Tunnel surrounding rock fissure positioning method and device

Abstract

The application discloses a tunnel surrounding rock fissure positioning method and device. The method comprises the following steps: acquiring a tunnel surrounding rock face image; acquiring a gray image of the tunnel surrounding rock face image; enhancing the gray image to obtain an enhanced image; extracting a fissure feature in the enhanced image; selecting at least two standard features of standard templates in a knowledge base which are close to the fissure feature; evaluating the similarity between the fissure feature and each standard feature; and selecting the standard feature with a similarity less than a threshold value as a real fissure of the fissure feature. The application firstly enhances the gray image of the tunnel surrounding rock face, and then compares the enhanced gray image with the standard template in the similarity comparison mode, so that the fissure of the tunnel surrounding rock face is quickly positioned.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of geological exploration, in particular, relates to a tunnel surrounding rock fracture positioning method and device. BACKGROUND

[0002] The soil types through which a tunnel passes are various, mainly including rock, silt, cohesive soil, sandy soil, pebble soil and gravel soil. Among them, rock is the hardest soil, with high density and large compressive strength, but also has special structures such as faults, joints and karst caves; silt and cohesive soil are prone to be affected by moisture due to their small particles and low density, so attention needs to be paid in tunnel design and construction.

[0003] The tunnel rock-soil body shows inherent rheological properties, and the deformation of the surrounding rock of the tunnel face also has certain rules. The excavation of the tunnel causes the surrounding rock supporting the tunnel body to be dug out, and the space behind the tunnel face is empty, resulting in the deformation of the surrounding rock towards the tunnel clearance. At the same time, with the tunnel construction, the current tunnel face will show deformation, cracking and other characteristics, and with the progress of the tunnel construction, the deformation, cracking and other characteristics of the tunnel face are positioned, which provides support for the analysis of the tunnel face and is one of the measures to improve the construction safety and efficiency. SUMMARY

[0004] Therefore, the first aspect of the embodiments of the present application discloses a tunnel surrounding rock fracture positioning method.

[0005] The method comprises,

[0006] obtaining a tunnel face image of the surrounding rock;

[0007] obtaining a gray-scale image of the tunnel face image;

[0008] enhancing the gray-scale image to obtain an enhanced image;

[0009] extracting a fracture feature in the enhanced image;

[0010] selecting at least two standard features of the standard templates in the knowledge base that are close to the fracture feature;

[0011] evaluating the similarity of the fracture feature and each standard feature;

[0012] selecting the standard feature with a similarity less than a threshold value as the real fracture of the fracture feature.

[0013] The enhanced image comprises,

[0014] obtaining the pixel point coordinates and gray-scale values of the gray-scale image;

[0015] establishing a pixel neighborhood of the center point with the pixel point as the center point.

[0016] obtaining the gray scale values of the pixels in the pixel neighborhood;

[0017] obtaining a weighted entropy value of the center point according to the number of the gray scale values in the pixel neighborhood, that is,

[0018] wherein, s i is the i-th gray scale value in the pixel neighborhood, s0 is the gray scale value of the center point, P si is the probability of the i-th gray scale value appearing in the pixel neighborhood, and m is the number of the types of the gray scale values of the pixels in the pixel neighborhood;

[0019] enhancing the gray scale value of the center point according to the weighted entropy value,

[0020] that is,

[0021] wherein, M is the pixel point set of the pixel neighborhood, h(x i ,y i ) is the weighted entropy value of the x i y i -th pixel point in the pixel neighborhood set;

[0022] establishing the enhanced image according to the enhanced gray scale value of the center point.

[0023] wherein, the crack feature is extracted by,

[0024] obtaining the first gradient component and the second gradient component of the gray scale values of the enhanced image in the horizontal direction and the vertical direction by using a gradient operator;

[0025] obtaining the crack feature according to the first gradient component and the second gradient component.

[0026] wherein, the crack feature is extracted by,

[0027] obtaining the crack feature of the enhanced image by using a Canny operator.

[0028] wherein, the evaluation of the crack feature and the standard feature includes,

[0029] calculating the feature similarity of the crack feature, that is,

[0030]

[0031]

[0032] wherein, L(x,y) is the gray scale value of the pixel point in the x is a gray value of a pixel point of an xy-th row and column of the standard template, N is a height or a width of the enhanced image or the standard template;

[0033] calculating a pixel distance value, i.e. D = 1 / S1 + 1 / S2;

[0034] selecting the standard feature of the standard template as the real crack of the crack feature when the pixel distance value of the standard template and the enhanced image is minimum and less than a distance threshold value.

[0035] wherein, evaluating the crack feature and the standard feature comprises,

[0036] selecting the standard feature of the standard template as the real crack of the crack feature when the pixel distance value of the standard template and the enhanced image is minimum and greater than the distance threshold value. wherein, n is a positive integer, an initial value of n is a random value or 1, and n is increased according to the iteration number;

[0037] returning to extracting the crack feature in the enhanced image after iterating the weighted entropy value of the center point.

[0038] wherein, when the pixel distance value of at least two standard templates and the enhanced image is less than the distance threshold value, obtaining original images corresponding to the standard templates and the gray image;

[0039] calculating the similarity of two original images and the gray image respectively;

[0040] selecting the standard template corresponding to the original image with the maximum similarity;

[0041] selecting the standard feature of the standard template as the real crack of the crack feature.

[0042] wherein, calculating the similarity of the original image and the gray image comprises,

[0043] calculating the cosine similarity of the original image and the gray image;

[0044] selecting the standard template corresponding to the original image with the maximum cosine similarity.

[0045] wherein, selecting the standard feature close to the crack feature in the knowledge base comprises,

[0046] rotating the enhanced image to obtain the crack feature in at least four directions;

[0047] evaluating the similarity of each direction of the crack feature and each standard feature;

[0048] The standard feature with a similarity less than a threshold value is selected as a real fissure of the fissure feature of the direction.

[0049] In addition, the second aspect of the embodiment of the present application discloses a tunnel surrounding rock fissure positioning device.

[0050] The device comprises an image acquisition module, an image processing module, a feature extraction module, a feature selection module and a feature evaluation module.

[0051] The image acquisition module is used to acquire the tunnel surrounding rock face image.

[0052] The image processing module is used to acquire the gray image of the tunnel surrounding rock face image, and to obtain the enhanced image by enhancing the gray image.

[0053] The feature extraction module is used to extract the fissure feature in the enhanced image.

[0054] The feature selection module is used to select the standard feature of at least two standard templates in the knowledge base which is close to the fissure feature.

[0055] The feature evaluation module is used to evaluate the similarity of the fissure feature and each standard feature, and to select the standard feature with a similarity less than a threshold value as a real fissure of the fissure feature.

[0056] Compared with the prior art, the gray image of the tunnel surrounding rock face is first enhanced, and then the enhanced gray image is compared with the standard template in terms of similarity, so that the fissure of the tunnel surrounding rock face is quickly positioned. During the tunnel excavation, the tunnel surrounding rock face as the working face of the tunnel excavation will continue to change. Quickly positioning the fissure of the tunnel surrounding rock face will be an important measure to balance the safety and efficiency of the tunnel excavation.

[0057] In view of the above scheme, the present application will be described in detail below with reference to the accompanying drawings, and other features and advantages of the present application will be made clear. BRIEF DESCRIPTION OF DRAWINGS

[0058] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope, and for those skilled in the art, other related drawings can also be obtained without creative labor.

[0059] Figure 1 The flowchart of the tunnel surrounding rock fissure positioning method of the present embodiment.

[0060] Figure 2A flowchart of a process for obtaining an enhanced image from a gray-scale image in the embodiment.

[0061] Figure 3 A flowchart of a process for extracting a crack feature in the enhanced image in the embodiment.

[0062] Figure 4 A flowchart of a process for evaluating the similarity between the crack feature and each standard feature in the embodiment.

[0063] Figure 5 A structural diagram of a tunnel surrounding rock crack positioning device in the embodiment. DETAILED DESCRIPTION

[0064] For the purpose of facilitating the understanding of the present application, a more comprehensive description of the present application will be given below with reference to the relevant drawings. The drawings show embodiments of the present application. However, the present application can be realized in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present application more thorough and comprehensive.

[0065] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs. The terms used in the specification of the present application herein are only for the purpose of describing specific embodiments and are not intended to limit the present application.

[0066] The embodiment discloses a tunnel surrounding rock crack positioning method.

[0067] Figure 1 A flowchart of a process for obtaining an enhanced image from a gray-scale image in the embodiment.

[0068] Figure 1 The tunnel surrounding rock crack positioning method includes steps 10 to 70.

[0069] 10 Obtain a tunnel surrounding rock face image.

[0070] Tunnel surrounding rock refers to the rock mass that has an impact on the stability of the tunnel after the tunnel is excavated. The tunnel surrounding rock face is the front part of the tunnel surrounding rock, i.e., the working face that is constantly advancing forward during the process of excavating the tunnel.

[0071] Obtaining a tunnel surrounding rock face image refers to continuously illuminating the tunnel surrounding rock face by a light source arranged in the tunnel, and obtaining a tunnel surrounding rock face image by a camera shooting the tunnel surrounding rock face at a fixed angle. The tunnel surrounding rock face image is a visible light image, and the visible light image is a multi-channel image.

[0072] 20 Obtain a gray-scale image of the tunnel surrounding rock face image.

[0073] A grayscale image is a single-channel image containing only grayscale information. Unlike a general visible light image, a grayscale image has only one color channel, and the grayscale value of each pixel represents its brightness level.

[0074] The grayscale image of the tunnel face image refers to converting the tunnel face image into a grayscale image using a weighted average method or a simple average method.

[0075] The weighted average method refers to weighting and averaging the RGB values of each pixel in the tunnel face image according to certain weights to obtain the corresponding grayscale value. The weights are usually determined based on the survey results of human eye sensitivity to different colors.

[0076] The simple average method refers to taking the average of the RGB values of each pixel in the tunnel face image as the grayscale value.

[0077] 30 The grayscale image is enhanced to obtain an enhanced image.

[0078] The enhanced grayscale image refers to enhancing the grayscale mutations of the grayscale image to improve the feature extraction effect of the enhanced grayscale image.

[0079] Figure 2 The flowchart of enhancing the grayscale image to obtain an enhanced image is shown in this embodiment.

[0080] Figure 2 The steps of enhancing the grayscale image include steps 31 to 36.

[0081] 31 Obtain the pixel point coordinates and grayscale values of the grayscale image.

[0082] 32 Take a pixel point as the center point and establish the pixel neighborhood of the center point.

[0083] The pixel neighborhood is a square region around the center point, and the region size is AxA, where A is the number of pixels in the horizontal and vertical directions of the region.

[0084] 33 Traverse the grayscale values of each pixel point in the pixel neighborhood.

[0085] 34 Obtain the weighted entropy value of the center point according to the number of grayscale values in the pixel neighborhood,

[0086] i.e.

[0087] where s i is the i-th grayscale value in the pixel neighborhood, s0 is the grayscale value of the center point, is the probability of the i-th grayscale value in the pixel neighborhood, and m is the number of types of grayscale values in the pixel neighborhood. The probability of the i-th grayscale value in the pixel neighborhood is the ratio of the number of times the grayscale value appears in the region to the total number of pixels in the region.

[0088] 35 updating the gray value of the center point according to the weighted entropy value,

[0089] that is

[0090] wherein M is a pixel point set of a pixel neighborhood, h(x i ,y i ) is a weighted entropy value of an x i y i th pixel point in the set of the pixel neighborhood.

[0091] 36 establishing an enhanced image according to the gray value of all the center points after enhancement.

[0092] Then, the embodiment realizes the enhancement of the gray image through steps 31 to 36, can effectively suppress the background of the roof surface image, and enhances the effect of the crack feature extraction.

[0093] 40 extracting the crack feature in the enhanced image.

[0094] Figure 3 Fig. 4 is a flowchart of the process of extracting the crack feature in the enhanced image according to the embodiment.

[0095] Figure 3 The extraction of the crack feature in the enhanced image includes steps 41 to 43.

[0096] 41 obtaining the first gradient component and the second gradient component of the gray value of the enhanced image in the horizontal direction and the vertical direction by using the sobel gradient operator.

[0097] The sobel gradient operator is a gradient-based edge detection method, which detects the edge by calculating the gradient of the image in the x direction and the y direction. The sobel gradient operator uses two 3x3 convolution kernels to calculate the gradient in the horizontal direction and the vertical direction respectively, and then combines the two gradients to obtain the edge strength. The sobel gradient operator has the characteristics of simple calculation and high efficiency.

[0098] Specifically, obtaining the first gradient component of the enhanced image in the horizontal direction by using the sobel gradient operator includes,

[0099]

[0100] wherein G x is the gray value obtained by the convolution operation in the horizontal direction by using the sobel gradient operator, and x ij is the gray value of each pixel point of the enhanced image in the horizontal direction.

[0101] And, obtaining the first gradient component of the enhanced image in the vertical direction by using the sobel gradient operator includes,

[0102]

[0103] Gx= Gx- Gx- 1 y is the gray value obtained by convolution operation in horizontal direction using sobel gradient operator, y ij is the gray value of each pixel point in horizontal direction of the enhanced image.

[0104] 42obtaining the edge strength of each pixel point according to the first gradient component and the second gradient component.

[0105] Specifically, obtaining the edge strength of the pixel point comprises,

[0106]

[0107] Gx= Gx- Gx- 1 x,y is the edge strength value of the pixel point.

[0108] 43obtaining the crack feature according to the edge strength of each pixel point.

[0109] The edge strength value of the pixel point is the amplitude of the edge point gradient. The edge strength is the place where the pixel value in the image changes significantly, which corresponds to the boundary or contour of the enhanced image. Considering that the crack of the palm surface is usually a long crack extending in at least one direction, in 43, a pixel point with an edge strength exceeding a threshold value is selected, and then a search is performed along an angle range with the pixel point to obtain adjacent pixels that also exceed the threshold value. The set of these pixels is a crack feature extracted.

[0110] In some embodiments, step 40 can obtain one or more crack features of the enhanced image using a Canny operator. The Canny operator is a multi-stage edge detection algorithm, including noise filtering, gradient calculation, non-maximum suppression and double threshold detection. The Canny operator can extract edge information with high precision and low error rate, and has the characteristics of high edge detection precision, can detect small edges in the image, and has good noise suppression effect.

[0111] 50selecting standard features of at least two standard templates in the knowledge base that are close to the crack feature.

[0112] When there are multiple crack features extracted in step 40, step 50 is performed on each crack feature to select a standard template.

[0113] The standard features of the standard template selected according to the crack feature can be the standard features of the corresponding standard template preliminarily screened according to the length and curvature of the crack feature. The standard features are close to the crack feature in length and / or curvature.

[0114] 60 evaluate the similarity of the fissure feature to each standard feature.

[0115] Figure 4 A flowchart for evaluating the similarity of the fissure feature to each standard feature in this embodiment is shown.

[0116] Figure 4 The evaluation of the similarity includes steps 61 to 62.

[0117] 61 calculate the feature similarity of the fissure feature, i.e.

[0118]

[0119] where L(x, y) is the gray value of the pixel in the xth row and yth column of the enhanced image, is the gray value of the pixel in the xth row and yth column of the standard template, and N is the height and width of the image size of the enhanced image or the standard template.

[0120] 62 calculate the pixel distance value, i.e. D = 1 / S1 + 1 / S2.

[0121] 70 select the standard feature of the standard template as the real fissure of the fissure feature when the similarity is less than a threshold value.

[0122] Select the standard feature of the standard template as the real fissure of the fissure feature when there is a standard template with the minimum pixel distance value and the only one less than a distance threshold value.

[0123] When there are at least two standard templates with pixel distance values less than the distance threshold value, obtain the original images corresponding to the standard templates and the gray image; calculate the similarity of the two original images and the gray image, respectively; select the standard template corresponding to the original image with the maximum similarity; and select the standard feature of the standard template as the real fissure of the fissure feature.

[0124] Preferably, the calculation of the similarity of the original image and the gray image includes calculating the cosine similarity of the original image and the gray image, and then selecting the original image corresponding to the standard template with the maximum cosine similarity.

[0125] When there is a standard template with the minimum pixel distance value and greater than the distance threshold value, iterate the weighted entropy value of each center point as

[0126] where n is a positive integer, the initial value of n is a random value or 1, the value of n is increased according to the iteration number, and n≤5. After iterating the weighted entropy value of the center point, return to extract the fissure feature in the enhanced image, i.e. return to 40. If n>5, it is judged that there is no fissure feature in the enhanced image.

[0127] Preferably, when selecting the standard features close to the crack features in the knowledge base, the enhanced image is rotated to obtain at least four directions of crack features, and then the similarity of each direction of crack features to each standard feature is evaluated; and the standard feature with a similarity less than a threshold is selected as a real crack of the crack features in the direction.

[0128] Therefore, the tunnel surrounding rock crack positioning method of the embodiment first enhances the gray image of the working face, and then quickly positions the cracks of the working face by using the similarity comparison between the enhanced gray image and the standard template.

[0129] The embodiment of the application discloses a tunnel surrounding rock crack positioning device.

[0130] Figure 5 The figure is a structural schematic diagram of the tunnel surrounding rock crack positioning device of the embodiment.

[0131] The positioning device comprises an image acquisition module, an image processing module, a feature extraction module, a feature selection module and a feature evaluation module.

[0132] The image acquisition module is used to acquire the working face image of the tunnel surrounding rock.

[0133] The image processing module is used to acquire the gray image of the working face image, and enhance the gray image to obtain an enhanced image.

[0134] The feature extraction module is used to extract the crack features in the enhanced image.

[0135] The feature selection module is used to select the standard features of at least two standard templates close to the crack features in the knowledge base.

[0136] The feature evaluation module is used to evaluate the similarity of the crack features to each standard feature, and select the standard feature with a similarity less than a threshold as a real crack of the crack features.

[0137] Note that the above is only the preferred embodiment of the application and the applied technical principle. It is understood by those skilled in the art that the application is not limited to the specific embodiments herein, and those skilled in the art can make various obvious changes, readjustments and substitutions without departing from the protection scope of the application. Therefore, although the application is described in more detail through the above embodiments, the application is not limited to the above embodiments, and more other equivalent embodiments can be included without departing from the concept of the application, and the scope of the application is determined by the appended claims.

Claims

1. A method for locating fractures in tunnel surrounding rock, characterized in that, The method includes, Obtain images of the tunnel face of the surrounding rock; Obtain the grayscale image of the face of the machine; The grayscale image is enhanced to obtain an enhanced image; Extract the crack features within the enhanced image; Select standard features from at least two standard templates in the knowledge base that are close to the crack features; Evaluate the pixel distance value between the crack feature and each of the standard features; The crack type of the standard template corresponding to the standard feature whose pixel distance value is less than a threshold is selected as the true crack of the crack feature.

2. The method for locating fractures in tunnel surrounding rock according to claim 1, characterized in that, The enhanced image is obtained by: Obtain the pixel coordinates and grayscale values ​​of the grayscale image; Using one of the aforementioned pixels as the center point, establish the pixel neighborhood of the center point; The grayscale value of each pixel in the neighborhood of the pixel is obtained by traversing the neighborhood. The weighted entropy value of the center point is obtained based on the number of gray values ​​in the pixel's neighborhood. Right now Among them, s i Let be the gray value of the i-th element in the neighborhood of the pixel, and let s0 be the gray value of the center point. Let m be the probability of the i-th gray value appearing in the pixel neighborhood, and m be the number of gray value types of the pixel in the pixel neighborhood. The grayscale value of the center point is enhanced based on the weighted entropy value. Right now Where M is the set of pixels in the neighborhood of the pixel, h(x) i ,y i ) is the x-th pixel in the set of the pixel neighborhood. i y i The weighted entropy value of each pixel; The enhanced image is constructed based on the grayscale value of the center point after enhancement.

3. The method for locating fractures in tunnel surrounding rock according to claim 2, characterized in that, Extracting the crack features includes, The gradient operator is used to obtain the first gradient component and the second gradient component of the gray value in the horizontal and vertical directions of the enhanced image; The crack features are obtained based on the first gradient component and the second gradient component.

4. The method for locating fractures in tunnel surrounding rock according to claim 2, characterized in that, Extracting the crack features includes, The crack features of the enhanced image are obtained using the Canny operator.

5. The method for locating fractures in tunnel surrounding rock according to claim 4, characterized in that, Evaluating the crack characteristics and the standard characteristics includes, Calculate the feature similarity of the crack features, i.e. Where L(x,y) is the gray value of the pixel in the xy-th row and column of the enhanced image. The grayscale value of the pixel in the xy-th row and column of the standard template is denoted as , and N is the height and width of the enhanced image or the standard template. Calculate the pixel distance value, i.e., D = 1 / S1 + 1 / S2; When a standard template and the enhanced image have the smallest pixel distance value that is uniquely less than a distance threshold, the crack type of the standard template corresponding to the standard feature is selected as the true crack of the crack feature.

6. The method for locating fractures in tunnel surrounding rock according to claim 5, characterized in that, Evaluating the crack characteristics and the standard characteristics includes, When a standard template is selected with the smallest pixel distance value to the enhanced image that is greater than the distance threshold, the weighted entropy value of each center point is iterated. Where n is a positive integer, the initial value of n is a random value or 1, and the value of n increases with the number of iterations; After iterating over the weighted entropy value of the center point, the crack features within the enhanced image are extracted.

7. The method for locating fractures in tunnel surrounding rock according to claim 5, characterized in that, When at least two of the standard templates have pixel distance values ​​less than the distance threshold with respect to the enhanced image, the original image corresponding to the standard template and the grayscale image is obtained. Calculate the similarity between the two original images and the grayscale image respectively; Select the standard template corresponding to the original image with the highest similarity; Select the crack type of the standard template corresponding to the standard feature as the actual crack of the crack feature.

8. The method for locating fractures in tunnel surrounding rock according to claim 7, characterized in that, Calculating the similarity between the original image and the grayscale image includes, Calculate the cosine similarity between the original image and the grayscale image; Select the standard template corresponding to the original image with the highest cosine similarity.

9. The method for locating fractures in tunnel surrounding rock according to claim 1, characterized in that, The standard features selected from the knowledge base that are similar to the crack features include... Rotating the enhanced image yields the crack features in at least four directions; Evaluate the pixel distance value between the crack feature in each direction and each of the standard features; Select the true crack of the standard template corresponding to the standard feature whose pixel distance value is less than a threshold as the crack type of the crack feature in one direction.

10. A tunnel surrounding rock fissure positioning device, characterized in that, The device includes an image acquisition module, an image processing module, a feature extraction module, a feature selection module, and a feature evaluation module; The image acquisition module is used to acquire images of the tunnel face of the surrounding rock. The image processing module is used to acquire a grayscale image of the face of the working face and enhance the grayscale image to obtain an enhanced image; The feature extraction module is used to extract crack features within the enhanced image; The feature selection module is used to select standard features from at least two standard templates in the knowledge base that are close to the crack feature; The feature evaluation module is used to evaluate the pixel distance value between the crack feature and each of the standard features, and selects the crack type of the standard template corresponding to the standard feature whose pixel distance value is less than a threshold as the true crack of the crack feature.

Images