Roadway surrounding rock fracture form measuring method

Abstract

The invention records a roadway surrounding rock fracture morphology measurement method, which comprises the following steps of S1, installing a photogrammetry device in a roadway, and obtaining a continuous digital image of a roadway surrounding rock area to be measured; s2, carrying out image acquisition on the structural surface of the surrounding rock of the roadway to obtain a color original image, and converting the color original image into a grayscale image; s3, dividing the grayscale image into a plurality of local areas, establishing a grayscale histogram of the local areas, and equalizing the grayscale histogram of the local areas to form an enhanced grayscale image; s4, performing adaptive gamma correction on the enhanced grayscale image; s5, smoothing the enhanced grayscale image by using a median filter; s6, using a Sobel operator to carry out crack contour extraction on the enhanced grayscale image; and S7, calculating to obtain the real length, the real width and the crack inclination angle of the tunnel surrounding rock structural surface crack. According to the invention, safe, direct, efficient and accurate acquisition of the surrounding rock fracture image of the coal mine tunnel and rapid identification and extraction of surrounding rock fracture parameters can be realized.

Description

Technical Field

[0001] The invention relates to the technical field of coal mine tunnel surrounding rock structure engineering measurement, and in particular to a tunnel surrounding rock crack morphology measurement method. Background Art

[0002] Under the dual effects of mining stress and original ground stress, coal mine tunnel excavation working faces frequently suffer from safety accidents and disasters such as roof falls, rock bursts, and rock bursts. The initial manifestation of such safety accidents and disasters is the generation and development of structural surface cracks in the tunnel surrounding rock. Therefore, timely perception and development prediction of tunnel surrounding rock cracks is an indispensable cornerstone for achieving safe and efficient production in the coal mining industry, and is also an important means to prevent safety accidents and disasters in coal mines and improve economic benefits.

[0003] At present, underground coal mining industry generally adopts measurement methods such as line surveying, 3D laser scanning technology, and precision measurement to accurately identify cracks in the surrounding rock structure of tunnels. However, the above methods also have the following shortcomings or defects:

[0004] ① The line measurement method relies on manual measurement, which results in low safety factor and low efficiency in actual operation;

[0005] ② The 3D laser scanning technology is too slow to obtain the surrounding rock crack data of the tunnel, and the data is too long and difficult to analyze;

[0006] ③ The precision measurement method requires a lot of manpower and material resources, is difficult to apply to practical operations on the production site, and has poor practicality.

[0007] Therefore, there is an urgent need to provide a method for measuring the crack morphology of the surrounding rock of coal mine tunnels, so as to achieve safe, direct, efficient and accurate acquisition of the crack morphology of the surrounding rock of coal mine tunnels, and provide strong support for coal mine safety production. Summary of the invention

[0008] In order to solve the problems existing in the above-mentioned prior art, the present invention provides a method for measuring the crack morphology of surrounding rock of a tunnel, and its specific technical solution is as follows:

[0009] A method for measuring crack morphology of tunnel surrounding rock comprises the following steps:

[0010] S1, installing a photogrammetric device in the tunnel, adjusting the spatial parameters of the photogrammetric device in the tunnel, and obtaining continuous digital images of the surrounding rock area of ​​the tunnel to be measured;

[0011] S2, using the photogrammetry device to collect images of the structural surface of the tunnel surrounding rock, obtain a color original image, and convert the color original image into a grayscale image;

[0012] S3, dividing the grayscale image into a plurality of local regions, establishing a grayscale histogram of the local regions, equalizing the grayscale histogram of the local regions, and merging the equalized results to form an enhanced grayscale image;

[0013] S4, performing adaptive gamma correction on the enhanced grayscale image;

[0014] S5, using a median filter to smooth the enhanced grayscale image;

[0015] S6, using the Sobel operator to extract the crack contour from the enhanced grayscale image;

[0016] S7, calculate the true length, true width and crack inclination of the cracks in the tunnel surrounding rock structure surface.

[0017] Furthermore, the spatial parameters of the photogrammetric device in the tunnel include the photography working distance, photography height, visible rock surface area width, and field of view; the photography working distance is the vertical distance between the center point of the lens of the photography device and the surrounding rock of the tunnel to be measured; the photography height is the height between the center point of the lens of the photography device and the bottom plate of the tunnel; the calculation formula of the visible rock surface area width is:

[0018]

[0019] Where W f is the width of the visible rock surface area of ​​the tunnel surrounding rock, mm; a is the length of the camera projection sensor, mm; b is the width of the camera projection sensor, mm; H f is the height of the visible rock surface area, mm; D w is the photographic working distance, mm; F ​​is the focal length of the camera lens, mm;

[0020] The field of view L f The calculation formula is:

[0021] Where, L f is the field of view, mm 2 ; S d is the camera projection sensor size, which is the product of the sensor length a and width b, mm 2 ;D w is the photographic working distance, mm; F ​​is the focal length of the camera lens, mm.

[0022] Furthermore, the steps of converting the color original image into a grayscale image are:

[0023] Use MATLAB to obtain the R, G, and B component values ​​of each pixel in the color original image, convert the R, G, and B component values ​​into a single grayscale value, and use the grayscale value to replace the R, G, and B component values ​​of the pixels in the dot matrix image of the color original image.

[0024] Furthermore, the gray value conversion formula is:

[0025] I(x,y)=0.299×I r (x,y)+0.587×I g (x,y)+0.114×I b (x,y)

[0026] Where I(x,y) is the grayscale value corresponding to the pixel of the original image; I r (x,y),I g (x,y),I b (x,y) is the RGB value of the pixel in the original image.

[0027] Furthermore, the cumulative distribution function is calculated for the grayscale histogram of the local area to identify the total frequency of the local area from the lowest grayscale value to the current grayscale value; the cumulative distribution function calculation formula is:

[0028]

[0029] Where CDF(i) is the cumulative distribution function of the grayscale histogram; P(j) is the pixel ratio of grayscale value j in the grayscale histogram.

[0030] Furthermore, each pixel in the local area is transformed by the cumulative distribution function, and its grayscale value is remapped to the equalized value, so that the grayscale histogram of the local area is equalized. Specifically, the CDF is normalized to the range [0, L-1], where L is the grayscale level (8-bit image, L=256), then: NewValue=round(CDF(i)×(L-1)).

[0031] Furthermore, the grayscale image is divided into a plurality of local regions using MATLAB, and the size of the local region is preferably set to 5×5; the grayscale value of the pixel point in each of the local regions is obtained, and a grayscale histogram corresponding to the local region is established using MATLAB.

[0032] Furthermore, the conversion formula for adaptive gamma correction is:

[0033]

[0034] In the formula, GC(r k ) is the adaptive gamma correction normalization transformation; r k is the kth grayscale of the input; r k maxis the maximum grayscale of the input; γ is the adjustment parameter: γ<1 maps the input grayscale to a higher grayscale range, which brightens the image, while γ>1 has the opposite effect, which darkens the image.

[0035] Furthermore, the cumulative distribution function is used to perform image adaptive correction of the enhanced grayscale image, and the weight distribution function is used to fine-tune the grayscale histogram of the local area; the weight distribution function expression is as follows:

[0036] γ=1-cdf w (r k )

[0037] in,

[0038]

[0039] therefore,

[0040]

[0041] Right now,

[0042]

[0043] In the formula,

[0044] γ is the weight distribution function; r k is the kth gray level in the local area; pdf max is the maximum value of the grayscale probability density function; pdf min is the minimum value of the grayscale probability density function; α is the adjustment parameter; r j is the jth gray level in the local area; pdf w (r j ) is the maximum value of the j-th grayscale probability density function.

[0045] Furthermore, the Sobel operator includes the following steps:

[0046] First, define the Sobel convolution kernel:

[0047]

[0048] In the formula, G x is the horizontal convolution kernel; G y is the vertical convolution kernel.

[0049] Second, the grayscale smoothed image is convolved using the Sobel convolution kernel to obtain the x and y direction gradients:

[0050]

[0051] In the formula, G x(x,y) is the gradient in the x direction; G y (x,y) is the gradient in the y direction; G x (i, j) is the x-direction gradient of the convolution kernel at the horizontal coordinate i and the vertical coordinate j; G y (i, j) is the y-direction gradient of the convolution kernel at the horizontal coordinate i and the vertical coordinate j; I g (x+i, y+j) represents the pixel value of the grayscale smoothed image at the position (x+i, y+j).

[0052] Furthermore, non-maximum suppression is performed on the enhanced grayscale image, including the following steps:

[0053] First, the gradient magnitude G(x,y) and direction θ(x,y) of each pixel in the enhanced grayscale image are calculated by the following formula:

[0054]

[0055] Where G(x,y) is the gradient amplitude of each pixel in the enhanced grayscale image; G x (x,y) is the gradient in the x direction; G y (x, y) is the gradient in the y direction; θ(x, y) is the gradient direction of each pixel in the enhanced grayscale image.

[0056] Second, non-maximum suppression is performed on the enhanced grayscale image. Specifically, the indexes of two adjacent pixels in the gradient direction are set to p and q, and adjacent pixels in the direction of G(x,y) are selected for comparison. Then, if G(x,y) < G(p,q), G(x,y) is suppressed to 0; otherwise, G(x,y) is retained.

[0057] Furthermore, edge extraction is performed on the enhanced grayscale image, and a high threshold T is set. h and low threshold T l , specifically: if G(x,y)>T h , then it is marked as an edge point E(x,y)=1; if G(x,y)<T l , then it is marked as a non-edge point E(x,y)=0; if T l <G(x,y)<T h , then the connectivity is used to determine: if this pixel is connected to any pixel marked as an edge, then E(x, y) = 1; otherwise, set E(x, y) = 0.

[0058] Furthermore, the calculation of the real length, real width and crack inclination of the structural surface crack of the tunnel surrounding rock specifically includes the following steps:

[0059] First, define the actual length of the structural surface crack of the tunnel surrounding rock as l, the width as u, and the pixel length as l1 , pixel width is u 1 , can be obtained by the following formula:

[0060]

[0061] Where, l is the true length of the crack in the structural surface of the tunnel surrounding rock, mm; u is the true width of the crack in the structural surface of the tunnel surrounding rock, mm; l 1 is the pixel length of the structural surface crack of the tunnel surrounding rock, mm; u 1 is the pixel width of the structural surface crack of the tunnel surrounding rock, mm; P s D is the distance on the sensor represented by the unit pixel distance; w is the photographic working distance, mm; F ​​is the focal length of the camera lens, mm.

[0062] Second, the pixel connection line of the grayscale image after the crack enhancement of the tunnel surrounding rock structure surface is approximated as a straight line, and the crack inclination angle of the tunnel surrounding rock structure surface is calculated based on the linear regression algorithm;

[0063] The regression formula of the linear regression algorithm is:

[0064] y=cx+d

[0065]

[0066] In the formula, y represents the dependent variable, which is the target variable that needs to be predicted or explained; x represents the independent variable, which is the factor or predictor variable that affects y; and are the average coordinates of the crack pixel skeleton points on the pixel line; x i With y i are the horizontal and vertical coordinates of the i-th point respectively; N is the number of crack skeleton points; c is the regression coefficient; d is the intercept of the regression equation.

[0067] The calculation formula of the crack inclination angle of the tunnel surrounding rock structure surface is:

[0068] δ = arctanc.

[0069] Where δ is the crack inclination angle of the tunnel surrounding rock structure surface; c is the regression coefficient.

[0070] Based on the above technical solution, the present invention has the following beneficial effects:

[0071] 1. Based on high-precision cameras and precise spatial positioning, crack identification of the entire space of the tunnel surrounding rock can be realized, and the development status of the surrounding rock cracks can be accurately judged.

[0072] 2. Divide the grayscale image into regions, use the Sobel operator to quickly find the outer contour of each connected region, and smooth the image to improve image accuracy, avoiding the redundancy and inefficiency of traditional algorithms.

[0073] 3. Mining the mechanistic information hidden in a large amount of surrounding rock monitoring data to provide a reference for the prevention and control of disasters such as tunnel roof collapse and rock burst.

[0074] 4. By monitoring and analyzing the deformation of the surrounding rock, potential safety hazards can be discovered in time, providing strong guarantees for safe production in coal mines. At the same time, it can also provide a scientific basis for tunnel support design, construction plan optimization, etc., and further improve the economic and social benefits of coal mining.

[0075] 5. Through the technical means of combining digital photogrammetry technology with digital image processing technology, the safe, direct, efficient and accurate acquisition of coal mine tunnel surrounding rock crack images and the rapid identification and extraction of surrounding rock crack parameters can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 : Schematic diagram of the technical route of the recording method of the present invention;

[0077] Figure 2 :Schematic diagram of the layout of the photogrammetric measurement device for surrounding rock cracks in coal mine tunnels;

[0078] Figure 3 : Schematic diagram of the specific implementation steps of the method of recording the present invention. DETAILED DESCRIPTION

[0079] It should be noted that:

[0080] 1. Certain words are used in the specification and claims to refer to specific components or structures. Those skilled in the art should understand that technicians may use different terms to refer to the same component or structure. This specification and claims do not use the difference in terms as a way to distinguish components or structures, but use the difference in the functions of components or structures as the criterion for distinction.

[0081] 2. Unless otherwise clearly specified and limited, the terms "installation", "setting", "connection", etc. should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be the internal connection of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0082] 3. Unless otherwise defined, the technical or scientific terms used in this disclosure shall have the common meanings understood by persons with ordinary skills in the field to which this disclosure belongs. The words "first", "second" and similar words used in this disclosure do not indicate any order, quantity or importance, but are only used to distinguish different components.

[0083] The following is combined with Figure 1 -Attached Figure 3 , the present invention is described in detail by way of embodiments.

[0084] This embodiment describes a method for measuring the crack morphology of surrounding rock in a tunnel, which specifically includes the following steps.

[0085] S1, installing a photogrammetric device in the tunnel, and obtaining or setting spatial parameters of the photogrammetric device in the tunnel, including but not limited to photographic working distance, photographic height, width of visible rock surface area, and field of view.

[0086] As attached Figure 2 As shown, the main body of the photogrammetric device is a high-precision explosion-proof digital camera 1 and a laser rangefinder 2, and is equipped with lighting equipment 3 such as a fill light, a diffuse light, and an adjustable tripod 4 to ensure that the light source is sufficient and uniform during the photographing / videoing process to avoid shadows and reflections, and to adjust the height of the camera and the rangefinder relative to the tunnel floor and the distance relative to the tunnel surrounding rock according to the actual conditions of the tunnel.

[0087] When installing the photogrammetric device, in order to obtain a clear and complete image of the surrounding rock surface of the tunnel, it is preferred to install the photogrammetric device near the center area of ​​the tunnel.

[0088] In order to ensure the measurement process is efficient, fast and accurate, and to obtain continuous digital images of large areas of tunnel surrounding rock, it is necessary to reasonably set the photography working distance, photography height, visible rock surface area width, and field of view. The specific steps are as follows:

[0089] S101, using a measuring tool such as a ruler or a laser rangefinder to obtain a photographic working distance between a measuring object and a photogrammetric device, and a photographic height of the photogrammetric device from a tunnel floor.

[0090] Taking a digital camera as an example, the photographic working distance refers to the vertical distance between the center point of the camera lens and the measuring body, and the photographic height refers to the height distance between the center point of the camera lens and the bottom plate of the tunnel.

[0091] As a preferred method, the photogrammetry device is set on the central axis and the central waistline of the tunnel. This setting can more conveniently measure the parameters of the surrounding rock structure surface on both sides of the tunnel, obtain continuous digital images of large areas of the tunnel surrounding rock, and make the obtained digital image of the tunnel surrounding rock more complete.

[0092] ①The calculation formula for photographic working distance is:

[0093]

[0094] In the formula,

[0095] D W is the photographic working distance, mm;

[0096] D h is the lane width, mm;

[0097] ②The calculation formula for the photography height is:

[0098]

[0099] In the formula,

[0100] H W is the photographic height, mm;

[0101] H h is the tunnel height, mm.

[0102] S102, determining a reasonable and applicable shooting range of the surrounding rock of the tunnel to be measured, including but not limited to the width of the visible rock surface area and the field of view of the surrounding rock of the tunnel.

[0103] ① The calculation formula for the width of the visible rock surface area of ​​the tunnel surrounding rock is:

[0104]

[0105] In the formula,

[0106] W f is the width of the visible rock surface area of ​​the tunnel surrounding rock, mm;

[0107] a is the camera projection sensor length, mm;

[0108] b is the camera projection sensor width, mm;

[0109] H f is the height of the visible rock surface area, mm;

[0110] The calculation formula for the height of the visible rock surface area is:

[0111]

[0112] In the formula,

[0113] H f is the height of the visible rock surface area, mm;

[0114] a is the camera projection sensor length, mm;

[0115] Dw is the photographic working distance, mm;

[0116] F is the focal length of the camera lens, mm.

[0117] ② Field of view L f The calculation formula is:

[0118]

[0119] In the formula,

[0120] L f is the field of view, mm 2 ;

[0121] S d is the camera projection sensor size, which is the product of the sensor length a and width b, mm 2 ;

[0122] D w is the photographic working distance, mm;

[0123] F is the focal length of the camera lens, mm.

[0124] S2, according to the spatial position of the photogrammetry device determined in step S1, the structural surface of the tunnel surrounding rock is imaged to obtain a color original image; the collected original image data is transmitted to the computer in real time via a wireless communication protocol, and the color original image is converted into a grayscale image.

[0125] The specific steps are as follows:

[0126] Use MATLAB to obtain the R, G, and B component values ​​of each pixel in the color original image, convert the R, G, and B component values ​​into a single grayscale value, and use the grayscale value to replace the R, G, and B component values ​​of the pixels in the dot matrix image of the color original image.

[0127] The gray value conversion formula is:

[0128] I(x,y)=0.299×I r (x,y)+0.587×I g (x,y)+0.114×I b (x,y)

[0129] In the formula,

[0130] I(x,y) is the gray value corresponding to the pixel of the original image;

[0131] I r (x,y),I g (x,y),I b (x,y) is the RGB value of the pixel in the original image.

[0132] Converting the original color image into a grayscale image preserves the brightness information in the original image while reducing the interference of color information.

[0133] S3, using the local mean filter denoising method to denoise the grayscale image to obtain an enhanced grayscale image.

[0134] The specific steps include:

[0135] S301, using MATLAB to divide the grayscale image into multiple local areas, and preferably setting the size of the local area to 5×5.

[0136] Regional subdivision of grayscale images can avoid the formation of large-scale high-brightness or low-brightness concentration and suppress uneven brightness distribution.

[0137] S302, obtaining the grayscale value of the pixel points in each local area, and using MATLAB to establish a grayscale histogram corresponding to the local area.

[0138] The grayscale histogram can intuitively show the frequency of occurrence of each gray level in a local area.

[0139] S303, calculating the cumulative distribution function for the grayscale histogram of each local area, thereby representing the total frequency of the local area from the lowest grayscale value to the current grayscale value.

[0140] The cumulative distribution function calculation formula is:

[0141]

[0142] In the formula,

[0143] CDF(i) is the cumulative distribution function of the grayscale histogram;

[0144] P(j) is the proportion of pixels with gray value j in the gray histogram.

[0145] S304, transforming each pixel in the local area by using a cumulative distribution function, remapping its grayscale value to a balanced value, thereby equalizing the grayscale histogram of the local area.

[0146] Normalize the CDF to the range [0, L-1], where L is the number of gray levels (for 8-bit image, L=256), then: NewValue=round(CDF(i)×(L-1)).

[0147] S305, combining the equalization results of all local regions to form an enhanced grayscale image.

[0148] S4, performing adaptive gamma correction on the enhanced grayscale image, improving the image brightness of the enhanced grayscale image, and enhancing the difference between the rock surface and the cracks of the tunnel surrounding rock.

[0149] The specific steps include:

[0150] S401, based on the traditional gamma correction, adaptive gamma correction conversion is performed, and the normalized form of the conversion formula is GC(r k )for:

[0151]

[0152] In the formula,

[0153] GC(r k ) is the adaptive gamma correction normalization transformation;

[0154] r k is the kth grayscale of the input;

[0155] r k max is the maximum grayscale of the input;

[0156] γ is an adjustment parameter: γ<1 maps the input grayscale to a higher grayscale range, which brightens the image, while γ>1 has the opposite effect, which darkens the image.

[0157] S402, using the cumulative distribution function to perform image adaptive correction on the enhanced grayscale image, and using the weight distribution function to fine-tune the grayscale histogram of the local area to avoid the enhanced grayscale image being too bright or too dark.

[0158] The weight distribution function expression is as follows:

[0159] γ=1-cdf w (r k )

[0160] in,

[0161]

[0162] therefore,

[0163]

[0164] Right now,

[0165]

[0166] In the formula,

[0167] γ is the weight distribution function;

[0168] r jis the j-th gray level in the local area;

[0169] pdf max is the maximum value of the grayscale probability density function;

[0170] pdf min is the minimum value of the grayscale probability density function;

[0171] α is the adjustment parameter;

[0172] r j is the jth gray level in the local area;

[0173] pdf w (r j ) is the maximum value of the j-th grayscale probability density function.

[0174] S5, select a median filter with a size of 3×3 to smooth the enhanced grayscale image. S6, use the Sobel operator to extract the crack contour of the enhanced grayscale image.

[0175] The specific steps include:

[0176] S601, define the Sobel convolution kernel:

[0177]

[0178] In the formula,

[0179] G x It is a transverse convolution;

[0180] G y It is the longitudinal convolution.

[0181] S602, convolve the grayscale smoothed image using the Sobel convolution kernel to obtain the x and y direction gradients:

[0182]

[0183] In the formula,

[0184] G x (x,y) is the gradient in the x direction;

[0185] G y (x,y) is the gradient in the y direction;

[0186] G x (i, j) is the x-direction gradient of the convolution kernel at the horizontal coordinate i and the vertical coordinate j;

[0187] G y (i, j) is the x-direction gradient of the convolution kernel at the horizontal coordinate i and the vertical coordinate j;

[0188] I g (x+i, y+j) represents the pixel value of the grayscale smoothed image at the position (x+i, y+j).

[0189] S603, calculating the gradient magnitude G(x, y) and direction θ(x, y) of each pixel in the enhanced grayscale image by the following formula:

[0190]

[0191] In the formula,

[0192] G(x,y) is the gradient amplitude of each pixel in the enhanced grayscale image;

[0193] G x (x,y) is the gradient in the x direction;

[0194] G y (x,y) is the gradient in the y direction;

[0195] θ(x,y) is the gradient direction of each pixel in the enhanced grayscale image.

[0196] S604, performing non-maximum suppression on the enhanced grayscale image, specifically:

[0197] Set the indexes of two adjacent pixels in the gradient direction to p and q, and select adjacent pixels in the G(x,y) direction for comparison, then:

[0198] If G(x,y)<G(p,q), then suppress G(x,y) to 0;

[0199] Otherwise, keep G(x,y).

[0200] S605, edge extraction is performed on the enhanced grayscale image, and a high threshold T is set. h and low threshold T l , specifically:

[0201] If G(x,y)>T h , then it is marked as an edge point E(x,y)=1;

[0202] If G(x,y)<T l , then it is marked as a non-edge point E(x,y)=0;

[0203] If T l <G(x,y)<T h , then judge based on connectivity:

[0204] If this pixel is connected to any pixel marked as an edge, then E(x, y) = 1; otherwise, set E(x, y) = 0.

[0205] S7, calculating geometric parameters of cracks on the structural surface of the tunnel surrounding rock, including but not limited to the true length, true width, and crack inclination of the structural surface cracks of the tunnel surrounding rock, specifically comprising the following steps:

[0206] S701, define the actual length of the crack in the surrounding rock structure of the tunnel as l, the width as u, and the pixel length as l 1 , pixel width is u 1 , can be obtained by the following formula:

[0207]

[0208] In the formula,

[0209] l is the true length of the structural surface crack of the tunnel surrounding rock, mm;

[0210] u is the true width of the crack in the structural surface of the tunnel surrounding rock, mm;

[0211] l 1 is the pixel length of the structural surface crack of the tunnel surrounding rock, mm;

[0212] u 1 is the pixel width of the structural surface crack of the tunnel surrounding rock, mm;

[0213] P s is the distance between adjacent pixels, mm;

[0214] D w is the photographic working distance, mm;

[0215] F is the focal length of the camera lens, mm.

[0216] S702, calculate the inclination angle γ of the cracks on the surrounding rock structure surface of the tunnel.

[0217] The pixel connection line of the grayscale image of the tunnel surrounding rock structure surface after crack enhancement is approximated as a straight line, and the crack inclination is calculated based on the linear regression algorithm.

[0218] The regression formula is:

[0219] y=cx+d

[0220] In the formula,

[0221]

[0222] In the formula,

[0223] y represents the dependent variable, which is the target variable that needs to be predicted or explained;

[0224] x represents the independent variable, which is the factor or predictor variable that affects y;

[0225] c is the regression coefficient;

[0226] d is the intercept of the regression equation.

[0227] and are the average values ​​of the coordinates of the crack pixel skeleton points on the pixel line;

[0228] x i With y i are the horizontal and vertical coordinates of the i-th point respectively;

[0229] N is the number of crack skeleton points.

[0230] From this, we can get the crack inclination angle

[0231] δ = arctanc.

[0232] In the formula,

[0233] δ is the crack inclination angle of the tunnel surrounding rock structure surface;

[0234] c is the regression coefficient.

[0235] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments, and the above embodiments and descriptions are only for explaining the principles of the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A method for measuring crack morphology of surrounding rock of a tunnel, characterized in that: The following steps are involved: S1, installing a photogrammetric device in the tunnel, adjusting the spatial parameters of the photogrammetric device in the tunnel, and obtaining continuous digital images of the surrounding rock area of ​​the tunnel to be measured; S2, using the photogrammetry device to collect images of the structural surface of the tunnel surrounding rock, obtain a color original image, and convert the color original image into a grayscale image; S3, dividing the grayscale image into a plurality of local regions, establishing a grayscale histogram of the local regions, equalizing the grayscale histogram of the local regions, and merging the equalized results to form an enhanced grayscale image; S4, performing adaptive gamma correction on the enhanced grayscale image; S5, using a median filter to smooth the enhanced grayscale image; S6, using the Sobel operator to extract the crack contour from the enhanced grayscale image; S7, calculate the true length, true width and crack inclination of the cracks in the tunnel surrounding rock structure surface.

2. A method for measuring crack morphology of tunnel surrounding rock according to claim 1, characterized in that: The spatial parameters of the photogrammetric device in the tunnel include the photographic working distance, the photographic height, the width of the visible rock surface area, and the field of view; The photographic working distance is the vertical distance between the center point of the lens of the photographic device and the surrounding rock of the tunnel to be measured; The photographic height is the height between the center point of the lens of the photographic device and the bottom plate of the tunnel; The calculation formula for the width of the visible rock surface area is: In the formula, W f is the width of the visible rock surface area of ​​the tunnel surrounding rock, mm; a is the length of the camera projection sensor, mm; b is the width of the camera projection sensor, mm; H f is the height of the visible rock surface area, mm; D w is the photographic working distance, mm; F ​​is the focal length of the camera lens, mm; The field of view L f The calculation formula is: In the formula, L f is the field of view, mm 2 ; S d is the camera projection sensor size, which is the product of the sensor length a and width b, mm 2 ;D w is the photographic working distance, mm; F ​​is the focal length of the camera lens, mm.

3. A method for measuring crack morphology of tunnel surrounding rock according to claim 1, characterized in that: The steps to convert the color original image into a grayscale image are: Use MATLAB to obtain the R, G, and B component values ​​of each pixel in the color original image, convert the R, G, and B component values ​​into a single grayscale value, and use the grayscale value to replace the R, G, and B component values ​​of the pixels in the dot matrix image of the color original image.

4. A method for measuring crack morphology of tunnel surrounding rock according to claim 3, characterized in that: The gray value conversion formula is: I(x,y) = 0.299×I r (x,y)+0.587×I g (x,y)+0.114×I b (x,y) In the formula, I(x,y) is the grayscale value corresponding to the pixel point of the original image; I r (x,y),I g (x,y),I b (x,y) is the RGB value of the pixel in the original image.

5. The method for measuring crack morphology of tunnel surrounding rock according to claim 1, characterized in that: Calculating a cumulative distribution function for the grayscale histogram of the local area to identify the total frequency of the local area from the lowest grayscale value to the current grayscale value; The cumulative distribution function calculation formula is: Where CDF(i) is the cumulative distribution function of the grayscale histogram; P(j) is the pixel ratio of grayscale value j in the grayscale histogram.

6. A method for measuring crack morphology of tunnel surrounding rock according to claim 1, characterized in that: Each pixel in the local area is transformed by the cumulative distribution function, and its grayscale value is remapped to the equalized value, so that the grayscale histogram of the local area is equalized, specifically: Normalize the CDF to the range [0, L-1], where L is the number of gray levels (for 8-bit image, L=256), then: NewValue=round(CDF(i)×(L-1)).

7. A method for measuring crack morphology of tunnel surrounding rock according to claim 1, characterized in that: Use MATLAB to divide the grayscale image into multiple local regions, and preferably set the size of the local region to 5×5; The grayscale value of each pixel in the local area is obtained, and a grayscale histogram corresponding to the local area is established using MATLAB.

8. A method for measuring crack morphology of tunnel surrounding rock according to claim 1, characterized in that: The conversion formula for adaptive gamma correction is: In the formula, GC(r k ) is the adaptive gamma correction normalization transformation; r k is the kth grayscale of the input; is the maximum grayscale of the input; γ is the adjustment parameter: γ<1 maps the input grayscale to a higher grayscale range, which brightens the image, while γ>1 has the opposite effect, which darkens the image.

9. A method for measuring crack morphology of tunnel surrounding rock according to claim 1, characterized in that: The cumulative distribution function is used to perform image adaptive correction on the enhanced grayscale image, and the weighted distribution function is used to fine-tune the grayscale histogram of the local area; The weight distribution function expression is as follows: In the formula, γ is the weight distribution function; r j is the jth gray level in the local area; pdf w (r j ) is the maximum value of the j-th grayscale probability density function.

10. A method for measuring crack morphology of tunnel surrounding rock according to claim 1, characterized in that: The Sobel operator includes the following steps: First, define the Sobel convolution kernel: In the formula, G x is the horizontal convolution kernel; G y is the vertical convolution kernel. Second, the grayscale smoothed image is convolved using the Sobel convolution kernel to obtain the x and y direction gradients: In the formula, G x (x,y) is the gradient in the x direction; G y (x,y) is the gradient in the y direction; G x (i, j) is the x-direction gradient of the convolution kernel at the horizontal coordinate i and the vertical coordinate j; G y (i, j) is the y-direction gradient of the convolution kernel at the horizontal coordinate i and the vertical coordinate j; I g (x+i, y+j) represents the pixel value of the grayscale smoothed image at the position (x+i, y+j).

11. A method for measuring crack morphology of tunnel surrounding rock according to claim 10, characterized in that: The non-maximum suppression is performed on the enhanced grayscale image, including the following steps: First, the gradient magnitude G(x,y) and direction θ(x,y) of each pixel in the enhanced grayscale image are calculated by the following formula: In the formula, G(x,y) is the gradient amplitude of each pixel in the enhanced grayscale image; G x (x,y) is the gradient in the x direction; G y (x, y) is the gradient in the y direction; θ(x, y) is the gradient direction of each pixel in the enhanced grayscale image. Second, non-maximum suppression is performed on the enhanced grayscale image, specifically: Set the indices of two adjacent pixels in the gradient direction to p and q, and select adjacent pixels in the direction of G(x,y) for comparison. Then: if G(x,y)<G(p,q), suppress G(x,y) to 0; otherwise, retain G(x,y).

12. A method for measuring crack morphology of tunnel surrounding rock according to claim 11, characterized in that: Perform edge extraction on the enhanced grayscale image and set a high threshold T h and low threshold T l , specifically: If G(x,y)>T h , then it is marked as an edge point E(x,y)=1; If G(x,y)<T l , then it is marked as a non-edge point E(x,y)=0; If T l <G(x,y)<T h , then the connectivity is used to determine: if this pixel is connected to any pixel marked as an edge, then E(x, y) = 1; otherwise, set E(x, y) = 0.

13. A method for measuring crack morphology of tunnel surrounding rock according to claim 1, characterized in that: Calculating the true length, true width, and inclination of the structural surface cracks of the tunnel surrounding rock specifically includes the following steps: First, define the real length of the structural surface crack of the tunnel surrounding rock as l, the width as u, the pixel length as l1, and the pixel width as u1, which can be obtained by the following formula: In the formula, l is the real length of the structural surface crack of the tunnel surrounding rock, mm; u is the real width of the structural surface crack of the tunnel surrounding rock, mm; l1 is the pixel length of the structural surface crack of the tunnel surrounding rock, mm; u1 is the pixel width of the structural surface crack of the tunnel surrounding rock, mm; P s D is the distance on the sensor represented by the unit pixel distance; w is the photographic working distance, mm; F ​​is the focal length of the camera lens, mm. Second, the pixel connection line of the grayscale image after the crack enhancement of the tunnel surrounding rock structure surface is approximated as a straight line, and the crack inclination angle of the tunnel surrounding rock structure surface is calculated based on the linear regression algorithm; The regression formula of the linear regression algorithm is: y=cx+d In the formula, y represents the dependent variable, which is the target variable that needs to be predicted or explained; x represents the independent variable, which is the factor or predictor variable that affects y; and are the average coordinates of the crack pixel skeleton points on the pixel line; x i With y i are the horizontal and vertical coordinates of the i-th point respectively; N is the number of crack skeleton points; c is the regression coefficient; d is the intercept of the regression equation; The calculation formula of the crack inclination angle of the tunnel surrounding rock structure surface is: v=arctanc. In the formula, δ is the crack inclination angle of the tunnel surrounding rock structure surface; c is the regression coefficient.

Images