JP2996836B2 - Rock type and rock classification method in image analysis of tunnel face - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image analysis for evaluating the rock condition of a tunnel face.
It relates to a rock type and a rock class identification method.

[0002]

2. Description of the Related Art When performing tunnel excavation work, it is necessary to observe the face at any time to grasp the condition of the bedrock, confirm the safety of the face, and predict the geology ahead. Conventionally, a geotechnical engineer sketches a face and mainly focuses on rock type, rock grade (hereinafter referred to as "rock grade"), degree of weathering, crack characteristics,
We observe the presence of spring water.

[0003]

The conventional method for evaluating rock face of a tunnel face based on sketches is based on subjective evaluation. Therefore, the accuracy of evaluation varies due to lack of geological knowledge or difference in experience. Occurs. Therefore, a person having advanced geological skills is required, and it takes time to obtain an accurate rock mass evaluation.

[0004] Therefore, the accuracy of discriminating the rock type and rock class in the rock face evaluation of the tunnel face is improved, and even a person who does not have specialized knowledge can objectively evaluate the rock face in a short time. Therefore, a technical problem to be solved arises, and an object of the present invention is to solve this problem.

[0005]

DISCLOSURE OF THE INVENTION The present invention has been proposed to achieve the above object. A color image of a tunnel face is obtained, and an RGB signal of the color image is converted to a standard television signal. Converts into a luminance signal and a color difference signal based on the signal, obtains a color vector representing saturation from the color difference signal, expresses a hue distribution by a color angle, and replaces a color vector of each pixel with a maximum value. By expressing it as a characteristic color, the rock type and
An object of the present invention is to provide a rock type and rock class identification method in image analysis of a tunnel face, which is characterized by identifying rock classes .

[0006]

[Function] A tunnel face is photographed by a still video camera or the like, and a color image of the tunnel face is obtained. First, the R, G, and B primary color signals of the color image are converted into a luminance signal and a color difference signal in accordance with a standard television signal. Next, the saturation is obtained from the color difference signal of each pixel and is represented by a color vector.
The hue distribution is represented by a color angle of 0 to 360 °. Further, the color vector of each pixel is replaced with that of the maximum value, and is expressed as a characteristic color that is the color of maximum saturation.

[0007] Thus, each pixel is represented by a characteristic color, which makes it easier for human eyes to recognize. This makes it possible to identify the rock type and the rock class from the difference in the hue, and can be applied to the determination of the rock type and the rock class area when creating an observation diagram of the tunnel face.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below in detail with reference to the drawings. FIG. 1 shows a configuration of a device used for image analysis of a tunnel face, in which a still video camera 1 takes a color image of the tunnel face and records it on a floppy disk 2 as digital data. A plurality of face image data can be recorded on the floppy disk 2. The still video player 3 reproduces the face image data and inputs the face image data to the computer 4.

The computer 4 incorporates a large-capacity frame buffer memory for performing image processing, and is provided with an additional memory and a magnetic disk device. The face image data processed by the computer 4 is displayed on the color display 5 as needed, and a hard copy is created by the color printer 6 as needed. Further, data obtained by image analysis, geological observation information data, and the like are recorded on the magneto-optical disk 7 to construct a geological database.

FIG. 2 shows a flow chart for observing the state of the face by image analysis and evaluating the stability of the tunnel. First, a preliminary investigation of the tunnel is performed by a geological survey, a geophysical survey and a laboratory test (step 101). The excavation work of the tunnel is performed by measuring the mountain (step 102).
The tunnel face is photographed in color by the above-mentioned still video camera (step 103), and the face image data is input to the computer. Also, the initial conditions of the ground are set by a preliminary survey of the tunnel (step 104), and this data is also input to the computer. Then, image processing and analysis of the face are performed by a computer (step 105), and numerical analysis data is created (step 106). The image processing and analysis data is recorded in a geological database (step 112).

Based on the data of the geological database, the state of the ground in front of the face is predicted (step 114).
Numerical analysis is performed on the stability of the discontinuous ground from the numerical analysis data (step 107). Thus, a comprehensive evaluation of the ground stability is performed (step 108). If the ground is unstable, a stabilization measure is selected (step 109 →).
110), continue the tunnel excavation work.

Next, step 10 in the above flowchart
The image analysis systems 3 to 106 will be described in more detail. As described above, the color image of the tunnel face taken by the still video camera is recorded on a floppy disk, and the floppy disk is reproduced by the still video player, and the color image of the tunnel face is input to the computer.

FIG. 3 shows a color image of the tunnel face displayed in monochrome, and the color image is used as it is in the embodiment of the present invention. There is a face in front of the inverted U-shaped support, and a large discontinuous surface exists in an oblique direction from the upper left to the lower right of the face. In the color image, the rock on the upper right of the discontinuous surface has a bluish color, and the rock on the lower left of the discontinuous surface has been oxidized by weathering to have a brown color. The square object at the lower center is a blackboard on which the numbers of the face faces are described.

FIG. 4 shows an observation view of the bedrock sketched by a geotechnical engineer at the site of the face where the color image was taken. The the observation view, the and discontinuous surfaces of magnitude appearing in the color image is drawn, along with identifying the rock type sandstone S _S, Iwakyu bordering a large discontinuity described above and the C _M-class C _L
It turns out that it is different from the class. Next, a procedure for analyzing the color image to identify a rock type and a rock class will be described. For each pixel of the color image, the R, G, and B primary color signals are converted into a luminance signal Y representing brightness and color difference signals RY and BY by the following equation based on a standard television signal.

[0015]

(Equation 1)

Then, as shown in FIG. 5, a color vector representing the saturation is obtained from the color difference signals RY and BY, and the hue distribution is calculated. From 0 to 360 ° in a counterclockwise direction. Further, in a general face color image, since a color difference is unlikely to appear, it is represented by a characteristic color of each pixel. The characteristic color refers to a color having the maximum saturation to be expressed. By expressing the color vector of each pixel by replacing the color vector with the maximum value, the color can be intuitively recognized by human eyes.

FIG. 6 shows the result of color analysis of the color image of the face according to the above-mentioned procedure. In FIG. 6, although the monochrome image is displayed, the analysis result is displayed as a color image in the embodiment of the present invention. You. The upper right part of the large discontinuous surface in the oblique direction is displayed in a bluish characteristic color, and the blackish part at the top center is displayed in a reddish characteristic color. Further, a white portion at the lower left of the large discontinuous surface is displayed with a yellow characteristic color, and a portion having an intermediate density of the white portion is displayed with a red characteristic color. The color analysis results clearly show the difference in hue, and the region indicated by the blue characteristic color differs from the region indicated by the yellow characteristic color in rock type, rock class, weathering degree, and the like. Can be identified.

FIG. 7 shows a luminance signal Y and a color angle θ for each rock class when a color analysis of a continuous face in a section of a tunnel is performed.
In the _CH class shown in FIG. 3A, since the color image has a light gray color, the luminance signal has a higher value than the other rock classes. A part of the color angle is 115-1
It is concentrated at 65 °, but is relatively widely distributed. C _L shown in C _M-class and the drawing shown in FIG. (B) (c)
The analysis results for each class are continuously collected within a certain range, and the difference in color angle is clearly shown.

As described above, by performing the color analysis on the color image of the tunnel face, it is possible to identify the rock type and the rock class, and it is possible to apply the rock type and the rock class area determination when creating an observation diagram of the face. . The present invention can be variously modified without departing from the spirit of the present invention, and it goes without saying that the present invention extends to the modified ones.

[0020]

According to the present invention, as described in detail in the above-described embodiment, each feature of the color of the rock is determined by using the R, G, and B primary color signals of the color image of the tunnel face for each pixel. Be
It is expressed as the characteristic color, which is the color of the maximum color from the vector.
And, from this characteristic color, that is, the difference in hue ,
More it becomes easy to identify the rock type, rock class. Therefore, the characteristic colors are identified by human vision. The accuracy of the identification of rock types and rock classes is improved, and even those who do not have advanced geological technology can evaluate the condition of the rock mass, and save labor. Can be achieved.

Furthermore, by adding information on the discontinuous surface of the face and information on the spring point, the evaluation such as the prediction in front of the face is instantaneously performed for each excavation, and various effects such as contributing to the safety of tunnel construction can be obtained. It is the invention which plays.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a device used for image analysis of a tunnel face.

FIG. 2 is a flowchart of rock mass evaluation by image analysis.

FIG. 3 is a diagram in which a color image of a tunnel face is displayed in monochrome.

FIG. 4 is an observation view of a bedrock in which a tunnel face is sketched.

FIG. 5 is a diagram showing a color vector and a color angle.

FIG. 6 is a diagram in which a color image of a color analysis result is displayed in monochrome.

FIG. 7 shows a relationship between a luminance signal and a color angle for each rock class,
(A) is a graph of _CH class, (b) is a graph of _CM class,
(C) is a _CL class graph.

──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Hitoshi Murata 1043 Shimoyama, Onigabo-ku, Tsukuba-shi, Ibaraki 1 Inside Kumagaya Gumi Technical Research Institute, Inc. 1 Kumagai Gumi Technical Research Institute, Inc. (56) References JP-A-6-147846 (JP, A) JP-A-5-180776 (JP, A) JP-A-4-120495 (JP, A) 4-336869 (JP, A) JP-A-2-96880 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G06T 7/00 E21D 9/00, 9/06

Claims (1)

(57) [Claims] 1. A color image of a tunnel face is obtained, and R, G, and B primary color signals of the color image are converted into a luminance signal and a color difference signal in accordance with a standard television signal, and saturation is represented from the color difference signal. represents the distribution of the hue in the color angle with obtaining the color vector, furthermore, by expressing the characteristic color of the color vector of each pixel is replaced with that of the maximum value, the human visual
A rock type and rock class identification method in image analysis of a tunnel face, wherein rock types and rock classes are identified from differences in hue .