＋ dR)) ・ ・ ・ Equation (7)

R ＝ W × (dL ÷ (dL ＋ dR)) ・ ・ ・ Equation (8)

U ＝ H × ((1-dU) ÷ (dL ＋ dR)) ・ ・ ・ Equation (9)

The values of the distance 58 with the left inner surface, the distance 60 with the right inner surface, and the distance 61 with the upper inner surface calculated in this way are sent to the display device 36, and the result is displayed on the screen.

FIG. 12 is an example of the screen of the display device 36. In this example, the distance from the wall surface to the imaging device is shown as 0.7 m on the left, 1.3 m on the right, and 0.6 m on the top. The figure below has the same display contents as in the first embodiment, and by providing such an illustration, it is possible to grasp at a glance at which position on the cross section the image was taken without error.

FIG. 13 is a block diagram of each means of image processing according to the third embodiment of the present invention. This is the same as in Example 1 until the ratio 34 of the distance from the left and right inner surfaces of the structure to the imaging device and the ratio 35 of the distance from the upper and lower inner surfaces of the structure to the imaging device are calculated. In the third embodiment, these values are given to the development drawing calculation means 72. On the other hand, the second image 10 is given to the vanishing point calculation means 76, and the vanishing point coordinates 78 are calculated. The calculated vanishing point coordinates 78 are also given to the development drawing calculation means 72.

In the vanishing point calculation means 76, the coordinates of the vanishing point, which is the farthest point in the longitudinal direction in the direct-view image as shown in FIG. 2, are obtained by image processing. There are various methods for this image processing method, and any of them may be used. For example, the image processing method can be obtained by the following procedure.
(1) Search for the pixel with the lowest brightness (= darkness) in the direct-view image, and search for the largest circle composed of the pixel with that brightness. The center position (x, y) and radius r are memorized.
(2) Raise the reference brightness a little and search for the largest circle composed of pixels of that brightness. The center position (x, y) and radius r are memorized.
(3) Repeat (1) and (2).
(4) Find the first-order approximation function f such that x = f (r) by the least squares method.
(5) Find the first-order approximation function g such that y = g (r) by the least squares method.
(6) Let x and y as a result of substituting the radius r = 0 into the first-order approximation function of (4) and (5) as the coordinates of the vanishing point.

This is because the vanishing point is basically the darkest because there is no illumination inside the tube, and when the cross section is a circular tube, it gradually becomes circular from the pixel of the vanishing point to the pixel near the imaging device on the image. It utilizes the feature that the brightness increases. When the cross section is rectangular, the vanishing point can be obtained by the same procedure. In that case, the figure to be searched for in (1) and (2) may be a rectangle instead of a circle, and the value of the long side or the short side may be used instead of the radius r.

In the developed view calculation means 72, in addition to the coordinates 78 of the vanishing point, the developed view is drawn using the ratio 34 of the distance from the left and right inner surfaces of the structure to the photographing device and the ratio 35 of the distance from the upper and lower inner surfaces of the structure to the photographing device. Generate. The procedure for creating the development view is schematically shown in FIG. The case of a circular tube having a circular cross section will be described below, but a developed view can be obtained by the same procedure for a rectangular tube whose cross-sectional shape and dimensions are known.

FIG. 14A is the same as the direct-view image shown in FIG. From this direct-view image, as shown in FIG. 14B, pixels of a group of circles that do not intersect each other are read out. The center position of the circle group is generally different, and in order to determine the position, the coordinate 78 of the vanishing point described above, the ratio of the distance from the left and right inner surfaces of the structure to the imaging device 34, and the upper and lower inner surfaces of the structure A distance ratio of 35 to the imaging device is used.

The center position of the circle with the smallest radius (that is, 0) in the circle group coincides with the vanishing point. The smaller the ratio 34 of the distance from the left and right inner surfaces of the structure to the imaging device and the ratio 35 of the distance from the upper and lower inner surfaces of the structure to the imaging device, the more the center position of the circle group moves to the opposite side. For example, as shown in FIG. 14 (b), when the ratio of the distance to the upper inner surface is smaller than the ratio of the distance to the lower inner surface (that is, the photographing device is closer to the ceiling), the larger the radius of the circle, the more the center. The position moves downward.

In FIG. 14B, a group of circles with sparse radii and center positions is shown for easy understanding, but the pixels of the direct-view image (a) can be read out as a group of circles by taking these closely. The radius of the circle group corresponds to the depth of the structure extending in the longitudinal direction, and by geometrically transforming this and plotting it, a developed view as shown in FIG. 14 (c) can be obtained. This is a developed view of one direct view image. By moving the moving body provided with the photographing device, a plurality of first images 12 and a plurality of second images 10 can be obtained in the traveling direction. Since a development view as shown in FIG. 14 (c) can be obtained for each set of images, it is provided as a development view in which a cylindrical tube is cut open by connecting them in the depth direction. Can be done. In this connection, all of the developed drawings may be used, or some of them may be used. The area near the vanishing point in the direct-view image (a) is dark, and the number of pixels contained in one of the circles is small, so the image quality is poor. Therefore, if information from other developments can be obtained, that part should not be used. Is good. An example of this is schematically shown in FIG. FIG. 15 shows a procedure for obtaining a developed view 82 in which, for example, when there are three developed views 1 to 3, only a bright portion is cut out from the front side in the depth direction and the developed views 82 are connected and connected. ing. Although only up to the developed view 3 is shown in this figure, 11 developed views are required to obtain the connected developed view 82 in the figure. There is no particular limitation on the number of sheets to be connected. When connecting, it is desirable to match the images in the vicinity of each connecting portion and connect them so that there is as little deviation as possible. The developed view 82 connected in this way is sent to the display device 36 and output as a screen or a printed matter.

As described above, according to the present invention, even if the moving body equipped with the imaging device does not have a distance sensor, it is possible to obtain information on the distance from the inner surface of the structure extending in the longitudinal direction based on the captured image. it can. By using the obtained distance information, it is possible to obtain a high-quality development drawing with little distortion.

10 ... Second image 12 ... First image 14 ... Region extraction means 16 ... First region of the second image 18 ... Second region of the second image 19 ... Third region of the second image 20 ... Matching means 22 ... Correspondence information of the first region 24 ... Correspondence information of the second region 25 ... Correspondence information of the third region 26 ... Position difference calculation means 28 ... Position deviation of the first region 30 ... Deviation of the position of the second region 31 ... Deviation of the position of the third region 32 ... Inner surface distance ratio calculating means 34 ... Ratio of the distance from the left and right inner surfaces of the structure to the photographing device 35 ... From the upper and lower inner surfaces of the structure to the photographing device 36 ... Display device 38 ... Structure extending in the longitudinal direction 40 ... Sewage 42 ... Photographing device 44 at the position where the first image is taken ... Photographing device 46 at the position where the second image is taken. ... The seam 48 of the structure extending in the longitudinal direction ... The extracted first region corresponds to the corresponding region 50 in the first image ... The extracted second region corresponds to the corresponding region 51 ... Extraction The third area corresponds to the corresponding area in the first image 56 ... Focus distance 58 ... Distance to the left inner surface 60 ... Distance to the right inner surface 61 ... Distance to the upper inner surface 62 ... Position for capturing the first image Difference between the position where the second image is taken 64 ... Equivalent image plane on which the first image is projected 65 ... Equivalent image plane on which the second image is projected 68 ... On the inner surface of the structure extending in the longitudinal direction Sectional dimensions 70 ... Inner surface distance calculating means 72 ... Expanded view calculating means 74 ... Expanded view 76 ... Disappearing point calculating means 78 ... Disappearing point coordinates 80 ... Expanded view Connecting means 82 ... Connected developed view

Claims (4)

The first image and the second image of the inner surface of the structure extending in the longitudinal direction taken by the imaging device from two different points in the longitudinal direction, and
Region extraction means for extracting the first, second and third regions from the second image,
A matching means in which the extracted first, second, and third regions search for corresponding regions in the first image,
A position deviation calculation means for calculating the position deviations of the first, second, and third regions in the first image and the second image, and
An inner surface distance ratio calculating means for calculating the ratio of the distance from the inner surface of the structure to the photographing apparatus based on the deviation of the positions of the first, second and third regions, and
A display device that displays the ratio of the distance from the inner surface of the structure to the imaging device numerically or graphically,
An image processing device characterized by comprising. In claim 1, from the inner surface of the structure to the imaging device based on the ratio of the distance from the inner surface of the structure to the imaging device and the cross-sectional dimensions of the inner surface of the structure extending in the longitudinal direction obtained by the inner surface distance ratio calculating means. Inner surface distance calculation means for calculating the distance of
A display device that displays the distance from the inner surface of the structure to the imaging device numerically or graphically,
An image processing device characterized by comprising. In claim 1, the vanishing point calculation means for calculating the position of the vanishing point in the second image and
A development map calculation means that calculates a development map based on the ratio of the distance from the inner surface of the structure to the imaging device and the position of the vanishing point.
A development view connecting means for connecting a plurality of development views in the longitudinal direction,
A display means for illustrating the connected development view and
An image processing device characterized by comprising. The image processing apparatus according to claim 3, wherein the vanishing point calculation means determines the position of the vanishing point using light and dark information in the image.

Images