JP3004191B2 - Bolt box position detection method - 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 a position of a bolt box for connecting segments in a shield method, for example.

[0002]

2. Description of the Related Art Conventionally, to detect the position of a bolt box for connecting a segment in a shield method,
The position of the bolt box has been detected by photographing the bolt box with the camera device and performing image processing on the photographed image.

The above-mentioned image processing is a binarization process of shading information of a photographed image based on a predetermined threshold value. The feature is that the edge position of the bolt box was required.

[0004]

According to the above-described image processing, the photographed image is binarized by a predetermined threshold value, and the edge position of the bolt box is determined based on the binarized image information. If, for example, when the bolt box is photographed, if the illumination unevenness or the bolt box itself is dirty, the edge position is detected (extracted) only by binarization processing based on a certain threshold value. ) Was difficult to do.

Accordingly, an object of the present invention is to provide a method for detecting a box for a bolt that can solve the above problem.

[0006]

According to a first aspect of the present invention, there is provided a method for photographing a bolt box with a camera device and detecting a position of the bolt box based on the photographed image. Then, image processing is performed on an image photographed by the camera device to form a grayscale image having a plurality of gradations, and from this grayscale image data, grayscale image data for one line along a predetermined direction is acquired. This is a method in which a portion where the change in the shade of the shade image data of one line is the largest is determined as the edge position of the bolt box.

According to a second aspect of the present invention, there is provided a method of photographing a box for a bolt with a camera device and detecting the position of the box for the bolt based on the photographed image. Image processing is performed on the image captured by the device to produce a multi-tone image,
From this grayscale image data, grayscale image data for a plurality of lines along a predetermined direction is acquired, and taking into account the portion where the grayscale change of the grayscale image data for each of the acquired lines is greatest, the bolt box This is a method of determining an edge position.

Further, in the above-mentioned first or second means, when obtaining the position of the change in shading in the line of the shading image data, the sum of the shading data is determined on both sides and over a predetermined range with the change position as a boundary. This is a method for detecting the position of the bolt box by comparing the sum of the two shaded data and determining the edge position of the bolt box.

[0009]

According to each of the above constructions, the photographed image data of the bolt box is represented by a plurality of gradations in accordance with the shading, and one or a plurality of the image data is represented by the plurality of gradations. Image data on the line
Since the portion where the change in shading is extremely large is determined as the edge position of the bolt box, that is, the edge position is detected by the relative change in shading on the same line, so that the brightness of the illumination or Edges can be accurately detected without being affected by uneven illumination or the like.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to FIGS. In the present embodiment, as shown in FIG.
A description will be given of a case where the position of the rectangular bolt box 3 in a plan view for inserting the connecting bolt 2 for connecting the segments 1 constructed on the inner surface of the tunnel excavated by the shield machine is described. The method of detecting the position of the bolt box 3 is necessary for automating the connection operation between the segments 1.

First, a position detecting device used in the position detecting method will be described with reference to FIG. This position detecting device is provided on a device body (not shown) side and illuminates a bolt box 3 formed on a segment 1 with a lighting device 1.
1; a camera device 12 for photographing the bolt box 3 illuminated by the lighting device 11;
2. The image data captured by step 2 is input, and an image processing device 13 that performs predetermined image processing based on the input image data and the image data that has been subjected to image processing by the image processing device 13 are input. And an arithmetic processing unit 14 for obtaining the position of the bolt box 3 based on the image data, and a display device 1 for displaying the edge of the bolt box 3 obtained by the arithmetic processing unit 14.
And 5.

Next, a method for performing predetermined image processing on image data obtained by the camera device 12 and detecting the position of the bolt box 3 will be described. First, the illuminated bolt box 3 is photographed by the camera device 12, and the photographed gray image data (shown in FIG. 3) is sent to the image processing device 13.

In the image processing apparatus 13, the sent image data is represented by a predetermined gradation, for example, 256 gradations, according to the density. Then, as shown in FIG.
Image data for a plurality of lines L along one side 3b of the bolt box 3 is obtained based on the grayscale image data represented by six gradations, and the image data along each of the lines L is sent to the arithmetic processing unit 14. Can be

In the arithmetic processing unit 14, as shown in FIG. 5, for the acquired image data a of each line L, a portion (portion) b where the shading portion (light and dark portion) greatly changes is obtained. Then, by connecting the changing portions b obtained for each of the lines L, the position of the other side orthogonal to one side of the bolt box 3, that is, the position of the edge 3a can be clearly known.

At this time, in order to more accurately detect the position of the edge 3a, as shown in FIG. 6, the image data of the front and rear portions of the portion b determined to be the edge 3a, that is, the dark portion (dark portion) ) And image data d in a predetermined range of a light portion (bright portion).
Are compared with each other.

As described above, since the edge position of the bolt box 3 is detected in consideration of the sum of the image data at the front and rear positions of the shading change portion, adverse effects due to sudden changes in image data due to noise can be prevented. Can be.

As described above, the photographed image data of the bolt box 3 is represented by a plurality of gradations in accordance with the density.
Then, from the image data represented by the plurality of gradations, image data is acquired on a plurality of lines L along one side of the bolt box 3, and a portion where the change in shading is extremely large is determined.
Since the position of the edge 3a of the bolt box 3 is determined, that is, the edge is detected by comparing the relative densities on the same line, it may be affected by illumination brightness or illumination unevenness. The edge can be accurately detected without the influence of disturbance such as dirt or the like. Of course, the entire edge of the bolt box 3 can be detected by connecting the shading change points obtained on a plurality of lines. If this detection method is applied around the box 3 for bolts, the entire position of the box 3 for bolts can be known.

[0018]

As described above, according to the construction of the present invention, the photographed image data of the bolt box is represented by a plurality of gradations in accordance with the shading, and the image data represented by the plurality of gradations is represented. From, image data is acquired on one or more lines, and the part where the change in shading is extremely large is
Since it was decided to be the edge of the bolt box,
In other words, the edges are detected by the relative change (relative comparison) of shading on the same line, so that the edge is not affected by the brightness or unevenness of the illumination, or the influence of disturbance such as dirt is small, and the edge is accurately detected. Edges can be detected.

[Brief description of the drawings]

FIG. 1 is a view showing a bolt box according to an embodiment of the present invention, wherein (a) is a front view, (b) is a plan view, and (c) is a cross-sectional view.

FIG. 2 is a diagram illustrating a schematic configuration of a bolt box position detecting device in the embodiment.

FIG. 3 shows a photographed image of a bolt box in the embodiment.

FIG. 4 shows an acquisition line of image data of a bolt box in the embodiment.

FIGS. 5A and 5B are diagrams showing a light and shade state in the image data of the bolt box in the embodiment, wherein FIG. 5A is a graph showing an acquisition line, and FIG. 5B is a graph showing a light and shade state in the acquisition line.

FIG. 6 is a graph showing the vicinity of an edge of image data on an acquisition line in the embodiment.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 segment 3 Bolt box 3a Edge 11 Illumination device 12 Camera device 13 Image processing device 14 Arithmetic processing device

Claims (3)

(57) [Claims] 1. A method of photographing a bolt box with a camera device and detecting the position of the bolt box based on the photographed image, wherein the photographed image photographed by the camera device is subjected to image processing to obtain a plurality of gradations. A gray-scale image of one line along a predetermined direction is acquired from the gray-scale image data, and a portion of the obtained gray-scale image data of one line where the change of the gray scale is largest is used for the bolt. A method for detecting the position of a box for a bolt, comprising determining the position of an edge of the box. 2. A method of photographing a bolt box with a camera device and detecting the position of the bolt box based on the photographed image, wherein the photographed image photographed by the camera device is subjected to image processing to obtain a plurality of gradations. From this grayscale image data, grayscale image data for a plurality of lines along a predetermined direction is obtained, and a portion where the change in grayscale of the obtained grayscale image data for each line is greatest is considered. And determining the edge position of the bolt box. 3. A method for determining a position of a change in gray level in a line of gray image data, calculating a sum of the gray data on both sides and over a predetermined range with the change position as a boundary. The method for detecting the position of a bolt box according to claim 1, wherein the edge position of the bolt box is determined by comparison.