JP3276621B2 - Crack detection method for inner wall of tunnel - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method for detecting cracks in a tunnel inner wall surface made of reinforced concrete or steel when inspecting an inner wall surface of a tunnel or the like.

[0002]

2. Description of the Related Art At the time of maintenance, repair, and reinforcement of an existing tunnel, periodic inspections and investigations are conducted to ascertain the current state of damage and deterioration of the inner wall of the tunnel, and based on the results, the durability of the tunnel. Need to be evaluated and diagnosed accurately. 2. Description of the Related Art Conventionally, as disclosed in, for example, Japanese Patent Application Laid-Open No. 3-160349, a vehicle is equipped with a camera capable of capturing a digital image such as a CCD camera, and an image of an inner wall surface is captured by the camera while moving the vehicle. 2. Description of the Related Art There is known a method in which an image representing a crack or a flaw on an internal wall surface is extracted from stored image data, and a feature of the image is emphasized to detect a crack or a flaw.

[0003]

However, in the above-mentioned conventional example, cracks and flaws are detected by image data captured by a camera, and are displayed as line images when they are displayed on a display or the like. However, there is a limit to the size of the display, and when the inspection area is wide such as a road or a tunnel, if the entire inspection area is displayed at a time on one screen of the display, the line image displayed on one screen of the camera Are displayed extremely small, which causes a problem that the line image becomes unclear and it is difficult to grasp the state of cracks and flaws. The present invention has been made in view of the above circumstances, and performs image processing on image data captured by a video camera or the like, performs output display for each small section, creates crack information of an inspection range, and easily grasps a crack situation. It is another object of the present invention to provide a method for detecting cracks on the inner wall surface of a tunnel, which has high reliability and workability.

[0004]

According to the present invention, there is provided a method for detecting a crack on an inner wall surface of a tunnel according to the present invention, comprising the steps of: The obtained image data is subjected to image processing, and the line data of the crack is extracted for each of the small sections in which the inner wall surface is divided in the tunnel penetrating direction, and different colors or lightness are different depending on the degree of the crack as a feature point. The line data is subjected to gradation processing, and a quantification process for calculating the length and area of the line data is performed for each of the gradation-converted line data, and the length and area of the line data obtained for each feature point are summed. Then, the information is stored as the corresponding crack information for each of the small sections, and the crack information is attached to each of the small section positions on the internal wall surface and output and displayed. Thereby, the crack information is clarified for each subsection, and it is possible to immediately take a measure according to the degree of the crack. In the method for detecting cracks on the inner wall surface of a tunnel according to the present invention,
A plurality of cameras are arranged side by side in a direction perpendicular to the tunnel penetration direction, and are mounted on a truck traveling in the tunnel penetration direction facing the inner wall surface, and images are taken.Image data is obtained together with the traveling position of the truck. It may be composed of data obtained.

[0005]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments of the present invention will be described with reference to the accompanying drawings to provide an understanding of the present invention. Here, FIG. 1 is a front view showing a crack detecting device used in a method for detecting a crack on an inner wall surface of a tunnel according to one embodiment of the present invention, and FIG. 2 is a view showing a plurality of cameras of the crack detecting device used in the crack detecting method. FIG. 3 is an explanatory view showing a relative position of an image, FIG. 3 is an explanatory view showing a state of image processing, FIG. 4 is a block diagram of a crack detection device used in the crack detection method, and FIGS. 5 (A), (B) and (C). Is an explanatory diagram, a distribution diagram, and a wide distribution diagram showing image data obtained by the crack detection method, respectively, FIG. 6 is a flowchart showing the crack detection method, and FIG. 7 is a distribution diagram of cracks used in the crack detection method. FIG.

As shown in FIG. 1, a crack detecting device 10 used in a method for detecting a crack on an inner wall surface of a tunnel according to an embodiment of the present invention takes an image of an arch-shaped inner wall surface 20 of a tunnel 21 to obtain a digital quantity. An imaging device 30 for acquiring the image data of the image data is provided. The imaging device 30 includes a plurality of (for example, four) cameras 31 which can face the inner wall surface 20 of the tunnel 21 and can take images along the circumferential curved surface at equal intervals from the top of the arch to the left and right, and a screen where the cameras 31 are clear. As can be obtained, an illumination device 32 that illuminates an imaging surface in a range wider than the field of view of the camera 31 and an image storage medium 33 are provided. The image data captured by each camera 31 is output as a digital quantity such as a CCD camera or a digital video camera, and stored in the image storage medium 33. An imaging device 30 including a camera 31 and a lighting device 32 is provided with a moving device 40 including a carriage 41 that can move in a direction parallel to the inner wall surface 20 in a direction in which the tunnel 21 penetrates.
It is placed on. Note that the cameras 31 are arranged on the moving device 40 in a direction perpendicular to the direction in which the tunnel penetrates, so that the adjacent image data is slightly overlapped with each other and photographed. It is controlled to start shooting.

As shown in FIG. 2, a plurality of units (in this case, 4 units)
The images 34 captured by the cameras 31 of the two cameras have the heads of the respective image data of the cameras 31 aligned by the imaging device 30. The camera 31 is, for example, 3
A plurality of continuous images 34 of about 0 frames are recorded. If each camera 31 records while moving by the carriage 41 in the direction of penetration of the tunnel 21, image data is obtained together with the traveling position of the carriage 41. However, the number of overlapping portions between the images 34 adjacent to each other in the moving direction increases with the movement, and it is difficult to evaluate the image data as it is. Therefore, as shown in FIG. 3, when the images 34 are extracted for each of a plurality of frames from the continuous images 34 according to the moving speed of the camera 31 and the extracted images 34 are successively arranged, an adjacent image 34 A continuous image 3 that slightly overlaps each other
Edit to get 5. For example, when the moving speed of the carriage 41 is about 5 m / sec (18 km / h) and the photographing speed of the camera 31 is about 30 frames / sec, when one frame is photographed in a small section of about 1000 mm × 700 mm, every four frames A continuous image 35 is obtained by extracting and arranging one frame at a time. One frame of image data R0 of the continuous image 35 is data including an overlapped portion with an adjacent portion. By performing image synthesis by joining the image data R0, the inner wall surface 20 continuous along the tunnel 21 penetration direction is formed. Are obtained.
The plurality of continuous image data R0 are divided into large sections divided in the direction of penetration of the tunnel 21, and a small section number is assigned to each frame (image data R0) of the image 34 in the section for each large section. . That is, the image data R0 divides the inner wall surface 20 into large sections at intervals of 1 km along the direction in which the tunnel 21 penetrates, for example, as shown in FIG. 5C, and assigns large section numbers P1, P2 to Pi. Further, as shown in FIG. 5B, four cameras 31 (# 1
Camera to # 4 camera) image data R0 of image 34,
The small section numbers (Q11 to Q11) are sequentially set in the penetration direction of the tunnel 21.
(Q41, Q12 to Q42... Q1n to Q4n) and sequentially stored in files (P1, P2 to Pi) assigned the same numbers as the large division numbers of the image storage medium 33.

As shown in FIG. 4, the image data R0 stored in the image storage medium 33 is divided into adjacent small sections (for example, Q11 and Q21, Q11 and Q12) by the image editing device 50.
Etc.), the editing process is performed to delete the overlapping portions of the image data R0 for each image data R0, and the required image data R with little noise
1 is taken out, and the image data R1 is subjected to image processing by an image processing device 60 such as a personal computer. In the image processing device 60, as shown in FIG.
Is subjected to a density conversion process for each of the small sections into which the inner wall surface 20 is divided to smooth the image density, and is separated for each of the thicknesses of the crack lines, and the feature points of the crack are extracted and quantified. The characteristic amount of the crack, for example, the thickness and length of the line is calculated and displayed for each color. For example, large cracks are colored red, medium ones are blue, and small ones are green. The characteristic amount of the crack is subjected to data processing for each subsection, and this crack information is pasted to each subsection position, and as shown in FIG.
A distribution map 70 indicating the degree of cracking for each of the large sections of 0 is created, and as shown in FIG.
A continuous wide-area distribution map 71 is obtained by calculating 0. The distribution map 70 and the wide distribution map 71 are sequentially recorded on a storage medium 81 such as a recording tape or a magnetic tape, or recorded in a digital amount, so that they can be output and displayed on a display device 80 such as a CRT, a liquid crystal display, or a printer. .

Here, a method for detecting and displaying a crack on the inner wall surface of a tunnel according to an embodiment of the present invention will be described (see a flowchart shown in FIG. 6). (1) The inner wall surface 20 is imaged by a plurality of (four in this case) cameras 31 while the imaging device 30 is mounted on a carriage 41 and moved, and the image data R0 is stored in an image storage medium 33.
(Step 1). (2) The image data R0 of the image storage medium 33 is input to the image editing device 50, and the adjacent small sections (for example, Q11 and Q1)
2, Q12, Q13, etc.), the overlapping portions of the image data R0 are deleted, and the portions which are easily misjudged, such as dark portions and dark lines caused by the overlapping, are removed, and the necessary portions of the image data R1 are removed. (Step 2). (3) The image data R1 is input to the image processing device 60, and a density conversion process is performed (step 3). In the density conversion processing, since the image represented by the captured image data R1 varies from a bright image to a dark image, an average value of the image densities is obtained, and the processing is performed so that the average value becomes the central density of the image density. To reduce variations in image density.

(4) In order to emphasize the data of the crack line of the image data R1, for example, an operator having pixels of a 5 × 5 matrix is used to form an image which is continuous vertically, horizontally, diagonally to the right or diagonally to the left. A feature point process for determining a feature amount to be emphasized is performed (step 4). In the feature point processing, for example, the width of each of the vertical line (X direction), the horizontal line (Y direction), and the diagonal direction is set to 0.1 to 0.19, and the feature point A is set to 0.2 to 0.29. With the feature points B and 0.3 or more as feature points C, the image data R1 is passed through a filter that allows the image data of each feature point to pass. (5) Extract image data R2 as a line obtained by binarizing the image data R1 that has passed through the filter that passes each of the feature points A, B, and C according to the density of black and white using the average density as a threshold ( Step 5). (6) A line is drawn using the image data R2 as a line extracted and binarized at each of the feature points A, B, and C, and a line extraction process is performed (step 6). Therefore, the drawn line after the binarization is expressed as a line having the same thickness as the thick line and the thin line. (7) The drawn line is colored for each feature point and gradation is made into a different color or brightness depending on the degree of the crack, and the type of the flaw is easily distinguished based on the crack information (step 7). For example, the line data extracted at the feature point A is colored green, the line data extracted at the feature point B is colored blue, and the line data extracted at the feature point C is colored red to determine the type of flaw. Make it easy to do.

(8) The length and area of the line data are calculated for each feature point to quantify the crack information (step 8). (9) The length and area of the line data obtained for each feature point are summed to obtain the total length and area of the crack for each image data R2 (step 9). (10) The total length of cracks and the total area data (crack information) for each image data R2 are stored as line images for each corresponding small section, and pasted to each small section position on the inner wall surface and displayed on the display device 80. (Step 10). (11) As shown in FIG. 7, the line image of each subsection subjected to the image processing is reduced, and the line images are compared next to each other according to the subsection address of the internal wall surface 20, and the distribution map 70 of the crack for each internal wall number is shown. Is created (step 11). The distribution map 70 compares the crack length or area of each line image with a predetermined evaluation length or area and, if the comparison result is within the evaluation criteria, displays a screen of a small section of the line image with a predetermined color in a predetermined color. Paint. For example, when the evaluation length a is set to 1 to 2 m, when the crack length obtained by the image processing is in the meantime (for example, 1.5 m), green, and when the evaluation length b is set to 2 to 4 m, If the crack length obtained by the image processing is during this period (for example, 2.5 m), the color is blue, and the evaluation length c is 4 m.
With the above setting, if the crack length obtained by the image processing is during this period (for example, 5.5 m), the color is set to red. The above processing is performed on the large section numbers of all the internal wall surfaces 20, that is, the file numbers (P1 to Pi), and the small section numbers are pasted together so as to be arranged in succession. A wide-range distribution diagram 71 identified by color is displayed on the screen to grasp the deterioration location, the frequency of deterioration, and the like of all the internal wall surfaces 20. (12) With respect to the entire inner wall surface 20, for example, the display scale of a portion where the degree of damage and deterioration is particularly severe in the range of each small section is enlarged, and an image as a crack line for each small section of the enlarged distribution map 70 is displayed. The detailed data, which is the data R2, is recorded or recorded in a digital amount on a storage medium 81 such as a magnetic tape provided on the display device 80 so that the detailed data can be easily viewed visually, and displayed on the display device 80 as a line image (step 1).
2).

As a result, cracks appearing on the surface of the inner wall surface 20 are emphasized and clarified by the density conversion processing, and features such as the shape and size of the cracks are represented by counting by the crack extraction and quantification processing. As shown in FIG. 5, the state of cracks on the entire inner wall surface 20 is displayed as a distribution map 70, and it is possible to immediately take a measure according to the degree of cracks. In particular, displaying the file numbers (P1 to Pi) of the large sections of all the internal wall surfaces 20 as the line images of the small sections on the display device 80 requires only one screen because the display area of the display device 80 is limited. Although it cannot be displayed, a distribution map in which the line image of each subsection is reduced is obtained, and the distribution map 70 is attached so that adjacent subsection numbers are sequentially arranged in order, and the evaluation for each subsection is included. The entire image of the inner wall surface 20 can be grasped on one screen by displaying the wide area distribution map 71. When it is desired to know the state of the crack in detail, the display scale can be enlarged, and detailed data of the range of each subsection can be displayed as a line image on the enlarged distribution map 70, so that the data can be seen clearly and visually. . In addition, the output display is gradation-converted into different colors or lightness according to the thickness and length of the cracks for each small section by the density conversion processing, and further, the gradation-converted color or lightness is applied to the inner wall surface 20. By attaching the image to each of the small segment positions, the degree of cracking can be more easily grasped than displayed in a black and white line image. Further, in the description of the method for detecting a crack on the inner wall surface of the tunnel according to the embodiment, an example is described in which a plurality of cameras capable of shooting along the circumferential curved surface at equal intervals left and right from the top of the tunnel arch have been described. However, if there is a double track or track inside the tunnel, install multiple cameras that can take pictures of the inner wall of one side from the top of the arch along the curved surface in the circumferential direction, After the crack detection on the wall surface is performed, the crack detection on the inner wall surface on the other side may be performed on the return path.

[0013]

According to the first aspect of the present invention, in the method for detecting cracks in the inner wall surface of a tunnel, the inner wall surface is imaged by a camera, the obtained image data is subjected to image processing, and the inner wall surface is divided in the tunnel penetrating direction. Every minute, the line data of the crack is extracted, and the line data is toned to different colors or lightness according to the degree of the crack as the feature point. Performs a quantification process to calculate the area, sums the length and area of the line data obtained for each feature point, stores it as crack information for each corresponding subsection, and pastes the crack information at each subsection position on the inner wall surface Since the output is attached and displayed, it is possible to improve the reliability of the crack data for each subsection and provide a low-cost crack detection method for the inner wall surface of the tunnel with good workability. In addition, the cracks in each of the small sections become clear, and it is possible to immediately take a measure according to the degree of the cracks.

[Brief description of the drawings]

FIG. 1 is a front view showing a crack detection device used in a method for detecting a crack on an inner wall surface of a tunnel according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing relative positions of images of a plurality of cameras of a crack detection device used in the crack detection method.

FIG. 3 is an explanatory diagram showing a state of image processing.

FIG. 4 is a block diagram of a crack detection device used in the crack detection method.

FIGS. 5A, 5B, and 5C are an explanatory diagram, a distribution diagram, and a wide distribution diagram showing image data obtained by the same crack detection method, respectively.

FIG. 6 is a flowchart showing the crack detection method.

FIG. 7 is an explanatory diagram showing a distribution map of cracks used in the crack detection method.

[Explanation of symbols]

10: crack detection device, 20: inner wall surface, 21: tunnel, 30: imaging device, 31: camera, 32: lighting device, 33: image storage medium, 34: image, 35: continuous image, 40: moving device, 41 : Cart, 50: image editing device, 60: image processing device, 70: distribution map, 71: wide distribution map, 80: display device, 81: storage medium

──────────────────────────────────────────────────続 き Continuation of the front page (56) References JP-A-9-161068 (JP, A) JP-A-11-102429 (JP, A) JP-A-11-39471 (JP, A) JP-A-7-161 77498 (JP, A) JP-A-4-285845 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G01N 21/84-21/88 G01C 7/06 G06T 7/00

Claims (1)

(57) [Claims] 1. A method for detecting a crack in an inner wall surface of a tunnel, wherein the inner wall surface is imaged by a camera, and image data obtained is subjected to image processing to divide the inner wall surface in a direction in which the tunnel penetrates. For each subsection, line data of a crack is extracted, the line data is toned into different colors or lightness in accordance with the degree of the crack as a feature point, and the line is further divided for each of the gradation-converted line data. Performing a quantification process for calculating the length and area of the data, summing the length and area of the line data obtained for each feature point, storing the sum as crack information for each corresponding subsection, and storing the crack information in the inner wall surface. A method for detecting cracks on the inner wall surface of a tunnel, wherein the method is attached to each of the subsections and output-displayed.