A deformation measurement method and system based on rock surface texture characteristics

Abstract

The present invention discloses a deformation measurement method and system based on rock surface texture features, and specifically relates to the field of rock deformation measurement. The present invention first extracts local texture features before and after rock deformation through a local binary model to obtain a local binary image, and then based on the local binary image, uses a Harris corner point extraction algorithm to determine a first tracking point in a digital image before rock deformation, and uses a sparse optical flow tracking algorithm to determine a second tracking point in a digital image after rock deformation whose position is the same as the position of the first tracking point in the digital image before rock deformation, and finally obtains the displacement change before and after rock deformation according to the first tracking point and the second tracking point. It can be seen that the present invention realizes despeckle of rock deformation measurement by extracting local texture features, thereby improving the accuracy of rock deformation measurement.

Description

Technical Field

[0001] The invention relates to the field of rock deformation measurement, and in particular to a deformation measurement method and system based on rock surface texture characteristics. Background Art

[0002] Rock deformation measurement is an important way to study rock mechanics experiments, mechanical mechanisms, and fracture mutations. Digital image measurement technology is a commonly used and effective means of optical measurement, and is therefore widely used in the field of rock deformation measurement. Digital image measurement technology requires that the surface of the object being measured has randomly distributed scattered spots as a carrier for measuring the deformation of the object. The deformation information of the object being measured can be obtained by measuring the displacement of the scattered spots. In other words, the deformation measurement accuracy of digital image measurement technology is closely related to the quality of the scattered spots.

[0003] However, due to the low repeatability of scattered speckles, complicated manufacturing process and poor operability, the quality of actual scattered speckles varies, which in turn makes it impossible to guarantee the deformation measurement accuracy of the digital image measurement technology.

[0004] Therefore, how to achieve despeckle in rock deformation measurement and thus improve the accuracy of rock deformation measurement is an urgent problem to be solved. Summary of the invention

[0005] The purpose of the present invention is to provide a deformation measurement method and system based on rock surface texture characteristics, which can realize despeckle of rock deformation measurement and thus improve the accuracy of rock deformation measurement.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A deformation measurement method based on rock surface texture features, the method comprising the following steps:

[0008] Gray-scale processing is performed on the digital image before rock deformation to obtain a first gray-scale image; gray-scale processing is performed on the digital image after rock deformation to obtain a second gray-scale image;

[0009] Performing local binarization processing on the first grayscale image to obtain a first local binarization image; performing local binarization processing on the second grayscale image to obtain a second local binarization image;

[0010] Calculate the corner point feature value of each pixel in the first local binary image based on the Harris corner point extraction algorithm;

[0011] Taking the pixel point whose corner point feature value in the first local binary image is greater than a set threshold as the first tracking point;

[0012] Determine, based on a sparse optical flow tracking algorithm, a pixel point in the second local binary image whose position is the same as the position of the first tracking point in the first local binary image as a second tracking point;

[0013] The displacement change amount of the rock before and after deformation is obtained based on the first tracking point in the first local binary image and the second tracking point in the second local binary image.

[0014] Optionally, the grayscale processing of the digital image of the rock before deformation to obtain the first grayscale image specifically includes:

[0015] The digital image of the rock before deformation is grayed using the following formula to obtain a first grayed image;

[0016] gray(x,y)=0.299*R(x,y)+0.587*G(x,y)+0.114*B(x,y)

[0017] Where gray(x, y) is the gray value of the pixel point (x, y) in the first grayscale image, R(x, y), G(x, y) and B(x, y) are the R component value, G component value and B component value at the pixel point (x, y) in the digital image before rock deformation, respectively.

[0018] Optionally, performing local binarization processing on the first grayscale image to obtain a first local binarized image specifically includes:

[0019] Performing local binarization processing on the first grayscale image using the following formula to obtain a first local binarization image;

[0020]

[0021] Where I(x, y) is the local binary feature value of the pixel (x, y) in the first local binary image, p is the pth peripheral pixel of the window centered on the pixel (x, y), p = 1, 2, ..., 8; gray(x, y) is the gray value of the pixel (x, y) in the first grayscale image; gray(x p ,y p ) is the gray value of the pth peripheral pixel point in the window centered at the pixel point (x, y) in the first grayscale image; S(·) is the threshold function.

[0022] Optionally, calculating the corner point feature value of each pixel in the first local binary image based on the Harris corner point extraction algorithm specifically includes:

[0023] The following formula is used to calculate the corner point feature value of each pixel in the first local binary image based on the Harris corner point extraction algorithm;

[0024]

[0025] R=det(M)-k*trace 2 (M)

[0026] In the formula, I x and I y are the gradients of the local binary eigenvalues ​​of the pixel point (x, y) in the first local binary image in the x direction and the y direction respectively, M is the first local binary image matrix, w(·, ·) is the Gaussian function, det(M) is the determinant value of the first local binary image matrix, trace(M) is the trace of the first local binary image matrix, k is the empirical value, R is the corner point eigenvalue matrix, and each element of the corner point eigenvalue matrix is ​​the corner point eigenvalue of each pixel point in the first local binary image.

[0027] Optionally, obtaining the displacement change of the rock before and after deformation based on the first tracking point in the first local binary image and the second tracking point in the second local binary image specifically includes:

[0028] Calculate the difference between the corner feature value of the first tracking point and the corner feature value of the second tracking point as the feature difference I l ;

[0029] Based on the characteristic difference I l , the displacement change of the rock before and after deformation is obtained using the following formula;

[0030] e=∑ w I l [I x I y ] T

[0031]

[0032] d=G -1 e

[0033] Where d is the displacement change of the rock before and after deformation, w is the window centered at the pixel point (x, y), and I x and I y are the gradients of the local binary eigenvalues ​​of the pixel point (x, y) in the first local binary image in the x direction and the y direction respectively, and G and e are the first intermediate matrix and the second intermediate matrix respectively.

[0034] A deformation measurement system based on rock surface texture features, the system comprises an image grayscale module, a local binarization processing module, a corner feature value calculation module, a first tracking point determination module, a second tracking point determination module, and a displacement change calculation module;

[0035] The image grayscale module is used to grayscale the digital image before rock deformation to obtain a first grayscale image; and grayscale the digital image after rock deformation to obtain a second grayscale image.

[0036] The local binarization processing module is used to perform local binarization processing on the first grayscale image to obtain a first local binarization image; and perform local binarization processing on the second grayscale image to obtain a second local binarization image;

[0037] The corner point feature value calculation module is used to calculate the corner point feature value of each pixel in the first local binary image according to the Harris corner point extraction algorithm;

[0038] The first tracking point determination module is used to take the pixel point whose corner point feature value in the first local binary image is greater than a set threshold as the first tracking point;

[0039] The second tracking point determination module is used to determine, according to the sparse optical flow tracking algorithm, a pixel point in the second local binary image whose position is the same as the position of the first tracking point in the first local binary image as the second tracking point;

[0040] The displacement change calculation module is used to obtain the displacement change of the rock before and after deformation according to the first tracking point in the first local binary image and the second tracking point in the second local binary image.

[0041] Optionally, the image grayscale module performs grayscale processing on the digital image before rock deformation to obtain a first grayscale image, specifically including:

[0042] The digital image of the rock before deformation is grayed using the following formula to obtain a first grayed image;

[0043] gray(x,y)=0.299*R(x,y)+0.587*G(x,y)+0.114*B(x,y)

[0044] Where gray(x, y) is the gray value at the pixel point (x, y) in the first grayscale image, R(x, y), G(x, y) and B(x, y) are the R component value, G component value and B component value at the pixel point (x, y) in the digital image before rock deformation, respectively.

[0045] Optionally, the local binarization processing module performs local binarization processing on the first grayscale image to obtain a first local binarization image, specifically including:

[0046] Performing local binarization processing on the first grayscale image using the following formula to obtain a first local binarization image;

[0047]

[0048] Where I(x, y) is the local binary feature value at the pixel point (x, y) in the first local binary image, p is the p-th window surrounding pixel of the pixel point (x, y), p = 1, 2, ..., 8; g(x, y) is the gray value of the pixel point at the center of the window; g(x p ,y p ) is the gray value of the pixels around the window; S(·) is the threshold function.

[0049] Optionally, the corner point feature value calculation module calculates the corner point feature value of each pixel in the first local binary image according to the Harris corner point extraction algorithm, specifically including:

[0050] The following formula is used to calculate the corner point feature value of each pixel in the first local binary image based on the Harris corner point extraction algorithm;

[0051]

[0052] R=det(M)-k*trace 2 (M)

[0053] In the formula, I x ,I y is the gradient of the first local binary image I(x, y) in the x, y direction, M is the first local binary image matrix, w(·, ·) is the Gaussian function, det(M) is the determinant of the first local binary image matrix M, trace(M) is the trace of the first local binary image matrix M, k is an empirical value, and R is the corner point eigenvalue of each pixel in the first local binary image.

[0054] Optionally, the displacement change calculation module obtains the displacement change before and after the rock is deformed according to the first tracking point and the second tracking point, specifically including:

[0055] Calculate the difference between the corner point feature value of the first tracking point and the corner point feature value of the second tracking point as the feature difference I l ;

[0056] Based on the characteristic difference I l , the displacement change of the rock before and after deformation is obtained using the following formula;

[0057] e=∑ w I l [I x I y ] T

[0058]

[0059] d=G -1 e

[0060] Where d is the displacement change of the rock before and after deformation, w is the window centered at the pixel point (x, y), and I l is the characteristic difference, I x ,I y is the gradient of the first local binary image I(x, y) in the x, y directions, and G and e are intermediate matrices.

[0061] According to the specific embodiments provided by the present invention, the present invention discloses the following technical effects:

[0062] The present invention discloses a deformation measurement method and system based on rock surface texture features. First, the local texture features of the rock before and after deformation are extracted by a local binary model to obtain a local binary image. Then, based on the local binary image, a Harris corner point extraction algorithm is used to determine a first tracking point in a digital image before rock deformation. A sparse optical flow tracking algorithm is used to determine a second tracking point in the digital image after rock deformation whose position is the same as the position of the first tracking point in the digital image before rock deformation. Finally, the displacement change of the rock before and after deformation is obtained according to the first tracking point and the second tracking point. It can be seen that the present invention realizes despeckle of rock deformation measurement by extracting local texture features, thereby improving the accuracy of rock deformation measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0064] Figure 1 This is a flow chart of a deformation measurement method based on rock surface texture characteristics according to the present invention;

[0065] Figure 2 The rock specimen described in the present invention;

[0066] Figure 3 A digital image of the rock during deformation according to the present invention;

[0067] Figure 4 It is a local binary image during the rock deformation process described in the present invention;

[0068] Figure 5 It is the lateral displacement cloud map of the rock after deformation described in the present invention;

[0069] Figure 6 Schematic diagram of the pixel position described in the present invention. DETAILED DESCRIPTION

[0070] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0071] The purpose of the present invention is to provide a deformation measurement method and system based on rock surface texture features, which can achieve despeckle of rock deformation measurement by extracting local texture features, thereby improving the accuracy of rock deformation measurement.

[0072] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0073] Example 1

[0074] This embodiment provides a deformation measurement method based on rock surface texture features, and the specific steps are as follows:

[0075] S101 performs grayscale processing on the digital image before rock deformation to obtain a first grayscale image; and performs grayscale processing on the digital image after rock deformation to obtain a second grayscale image.

[0076] like Figure 2The figure shows the rock specimen of the present invention. In the figure, L, W, and H respectively represent the length, width, and height of the rock specimen. Before the grayscale processing, a rock direct shear test platform is built, and a CCD camera is used to collect digital images of the rock specimen before and after deformation. The frame rate of the CCD camera is 200fp. The loading system of the rock direct shear test platform is a RLW-3000 shear creep biaxial testing machine with a maximum axial pressure of 3000KN and a maximum lateral pressure of 1000KN. The specific loading method for rupturing the rock specimen using the loading system is as follows: first, the vertical load is added to 10KN, and the horizontal load is preloaded to 2KN, so that the rock specimen is fully coupled with the loading surface of the press, and then the horizontal loading is continued at a rate of 0.15mm / min until the rock specimen is destroyed. Figure 3 The digital images collected during the rock deformation process are shown. Figure 3 a- Figure 3 f is a digital image collected at different times. The digital image of the rock before deformation and the digital image of the rock after deformation in the present invention are Figure 3 a- Figure 3 Any two of f. Figure 3 a is a digital image collected when the loading time of the loading system is 1000s, Figure 3 b is a digital image acquired when the loading time of the loading system is 1500s, Figure 3 c is a digital image collected when the loading time of the loading system is 1800s, Figure 3 d is a digital image collected when the loading system is loaded for 1900s. Figure 3 e is a digital image captured at 1950s when the loading system was loaded, Figure 3 f is a digital image acquired when the loading time of the loading system is 2000s.

[0077] In this embodiment, the following formula is used to grayscale the digital image of the rock before deformation to obtain a first grayscale image:

[0078] gray(x,y)=0.299*R(x,y)+0.587*G(x,y)+0.114*B(x,y)

[0079] Where gray(x,y) is the gray value of the pixel point (x,y) in the first grayscale image, and R(x,y), G(x,y) and B(x,y) are the R component value, G component value and B component value at the pixel point (x,y) in the digital image before rock deformation, respectively.

[0080] S102: performing local binarization processing on the first grayscale image to obtain a first local binarization image; performing local binarization processing on the second grayscale image to obtain a second local binarization image.

[0081] Specifically, the LBP operator is defined as a 3×3 window, with the grayscale value of the central pixel of the window as the threshold, and the grayscale values ​​of the 8 pixels adjacent to the central pixel of the window are compared with the threshold one by one. If the surrounding pixel value is greater than the central pixel, the position is marked as 1, otherwise it is marked as 0. Then an 8-bit binary value is obtained, which is converted into a decimal value and then assigned to the pixel in the middle of the window to obtain a local binary image of the grayscale image.

[0082] In this embodiment, the first grayscale image is subjected to local binarization processing to obtain a first local binarized image, specifically: the first grayscale image is subjected to local binarization processing using the following formula to obtain a second local binarized image; Figure 4 The local binary image of the rock deformation process described in the present invention is shown. Figure 4 a- Figure 4 The local binary image of f is Figure 3 a- Figure 3 The digital images collected at different times in f correspond one to one;

[0083]

[0084] Where I(x, y) is the local binary feature value of the pixel (x, y) in the first local binary image, p is the pth peripheral pixel of the window centered on the pixel (x, y), p = 1, 2, ..., 8; gray(x, y) is the gray value of the pixel (x, y) in the first grayscale image; gray(x p ,y p ) is the gray value of the pth peripheral pixel point in the first grayscale image with the pixel point (x, y) as the center window; S(·) is the threshold function. Figure 6 As shown, c is the pixel point (x, y), and 1, 2, ..., 8 are the surrounding pixel points.

[0085] In this embodiment, local binarization is performed on the second grayscale image to obtain a second local binarized image, which is the same as the above process and will not be described herein.

[0086] S103 calculates the corner point feature value of each pixel in the first local binary image based on the Harris corner point extraction algorithm.

[0087] In this embodiment, the following formula is used to calculate the corner feature value of each pixel in the first local binary image based on the Harris corner extraction algorithm:

[0088]

[0089] R=det(M)-k*trace 2 (M)

[0090] In the formula, I x and I y are the gradients of the local binary eigenvalues ​​of the pixel point (x, y) in the first local binary image in the x direction and the y direction respectively, M is the first local binary image matrix, w(·, ·) is the Gaussian function, det(M) is the determinant value of the first local binary image matrix, trace(M) is the trace of the first local binary image matrix, k is the empirical value, R is the corner point eigenvalue matrix, and each element of the corner point eigenvalue matrix is ​​the corner point eigenvalue of each pixel point in the first local binary image.

[0091] S104: taking pixel points in the first local binary image whose corner point feature values ​​are greater than a set threshold as first tracking points.

[0092] In this embodiment, the threshold is set to 0.1. It should be noted that there is at least one first tracking point.

[0093] S105 determines, based on a sparse optical flow tracking algorithm (KLT, Kanade-Lucas-Tomasi tracking algorithm), a pixel point in the second local binary image whose position is the same as the position of the first tracking point in the first local binary image as a second tracking point.

[0094] Specifically, a sparse optical flow tracking algorithm is used to perform cross-correlation measurement on the first tracking point on the second local binary image to obtain a pixel point in the second local binary image whose position is the same as the position of the first tracking point in the first local binary image as the second tracking point.

[0095] S106: Obtain the displacement change of the rock before and after deformation based on the first tracking point in the first local binary image and the second tracking point in the second local binary image.

[0096] In this embodiment, the error ε is defined as:

[0097]

[0098] Where: W is the window centered at pixel (x, y), d x and d y are the displacements of the first local binary image in the x and y directions respectively, and I(x, y) is the first local binary image before deformation. x ,y+d y) first-order Taylor expansion, substituting it into the formula of error ε, and setting the differential of ε to 0, we get a formula of the form Gd = e, from which we can calculate the displacement change d = G -1 e.

[0099] Specifically, firstly, the difference between the corner point feature value of the first tracking point and the corner point feature value of the second tracking point is calculated as the feature difference I l ; Then based on the characteristic difference I l , the displacement change of the rock before and after deformation is obtained using the following formula;

[0100] e=∑ w I l [I x I y ] T

[0101]

[0102] d=G -1 e

[0103] Where d is the displacement change of the rock before and after deformation, w is the window centered at the pixel point (x, y), and I x and I y are the gradients of the local binary eigenvalues ​​of the pixel point (x, y) in the first local binary image in the x direction and the y direction, respectively, and G and e are the first intermediate matrix and the second intermediate matrix, respectively.

[0104] Furthermore, in order to realize the cloud image representation of rock deformation, the local binary images before and after the rock deformation are gridded to obtain the following Figure 5 The lateral displacement cloud diagram shown is Figure 5 a- Figure 5 The lateral displacement cloud diagram of f Figure 4 a- Figure 4 f has a one-to-one correspondence with the local binary image in f. If there are multiple tracking points in a single grid, the average displacement change of the tracking points is used to represent the displacement change of the local binary image in the grid; if there is a rare case where there is no tracking point in a single grid, a 3×3 grid window is defined with the grid as the center, and the displacement change of the 8 grids around it is averaged to represent the displacement change of the local binary image in the grid.

[0105] A deformation measurement system based on rock surface texture features, the system comprises an image grayscale module, a local binarization processing module, a corner feature value calculation module, a first tracking point determination module, a second tracking point determination module, and a displacement change calculation module;

[0106] The image grayscale module is used to grayscale the digital image before rock deformation to obtain a first grayscale image; and grayscale the digital image after rock deformation to obtain a second grayscale image.

[0107] The local binarization processing module is used to perform local binarization processing on the first grayscale image to obtain a first local binarization image; and perform local binarization processing on the second grayscale image to obtain a second local binarization image;

[0108] The corner point feature value calculation module is used to calculate the corner point feature value of each pixel in the first local binary image according to the Harris corner point extraction algorithm;

[0109] The first tracking point determination module is used to take the pixel point whose corner point feature value in the first local binary image is greater than a set threshold as the first tracking point;

[0110] The second tracking point determination module is used to determine, according to the sparse optical flow tracking algorithm, a pixel point in the second local binary image whose position is the same as the position of the first tracking point in the first local binary image as the second tracking point;

[0111] The displacement change calculation module is used to obtain the displacement change of the rock before and after deformation according to the first tracking point in the first local binary image and the second tracking point in the second local binary image.

[0112] Furthermore, the image grayscale module performs grayscale processing on the digital image before rock deformation to obtain a first grayscale image, which specifically includes:

[0113] The digital image of the rock before deformation is grayed using the following formula to obtain a first grayed image;

[0114] gray(x,y)=0.299*R(x,y)+0.587*G(x,y)+0.114*B(x,y)

[0115] Where gray(x, y) is the gray value at the pixel point (x, y) in the first grayscale image, R(x, y), G(x, y) and B(x, y) are the R component value, G component value and B component value at the pixel point (x, y) in the digital image before rock deformation, respectively.

[0116] Further, the local binarization processing module performs local binarization processing on the first grayscale image to obtain a first local binarization image, which specifically includes:

[0117] Performing local binarization processing on the first grayscale image using the following formula to obtain a first local binarization image;

[0118]

[0119] Where I(x, y) is the local binary feature value at the pixel point (x, y) in the first local binary image, p is the p-th window surrounding pixel of the pixel point (x, y), p = 1, 2, ..., 8; g(x, y) is the gray value of the pixel point at the center of the window; g(x p ,y p ) is the gray value of the pixels around the window; S(·) is the threshold function.

[0120] Further, the corner feature value calculation module calculates the corner feature value of each pixel in the first local binary image according to the Harris corner point extraction algorithm, specifically including:

[0121] The following formula is used to calculate the corner point feature value of each pixel in the first local binary image based on the Harris corner point extraction algorithm;

[0122]

[0123] R=det(M)-k*trace 2 (M)

[0124] In the formula, I x ,I y is the gradient of the first local binary image I(x, y) in the x, y direction, M is the first local binary image matrix, w(·, ·) is the Gaussian function, det(M) is the determinant of the first local binary image matrix M, trace(M) is the trace of the first local binary image matrix M, k is an empirical value, and R is the corner point eigenvalue of each pixel in the first local binary image.

[0125] Further, the displacement change calculation module obtains the displacement change before and after the rock is deformed according to the first tracking point and the second tracking point, specifically including:

[0126] Calculate the difference between the corner point feature value of the first tracking point and the corner point feature value of the second tracking point as the feature difference I l ;

[0127] Based on the characteristic difference I l , the displacement change of the rock before and after deformation is obtained using the following formula;

[0128] e=∑ w I l [I x I y ] T

[0129]

[0130] d=G -1 e

[0131] Where d is the displacement change of the rock before and after deformation, w is the window centered at the pixel point (x, y), and I l is the characteristic difference, I x ,I y is the gradient of the first local binary image I(x, y) in the x, y directions, and G and e are intermediate matrices.

[0132] In this specification, each embodiment is described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the embodiments can be referred to each other. For the system disclosed in the embodiment, since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part.

[0133] The principles and implementation methods of the present invention are described in this article using specific examples. The description of the above embodiments is only used to help understand the method and core idea of ​​the present invention. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as limiting the present invention.

Claims

1. A deformation measurement method based on rock surface texture characteristics, characterized in that: The method comprises the following steps: Gray-scale processing is performed on the digital image before rock deformation to obtain a first gray-scale image; gray-scale processing is performed on the digital image after rock deformation to obtain a second gray-scale image; Performing local binarization processing on the first grayscale image to obtain a first local binarization image; performing local binarization processing on the second grayscale image to obtain a second local binarization image; Calculate the corner point feature value of each pixel in the first local binary image based on the Harris corner point extraction algorithm; Taking the pixel point whose corner point feature value in the first local binary image is greater than a set threshold as the first tracking point; Determine, based on a sparse optical flow tracking algorithm, a pixel point in the second local binary image whose position is the same as the position of the first tracking point in the first local binary image as a second tracking point; Obtaining a displacement change amount of the rock before and after deformation based on a first tracking point in the first local binary image and a second tracking point in the second local binary image; The step of obtaining the displacement change of the rock before and after deformation based on the first tracking point in the first local binary image and the second tracking point in the second local binary image specifically includes: Calculate the difference between the corner feature value of the first tracking point and the corner feature value of the second tracking point as the feature difference I l ; Based on the characteristic difference I l , the displacement change of the rock before and after deformation is obtained using the following formula; e=∑ w I l [I x I y ] T ; d=G -1 e; Where d is the displacement change before and after the rock deformation, w is the window centered at the pixel point (x, y), and I x and I y are the gradients of the local binary eigenvalues ​​of the pixel point (x, y) in the first local binary image in the x direction and the y direction respectively, and G and e are the first intermediate matrix and the second intermediate matrix respectively.

2. The deformation measurement method based on rock surface texture characteristics according to claim 1 is characterized in that: The grayscale processing of the digital image before rock deformation to obtain the first grayscale image specifically includes: The digital image of the rock before deformation is grayed using the following formula to obtain a first grayed image; gray(x,y)=0.299*R(x,y)+0.587*G(x,y)+0.114*B(x,y); Where gray(x,y) is the gray value of the pixel point (x,y) in the first grayscale image, and R(x,y), G(x,y) and B(x,y) are the R component value, G component value and B component value at the pixel point (x,y) in the digital image before rock deformation, respectively.

3. The deformation measurement method based on rock surface texture characteristics according to claim 1 is characterized in that: The performing local binarization processing on the first grayscale image to obtain a first local binarization image specifically includes: Performing local binarization processing on the first grayscale image using the following formula to obtain a first local binarization image; Where I(x,y) is the local binary feature value of the pixel (x,y) in the first local binary image, p is the pth peripheral pixel of the window centered on the pixel (x,y), p = 1, 2, ..., 8; gray(x,y) is the gray value of the pixel (x,y) in the first grayscale image; gray(x p ,y p ) is the gray value of the pth peripheral pixel point in the window centered at the pixel point (x, y) in the first grayscale image; S(·) is the threshold function.

4. The deformation measurement method based on rock surface texture characteristics according to claim 1 is characterized in that: The calculating the corner point feature value of each pixel in the first local binary image based on the Harris corner point extraction algorithm specifically includes: The following formula is used to calculate the corner point feature value of each pixel in the first local binary image based on the Harris corner point extraction algorithm; R=det(M)-k*trace 2 (M); In the formula, I x and I y are the gradients of the local binary eigenvalues ​​of the pixel point (x, y) in the first local binary image in the x direction and the y direction respectively, M is the first local binary image matrix, w(·,·) is the Gaussian function, det(M) is the determinant value of the first local binary image matrix, trace(M) is the trace of the first local binary image matrix, k is the empirical value, R is the corner point eigenvalue matrix, and each element of the corner point eigenvalue matrix is ​​the corner point eigenvalue of each pixel point in the first local binary image.

5. A deformation measurement system based on rock surface texture features, characterized in that: The system includes an image grayscale module, a local binarization processing module, a corner feature value calculation module, a first tracking point determination module, a second tracking point determination module, and a displacement change calculation module; The image grayscale module is used to grayscale the digital image before rock deformation to obtain a first grayscale image; and grayscale the digital image after rock deformation to obtain a second grayscale image. The local binarization processing module is used to perform local binarization processing on the first grayscale image to obtain a first local binarization image; and perform local binarization processing on the second grayscale image to obtain a second local binarization image; The corner point feature value calculation module is used to calculate the corner point feature value of each pixel in the first local binary image according to the Harris corner point extraction algorithm; The first tracking point determination module is used to take the pixel point whose corner point feature value in the first local binary image is greater than a set threshold as the first tracking point; The second tracking point determination module is used to determine, according to the sparse optical flow tracking algorithm, a pixel point in the second local binary image whose position is the same as the position of the first tracking point in the first local binary image as the second tracking point; The displacement change calculation module is used to obtain the displacement change of the rock before and after deformation according to the first tracking point in the first local binary image and the second tracking point in the second local binary image; The displacement change calculation module obtains the displacement change before and after the rock is deformed according to the first tracking point and the second tracking point, specifically including: Calculate the difference between the corner point feature value of the first tracking point and the corner point feature value of the second tracking point as the feature difference I l ; Based on the characteristic difference I l , the displacement change of the rock before and after deformation is obtained using the following formula; e=∑ w I l [I x I y ] T ; d=G -1 e; Where d is the displacement change before and after the rock deformation, w is the window centered at the pixel point (x, y), and I x and I y are the gradients of the local binary eigenvalues ​​of the pixel point (x, y) in the first local binary image in the x direction and the y direction respectively, and G and e are the first intermediate matrix and the second intermediate matrix respectively.

6. The deformation measurement system based on rock surface texture characteristics according to claim 5 is characterized in that: The image grayscale module performs grayscale processing on the digital image before rock deformation to obtain a first grayscale image, specifically including: The digital image of the rock before deformation is grayed using the following formula to obtain a first grayed image; gray(x,y)=0.299*R(x,y)+0.587*G(x,y)+0.114*B(x,y); Where gray(x,y) is the gray value of the pixel point (x,y) in the first grayscale image, and R(x,y), G(x,y) and B(x,y) are the R component value, G component value and B component value at the pixel point (x,y) in the digital image before rock deformation, respectively.

7. The deformation measurement system based on rock surface texture characteristics according to claim 5, characterized in that: The local binarization processing module performs local binarization processing on the first grayscale image to obtain a first local binarization image, specifically including: Performing local binarization processing on the first grayscale image using the following formula to obtain a first local binarization image; Where I(x,y) is the local binary feature value of the pixel (x,y) in the first local binary image, p is the pth peripheral pixel of the window centered on the pixel (x,y), p = 1, 2, ..., 8; gray(x,y) is the gray value of the pixel (x,y) in the first grayscale image; gray(x p ,y p ) is the gray value of the pth peripheral pixel point in the window centered at the pixel point (x, y) in the first grayscale image; S(·) is the threshold function.

8. The deformation measurement system based on rock surface texture characteristics according to claim 5, characterized in that: The corner point feature value calculation module calculates the corner point feature value of each pixel in the first local binary image according to the Harris corner point extraction algorithm, specifically including: The following formula is used to calculate the corner point feature value of each pixel in the first local binary image based on the Harris corner point extraction algorithm; R=det(M)-k*trace 2 (M); In the formula, I x and I y are the gradients of the local binary eigenvalues ​​of the pixel point (x, y) in the first local binary image in the x direction and the y direction respectively, M is the first local binary image matrix, w(·,·) is the Gaussian function, det(M) is the determinant value of the first local binary image matrix, trace(M) is the trace of the first local binary image matrix, k is the empirical value, R is the corner point eigenvalue matrix, and each element of the corner point eigenvalue matrix is ​​the corner point eigenvalue of each pixel point in the first local binary image.

Images