Intelligent Matching Design Method for Crack Identification and Charging in Blastholes of Underground Engineering Blasting

Abstract

The present invention provides a method for intelligent matching design of crack identification and charging in blast holes for underground engineering, which relates to the technical field of tunnel and underground engineering equipment. The crack information in the blast holes is obtained in an automated manner, and the cracks can be accurately identified and the geometric information of the cracks can be extracted by using image recognition and image processing methods; on the basis of the RQD method, four indexes, namely the crack ratio, the surface crack ratio, the fracture area and the broken area, are considered to obtain a new index LRQD. A method based on the correction coefficient α is proposed to correct the peak pressure P that the blast hole can actually withstand, and then the charging structure is automatically adjusted to achieve intelligent matching design; the crack conditions in the blast holes are identified and processed through the developed in-hole camera hardware device and software system, and then the blasting design scheme is optimized and adjusted, so as to reduce the influence of blasting on rock mass damage, improve the utilization rate of explosive blasting energy, and have certain guiding significance for underground engineering construction.

Description

Technical Field

[0001] The invention belongs to the technical field of tunnel and underground engineering equipment, and particularly relates to a method for identifying cracks in underground engineering blasting holes and intelligent matching design of charge loading. Background Technique

[0002] With the rapid development of social economy, a large number of important underground projects have been built in China, involving traffic tunnels, underground mine exploitation, etc. With the increasing construction of underground projects, theories, construction methods, equipment and equipment are also developing rapidly. However, in complex geological conditions, the blasting process needs to be adjusted in a timely manner according to the actual situation. At present, there is still a strong uncertainty in the construction process, which is a complex and high-risk systematic project.

[0003] Blasting technology, as a commonly used means of excavation in engineering, is widely used in projects such as tunnels and mine exploitation. While bringing high engineering economic benefits, blasting will also cause a series of problems such as rock mass damage. Therefore, the blasting effect can be improved through the charge structure. However, there are many problems with traditional charge loading methods, resulting in high construction costs and potential safety hazards. Among them, the distribution characteristics of rock mass cracks can reflect the crack development inside the rock mass. When the number of cracks is large, manual measurement often requires a lot of manpower and time. Relying on digital extraction of cracks can greatly improve the speed and efficiency of obtaining crack information. It is of great significance to gradually shift from manual to digital methods for intelligent extraction of crack information and intelligent adjustment of the charge structure in blasting engineering. The charge structure is an important factor in adjusting the effect of explosive energy in blasting. With the development of blasting technology, different types of explosive varieties have emerged continuously, and new charge structures using different explosives have also emerged continuously, but there is still a large amount of loss of blasting explosive energy.

[0004] Considering the influence of crack factors on the blasting effect, quantitative analysis of rock mass cracks is realized based on deep learning and image processing technology, and further exploration of the charge structure is carried out. A rock mass crack detection system and a charge intelligent matching design system are designed, and a downhole camera device is built to realize the supporting development of software and hardware. Furthermore, the blasting design scheme is optimized and adjusted to reduce the influence of blasting on rock mass damage and improve the utilization rate of explosive blasting energy, which has certain engineering significance in tunnel and underground engineering construction. Summary of the Invention

[0005] The purpose of the embodiment of the invention is to provide a method for identifying cracks in underground engineering blasting holes and intelligent matching design of charge loading, so as to alleviate the technical problems of large damage to the rock mass and low utilization rate of explosives in the existing technology during the blasting process.

[0006] To solve the above technical problems, the embodiment of the invention provides the following technical solutions:

[0007] The first aspect of the present invention provides a method for intelligent matching design of crack identification and charging in blast holes for underground engineering, including the following steps:

[0008] S10. Obtain the rock mass crack image and preprocess the rock mass crack image; establish a data set in the deep learning network model;

[0009] S20. Build a deep learning network model; use the improved network model to perform image recognition on the processed image;

[0010] S30. According to the rock mass crack identification result in step S20, by extracting the identified crack feature area, crack skeleton extraction and splicing, and crack set information acquisition, segment the crack from the background and quantitatively express it;

[0011] S40. According to the recognition result in step S30, analyze the charging structure in the blast hole and establish an intelligent matching design system for charging based on crack identification in the blast hole;

[0012] S50. Establish an in-hole camera system.

[0013] In some modified embodiments of the first aspect of the present invention, step S30 includes the following steps:

[0014] S31. Use the fuzzy C-means clustering algorithm to perform image segmentation operation on the rock mass crack image;

[0015] S32. Use the connected domain of the least squares method to remove small noise in the image;

[0016] S33. Use the Zhang-Suen thinning algorithm to extract the crack skeleton line and refine the crack image;

[0017] S34. Use the SURF algorithm for feature point detection and the matching method for crack splicing, including feature point detection, descriptor extraction, feature point matching, geometric transformation, and image fusion; adopt the weighted average fusion algorithm to splice and fuse two images, and fuse them into a complete image by superposition or mixing;

[0018] S35. Use the Canny algorithm to extract the edge features of the rock mass crack image;

[0019] Adopt the orthogonal skeleton algorithm to calculate the crack width;

[0020] Adopt the neighborhood scanning method based on the skeleton curve to calculate the crack length;

[0021] Calculate the crack area, and the steps are as follows:

[0022] S351. Calibrate the scale of the image according to the scale markers in the image or other known dimension information;

[0023] S352. Use an edge detection algorithm to extract the fracture contour and mark the fracture area;

[0024] S353. Use the pixel counting function in image processing software or programming languages to calculate the number of pixels inside the fracture area. This will give the pixel area of the fracture area.

[0025] S354. Multiply the pixel area by the actual area of each pixel to obtain the actual area of the fracture.

[0026] In some modified embodiments of the first aspect of the present invention, step S31 includes the following steps:

[0027] S311. First, input the data of the rock mass fracture image into the algorithm, set the number of rock mass fracture categories for clustering, the fuzzy index p, and the initial clustering center, etc.;

[0028] S312. Set the threshold for stopping iteration and the maximum number of iterations;

[0029] S313. Then update the membership degree and calculate the clustering center;

[0030] S314. Determine whether the conditions of the normalization measure function are satisfied. If satisfied, stop the operation of the algorithm and output the membership degree and the clustering center;

[0031] S315. Divide the image pixel points according to the principle of maximum membership degree to achieve image segmentation of the rock mass fractures;

[0032] Its expression is: Given a data set containing n pixels: x = {x1, x2,..., xn}, define the objective function of the fuzzy C-means (FCM) clustering algorithm as follows:

[0033]

[0034]

[0035] In the formula: u mn is the fuzzy membership degree of the pixel point x n relative to the m-th clustering center z m , p is the fuzzy index, d mn is the gray-scale distance between x n and the clustering center z m ;

[0036] Find the minimum value of the objective function J. Under the condition of satisfying the constraint conditions, use the Lagrange multiplier method for the objective function, and let J be for v m and umn The partial derivative is 0, so the membership matrix u can be obtained. mn and the cluster center v m The specific expressions are as follows:

[0037]

[0038]

[0039] In some modified embodiments of the first aspect of the present invention, step 32 includes the following steps:

[0040] S321. Find all connected regions in the rock mass fracture image and obtain the number of pixel points contained in each of their connected regions;

[0041] S322. Determine the threshold range of the algorithm by using the least absolute deviations method;

[0042] S323. Determine the threshold range of the algorithm according to the median value of the threshold;

[0043] S324. Process the connected regions of the rock mass fracture image according to the limitation of the threshold range;

[0044] The expression of the least absolute deviations method is:

[0045]

[0046] where: a i is the coordinate information of the pixel point;

[0047] Step S33 includes the following steps:

[0048] S331: Scan pixel by pixel. By scanning the image pixel by pixel, find the candidate pixels to be deleted; the judgment conditions include the black and white pattern around the pixel and the connectivity of the pixel. For the foreground color, it will be detected whether the following conditions are met:

[0049] The current pixel is white, at least one of the upper, lower, left, right, upper left, lower left, upper right, and lower right of the 8-neighborhood of the current pixel P is black, and the number of white pixels in the 8-neighborhood of the current pixel is between 2 and 6. If all the above conditions are met, then the current pixel is marked as black;

[0050] S332: Use this algorithm to traverse each pixel of the image again. For the foreground color, it will be checked whether it meets the following conditions:

[0051] The current pixel is white. Among the 8-neighborhood of the current pixel, at least one is black. Among the 8-neighborhood of the current pixel, the number of white pixels is between 2 and 6. The black pixels in the 8-neighborhood of the current pixel are not connected, that is, not a connected component. If all the above conditions are met, then mark the current pixel as black. After repeated iterations, based on the first step, continuously iterate and apply the rules until no pixel points can be deleted. After the iteration ends, an image after thinning processing is obtained.

[0052] In some modified embodiments of the first aspect of the present invention, step S34 includes the following steps:

[0053] S341. Determine the key point positions by detecting local extreme points in the image. Let (x, y) be a pixel point in the image, then the image of this point can be defined as the sum of the values of all pixel points in the rectangular area enclosed by the image origin and this point. The mathematical expression is as follows:

[0054]

[0055] S342. Construct a Hessian matrix to generate stable mutation points in the image. The Hessian matrix can be expressed as:

[0056]

[0057] In the formula: σ is the Gaussian scale; (x, y) is the position of the image pixel point; L xx (x, y, σ) represents the convolution of the second-order Gaussian differential at the pixel point (x, y) of the image I. The same applies to other Ls;

[0058] S343. Generate a descriptor for each key point to describe the gradient information and distribution of the area around the key point; match the key points in the image, and use a similarity measure based on the descriptor to find the corresponding key points in the image; through the matched key points, different transformation models can be used to estimate the geometric transformation relationship between the images;

[0059] S344. According to the estimated geometric transformation relationship, splice and fuse the two images, and fuse them into a complete image by superposition or mixing; adopt a weighted average fusion algorithm, and the weighting factor is u. The mathematical expression for the weighted fusion of the pixel values of the image is as follows:

[0060]

[0061] In the formula: u is the weighting factor; x, y are coordinates; f 1 、f 2 represent images;

[0062] In some modified embodiments of the first aspect of the present invention, the mathematical expression of the two-dimensional Gaussian function adopted by the Gaussian filter in the Canny algorithm is specifically as follows:

[0063]

[0064] S = G × I

[0065] Where: G represents the Gaussian filter obtained by operating on the two-dimensional Gaussian function; I represents the image to be detected; S represents the smoothed image;

[0066] For the smoothed image, the magnitude and direction of the gradient of each pixel point are calculated. The components of the gradient of the pixel points on the edge in the vertical and horizontal directions (x, y) are as follows:

[0067]

[0068] The pixel point gradient magnitude g θ and direction θ are respectively:

[0069]

[0070] According to the given high and low thresholds, strong and weak edges are divided, and isolated weak edges are eliminated to obtain the final detection result; two thresholds T 1 , T 2 are set; points with a gradient value larger than T 1 are strong edge points, and vice versa are weak edge points;

[0071] The orthogonal skeleton line method uses the KD-tree algorithm, and the expression is as follows:

[0072]

[0073] Where: K is the dimension; d is the distance between data points, a i and y i are data points;

[0074] After using the KD-tree algorithm to find the given k nearest neighbor points, the normal vector of the skeleton line is calculated through singular value decomposition, and then the two crack edge line points closest to the normal vector of the skeleton line are found, and the intersection point of the line segment formed by these two points and the normal vector of the skeleton line is calculated. The expression of singular value decomposition is as follows:

[0075] Suppose there is a bilinear function: f(x, y) = x T Ay

[0076] where x, y belong to R n×1 , A belongs to R n×n , by introducing a linear transformation, the bilinear function can be alternatively expressed as: f(x, y) = (Uξ)T A(Vη) = ξ T U T AVη

[0077] Next, let: S = U T AV

[0078] Then the bilinear function can be written as: f(x, y) = ξ T Sη

[0079] If both U and V are orthogonal matrices, S can be made a diagonal matrix through appropriate U and V, specifically expressed as: S = diag(σ 1 , σ 2 ,..., σ n )

[0080] Multiply both sides by U and VT, which can be expressed as: A = USV T ;

[0081] The neighborhood scanning method based on the skeleton curve is used to calculate the crack length; the steps are as follows:

[0082] S351. First, skeletonize the fissure, and then use morphological operations to remove the burrs on the fissure skeleton line;

[0083] S352. Scan the global image to find the endpoints of the crack, and search for fissure pixels in the 8-neighborhood method starting from the endpoints and count the number;

[0084] S353. Multiply the obtained number of fissure pixels by the resolution to obtain the true length of the fissure;

[0085] The calculation of the rock mass fissure strike is realized based on the extraction of the fissure skeleton line; by marking each point on the fissure surface, then calculating the tangent direction vector of the skeleton line, and then taking the average value of all these tangent direction vectors of the skeleton line, the average vector representing the overall tangent direction of the skeleton line can be obtained; using the average tangent direction vector, the dip angle is determined through the x and y components of the given average tangent direction vector; the mathematical expression is as follows:

[0086]

[0087]

[0088] In the formula: a is the vector; g 1 , g 2 , f 1 , f 2 are the coordinate points; n is all the tangent direction vectors.

[0089] After obtaining the direction of the average tangent line, the angle θ of the crack can be obtained through the inverse trigonometric function. The specific mathematical expression is as follows:

[0090]

[0091] In some modified embodiments of the first aspect of the present invention, step S10 includes the following steps:

[0092] S11. Use the weighted average method to grayscale the obtained color image of the rock mass cracks, and perform weighted average distribution on the values of R, G, and B. The expression is as follows:

[0093] R = G = B = W R R + W G G + W B B;

[0094] In the formula: R, G, and B are the three component values of the image, where W R , W G and W B are the weight values of R, G, and B. When W R = 0.3, W G = 0.59, W B = 0.11, the required grayscale image can be obtained;

[0095] S12. Obtain the attention of the image width and height through the CA attention mechanism, perform global pooling on the input image in the horizontal and vertical directions respectively, and then obtain the feature maps in the horizontal and vertical directions in sequence. The calculation formula is as follows:

[0096]

[0097]

[0098] In the formula: x c is the input feature vector; h is the image height; w is the image width.

[0099] S13. Concatenate the feature maps in the horizontal and vertical directions in the spatial direction, then perform 1×1 convolution operation for dimensionality reduction, and send the feature map after BN layer processing to the Sigmoid activation function to obtain the feature map f. The calculation formula is as follows:

[0100] f = σ(F 1 ([z h , z w ))

[0101] In the formula: F 1 is the result of the BN layer operation;

[0102] The feature map with the same number of channels as the original is obtained through a 1×1 convolutional kernel, and f is decomposed into the feature maps f h and f w , and finally, the attention weights in the horizontal and vertical directions are obtained through the Sigmoid activation function. The calculation formula is as follows:

[0103] g h =σ(F h (f h ))

[0104] g w =σ(F w (f w ));

[0105] In the formula: F h and F w are transformation functions for f h and f w to be transformed into the same dimension as the output;

[0106] S14. Through the multiplication weighting operation on the original feature map, the final feature map with attention weights in the horizontal and vertical directions is obtained. The calculation formula is as follows:

[0107] y c =x c ×g h ×g w ;

[0108] In the formula: x c is the dimension of the input feature map.

[0109] In some modified embodiments of the first aspect of the present invention, step S40 includes the following steps:

[0110] S41. Calculate the in-hole pressure in the blast hole; the expression is:

[0111]

[0112] In the formula: P is the average pressure acting on the blast hole; P e is the initial pressure; γ is the gas expansion index; l e is the length of the charge section; l a is the length of the air column;

[0113] S42. Calculate the charge amount, and the expression is:

[0114] Q = q×S×L;

[0115] In the formula: Q is the total charge amount for one blasting; q is the unit consumption of blasting explosive; S is the excavation cross-sectional area; L is the average depth of the blast hole;

[0116] S43. For the blast holes with discontinuous charging, the charge amount Q can be calculated according to the following formula:

[0117]

[0118] Where: r is the blast hole radius; q is the unit consumption of blasting explosive; l e is the charging section length; K d is the radial decoupling coefficient, K d = r b / r c , r b 、r c are the blast hole diameter and cartridge diameter;

[0119] S44. When calculating the coupled charge, the impact load acting on the rock mass is expressed as follows:

[0120]

[0121]

[0122] It can be obtained that:

[0123]

[0124] Where: P is the maximum pressure; P H is the detonation pressure of the explosive; D e is the detonation velocity of the explosive; ρ e 、ρ m are the explosive density and rock density; C p is the wave velocity of the rock mass;

[0125] When the radial air decoupled charge is used, the peak value of the blasting load pressure in the charging section is as follows:

[0126]

[0127] Where: P 0 is the peak value of the maximum pressure in the charging section; K d is the decoupling coefficient; n is the pressure increase multiple, n = 8; D e is the detonation velocity of the explosive; ρ e is the explosive density;

[0128] Since the time for the air expansion in the hole is very short, it can be considered that the pressure on the hole wall when the air fills the entire blast hole is the peak value of the air section pressure, and the formula is as follows:

[0129]

[0130] Where: P 1 is the peak value of the maximum pressure in the air section; γ is the gas expansion index; l eis the length of the charge section; l is the length of the blast hole;

[0131] S45. Calculate K t1 and the surface fracture rate K t2 , and the expression is

[0132]

[0133]

[0134] In the formula: A i , l i are the area and trace length of the i-th structural plane fracture respectively; V is the volume of the rock mass; A is the area of the cross-section of the rock mass; n is the number of structural planes;

[0135] S46. Calculate the rock mass quality index and the rock mass volume joint number J v , and the expression is as follows:

[0136]

[0137] In the formula: L i is the length of the i-th core with a length greater than or equal to 10 cm in the drilling; L is the total length of the drilling;

[0138] The joint direction influence coefficient fi is introduced, and the mathematical expression is as follows:

[0139]

[0140] In the formula: TL is the total length of the drilling; CW is the length of the scoured core section; Fr is the length of the core less than 5 cm; Cr is the length of the broken core section; K is the length of the core in the karst area;

[0141] The new index LRQD based on the RQD method considering the influence of the above indicators is as follows:

[0142]

[0143] In the formula: TL is the total length of the drilling; Cr is the length of the broken core section; Fr is the length of the core less than 5 cm; K t1 is the volume fracture rate; K t2 is the surface fracture rate; μ is the correction coefficient of the rock mass fracture rate; it is initially determined that when the rock mass fracture rate is greater than 25, the correction coefficient μ slows down the increasing speed of the rock mass fracture rate; when the rock mass fracture rate is less than 25, the correction coefficient μ is 1;

[0144] S47. Through the comparative analysis of the new index LRQD and RQD, the volume fracture rate K t1 and the surface fracture rate K t2, the fracture area Fr and the fragmentation area Cr are corrected and adjusted on the basis of the original RQD method, and the mathematical expression of the correction coefficient α is formulated as follows:

[0145]

[0146] According to the distribution of rock mass fissures, the anti-shock pressure is corrected to achieve an ideal effect. The formula is as follows:

[0147] P d = αP

[0148] In the formula: α is the correction coefficient; P is the pressure.

[0149] In some alternative embodiments of the first aspect of the present invention, in the borehole camera system, the connection steps are as follows:

[0150] S51. Connect the cable to the borehole camera, select appropriate accessories and install them on the borehole camera, and connect the cable and the power supply through the connecting wire;

[0151] S52. Connect the power supply to the industrial all-in-one computer to supply power to the borehole camera and the industrial all-in-one computer. Connect the connecting wire, the converter and the video capture card in sequence to convert the analog signal of the borehole camera into a digital signal that can be processed by the industrial all-in-one computer;

[0152] S53. Connect the converter and the video capture card to the industrial all-in-one computer through appropriate connecting wires. At this time, the borehole camera device connection is completed, and the designed program can be used on the industrial all-in-one computer for research activities; both the cable and the borehole camera are waterproof devices during use;

[0153] S54. After use, shut down the industrial all-in-one computer and disconnect all components. After simple cleaning, put them into the equipment box. Description of the Drawings

[0154] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments.

[0155] Figure 1 It is the overall flowchart of the intelligent matching design method for fracture identification and charging in the blast holes of underground engineering blasting;

[0156] Figure 2 It is the function and flowchart of the rock mass fracture detection system based on deep learning in the intelligent matching design method for fracture identification and charging in the blast holes of underground engineering blasting;

[0157] Figure 3 It is the schematic diagram of the rock mass fracture detection system based on deep learning in the intelligent matching design method for fracture identification and charging in the blast holes of underground engineering blasting;

[0158] Figure 4 It is the function and flow chart of the charge matching design in the method for identifying fractures in blast holes of underground engineering blasting and intelligent matching design of charge loading;

[0159] Figure 5 It is the system schematic diagram of the charge matching design in the method for identifying fractures in blast holes of underground engineering blasting and intelligent matching design of charge loading;

[0160] Figure 6 It is the equipment connection schematic diagram of the in-hole camera system in the method for identifying fractures in blast holes of underground engineering blasting and intelligent matching design of charge loading.

[0161] In the figure: 1-Industrial all-in-one computer; 2-In-hole camera; 3-Cable; 4-First connecting wire; 5-Power supply; 6-Second connecting wire; 7-Converter; 8-Video capture card. Specific implementation mode

[0162] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0163] To make the above objects, features, and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention.

[0164] Embodiment 1

[0165] The present invention relates to a method for identifying fractures in blast holes of underground engineering blasting and intelligent matching design of charge loading, including the development of a rock mass fracture detection system module based on deep learning, a charge intelligent matching design system module based on in-hole fracture identification, and in-hole camera equipment.

[0166] As Figure 1 shown, the method for identifying fractures in blast holes of underground engineering blasting and intelligent matching design of charge loading, its working process includes the following steps:

[0167] Step (1): Obtain rock mass fracture images, and make a data set used in deep learning. The establishment of the data set generally includes three parts: image acquisition, image preprocessing, and image annotation. Generally, grayscale processing is performed on the images, and the weighted average method is used. The specific expression is as follows:

[0168] R = G = B = W R R + W G G + W B B

[0169] where: R, G, and B are the three component values of the image, where W R , W G , and W B are the weight values of R, G, and B.

[0170] A deep learning network model is established. In this invention, the YOLOv5 network model is used. The backbone feature extraction network of YOLOv5 is Backbone. An important feature of Backbone is that it uses the Residual network. Its features are easy to optimize and can improve the accuracy by increasing the depth. Image annotation uses the professional semantic segmentation dataset annotation software Labelimg to annotate the dataset images. The crack parts in the images are marked with rectangular boxes, and then the categories are described. The CA attention mechanism is to obtain the attention in the width and height aspects of the image, encode more accurate data, perform global pooling on the input image in the horizontal and vertical directions respectively, and then obtain the feature maps in the horizontal and vertical directions successively. The calculation formula is as follows:

[0171]

[0172]

[0173] where: xc is the input feature vector.

[0174] Then is the attention generation operation. The feature maps in the horizontal and vertical directions are concatenated together in the spatial direction, and then a 1×1 convolution operation is performed for dimensionality reduction. The feature map after passing through the BN layer is sent into the Sigmoid activation function to obtain the feature map f. The calculation formula is as follows:

[0175] f = σ(F 1 ([z h , z w ))

[0176] where: F 1 is the result of the BN layer operation.

[0177] A feature map with the same number of channels as the original is obtained through a 1×1 convolution kernel. f is decomposed into feature maps f h and f w in the spatial dimension. Finally, the attention weights in the horizontal and vertical directions are obtained through the Sigmoid activation function. The calculation formula is as follows:

[0178] g h = σ(F h (f h ))

[0179] g w= σ(F w (f w ))

[0180] where: F h and F w are transformation functions obtained by transforming f h and f w into the same dimension as the output.

[0181] Finally, there is the feature map correction operation. Through multiplication and weighting operations on the original feature map, a feature map with attention weights in both the horizontal and vertical directions will be obtained. The calculation formula is as follows:

[0182] y c = x c × g h × g w

[0183] where: x c is the dimension of the input feature map.

[0184] To improve the feature extraction ability of YOLOv5, the CA attention mechanism is added to the model. To verify whether adding the CA attention mechanism can improve the recognition result, by comparing the recognition accuracy of rock mass fracture images, the test result data are specifically: Precision, Recall, Map, F1, etc.

[0185] Step (2): Use PyQt5 for interface design. The programming language is Python, the deep learning framework is Pytorch, the processor (CPU) is Intel(R) Core(TM) i5 - 10300H, the graphics processing unit (GPU) is NVIDIA GeForce GTX 1650; the operating system is Windows10. Qt Designer is the development interface of PyQt5. The interface is mainly divided into five functional areas. The first part is the "Widget Box"; the second part is the "Dialog"; the third part is the object viewer; the fourth part is the property editor; the fifth part is the signal editor. As Figure 2 shown, it is the main functional structure relationship diagram of the system. The main functions of the rock mass fracture detection system module based on deep learning are: selecting images, image detection, camera detection, taking pictures, and exiting the interface function. The packaging tool used is the Pyinstaller library. Using the Pyinstaller library to encapsulate the system, the "Rock Mass Fracture Detection System Based on Deep Learning" application program is obtained. As Figure 3 shown, it is the schematic diagram of the rock mass fracture detection system based on deep learning.

[0186] Step (III): On the basis of image processing and image recognition, based on digital image technology, combined with the results of rock mass fracture recognition, select an appropriate method for extracting rock mass fracture image features, segment the fractures from the background, and quantitatively express them. First, perform image segmentation on the rock mass fracture image. The Fuzzy C-Means (FCM) clustering algorithm is used. The Fuzzy C-Means clustering algorithm is also an algorithm based on the objective function. Suppose there is a rock mass image containing fractures with a total of n pixels, that is, a data set containing n pixels is given: x = {x1, x2,..., xn}. The objective function of the Fuzzy C-Means clustering algorithm is defined as follows:

[0187]

[0188]

[0189] where: u mn is the fuzzy membership degree of pixel point x n relative to the m-th clustering center z m , p is the fuzzy index, and d mn is the gray-scale distance between x n and the clustering center z m .

[0190] To find the minimum value of the objective function J, use the Lagrange multiplier method for the objective function under the condition of satisfying the constraint conditions. Let the partial derivatives of J with respect to v m and u mn be 0. Therefore, the membership degree matrix u mn , and the clustering center v m can be obtained. The specific formulas are as follows:

[0191]

[0192]

[0193] The specific steps of the Fuzzy C-Means (FCM) clustering algorithm are as follows:

[0194] Step 1: First, input the data of the rock mass fracture image into the algorithm, set the number of rock mass fracture categories for clustering, the fuzzy index p, and the initial clustering center, etc.;

[0195] Step 2: Set the threshold for stopping iteration and the maximum number of iterations;

[0196] Step 3: Then update the membership degree and calculate the clustering center;

[0197] Step 4: Determine whether the conditions of the normalization measure function are satisfied. If satisfied, stop the operation of the algorithm and output the membership degree and the clustering center;

[0198] Step Five: Divide the image pixel points according to the principle of maximum membership degree to achieve the image segmentation of rock mass fissures.

[0199] After segmentation by the fuzzy C-means (FCM) clustering algorithm, there are usually tiny noise points in the rock mass fissure image. Therefore, an image denoising operation is used to remove the tiny noise points. The image denoising algorithm used is the connected component denoising method based on the least absolute deviation method, which is a commonly used image processing method. When there is not much noise in the image and there is an obvious difference in the size of the connected component where the noise is located and the size of the connected component where the target is located, more accurate results can be obtained using the least absolute deviation method. The expression of the least absolute deviation method is:

[0200]

[0201] After image denoising, the extraction of the fissure skeleton line can be carried out. The Zhang-Suen thinning algorithm is adopted. The purpose of the thinning algorithm is to refine the rough features in the image into finer lines or curves for subsequent image analysis and recognition. This algorithm can effectively refine the features in the image while maintaining the topological structure and shape features of the image.

[0202] First Step: Scan pixel by pixel. By scanning the image pixel by pixel, find the candidate pixel points to be deleted. The judgment conditions include the black and white pattern around the pixel point and the connectivity of the pixel point, etc. For the foreground color, it will be detected whether it satisfies the following conditions:

[0203] The current pixel is white, at least one of the upper, lower, left, right, upper left, lower left, upper right, and lower right of the 8-neighborhood of the current pixel P is black, and the number of white pixels in the 8-neighborhood of the current pixel is between 2 and 6. If all the above conditions are satisfied, then mark the current pixel as black.

[0204] Second Step: Use the algorithm to traverse each pixel of the image again. For the foreground color, it will be checked whether it satisfies the following conditions:

[0205] The current pixel is white, at least one of the 8-neighborhood of the current pixel is black, the number of white pixels in the 8-neighborhood of the current pixel is between 2 and 6, and the black pixels in the 8-neighborhood of the current pixel are not connected; that is, they are not a connected component; if all the above conditions are satisfied, then mark the current pixel as black; after repeated iterations, based on the first step, continuously iterate and apply the rules until no pixel points can be deleted. After the iteration ends, the image after thinning processing is obtained.

[0206] After the refinement process of the rock mass fracture image, fractures may occur. Therefore, it is necessary to connect the disconnected fractures to make the fractures continuous and complete. The SURF feature point detection and matching method is used for fracture splicing, including feature point detection, descriptor extraction, feature point matching, geometric transformation, and image fusion. The specific steps are as follows:

[0207] The SURF algorithm determines the key point positions by detecting local extreme points in the image. Let (x, y) be a pixel point in the image, then the image of this point can be defined as the sum of the values of all pixel points in the rectangular area enclosed by the image origin and this point. The mathematical expression is as follows:

[0208]

[0209] The scale space theory is the method for detecting feature points in the SURF algorithm. It regards the determinant in the Hessian matrix as a discriminant to find local maxima. Constructing the Hessian matrix is to generate stable mutation points in the image. The Hessian matrix can be expressed as:

[0210]

[0211] In the formula: σ is the Gaussian scale; (x, y) is the position of the image pixel point; L xx (x, y, σ) represents the convolution of the second-order Gaussian differential at the pixel point (x, y) of the image I, and the same applies to other Ls.

[0212] After determining the key points, the SURF algorithm generates descriptors for each key point. These descriptors can describe the gradient information and distribution of the area around the key points. Next, the key points in the image are matched. Usually, a similarity measure based on descriptors is used to find the corresponding key points in the image. Through the matched key points, different transformation models can be used to estimate the geometric transformation relationship between the images. Finally, according to the estimated geometric transformation relationship, the two images are spliced and fused into a complete image by superposition or mixing. The weighted average fusion algorithm is adopted, and the weighting factor is u. The mathematical expression for the weighted fusion of the pixel values of the image is as follows:

[0213]

[0214] The edge extraction of the rock mass fracture is carried out by the Canny algorithm. In the Canny algorithm, a Gaussian filter is used. It is a linear filter obtained by sampling and normalizing the two-dimensional Gaussian function. The mathematical expression of the two-dimensional Gaussian function used by the Gaussian filter in the Canny algorithm is specifically as follows:

[0215]

[0216] S = G × I

[0217] Where: G represents the Gaussian filter obtained by operating on the two-dimensional Gaussian function; I represents the image to be detected; S represents the smoothed image.

[0218] For the smoothed image, calculate the magnitude and direction of the gradient of each pixel point. The components of the gradient of the pixel points on the edge in the vertical and horizontal directions are as follows:

[0219]

[0220] The magnitude g of the pixel point gradient θ and the direction θ are respectively:

[0221]

[0222] According to the given high and low thresholds, divide the strong and weak edges, and eliminate the isolated weak edges to obtain the final detection result. In the Canny algorithm, two thresholds T 1 and T 2 are set. The points with gradient values larger than T 1 are strong edge points, and vice versa for weak edge points.

[0223] The calculation of the width of the rock mass fissure is realized based on fissure skeleton extraction and edge detection, and the orthogonal skeleton line method is used to calculate the fissure width. The orthogonal skeleton line method in this paper uses the method of finding the K nearest neighbor points of a given point, and specifically, the KD-tree algorithm can be used. The KD-tree algorithm is a tree-shaped data structure for storing points in a K-dimensional space for rapid retrieval. The process of constructing a KD-tree is to continuously divide the K-dimensional space with hyperplanes perpendicular to the coordinate axes to form a series of K-dimensional hyper-rectangular regions. Among them, the method for calculating the distance between data points in the KD-tree algorithm is the Euclidean distance, and its mathematical expression is as follows:

[0224]

[0225] Where: K is the dimension; d is the distance between data points, a i and y i are data points.

[0226] The general steps for using the KD-tree algorithm to find the k nearest neighbor points of a given point are:

[0227] First, it is necessary to construct a KD - tree for the dataset. The process of constructing a KD - tree involves selecting an axis, and then dividing the dataset into two subsets according to the axis. The left subtree contains data points smaller than the axis value, and the right subtree contains data points larger than the axis value. Then, the subsets are recursively divided until each subset contains only one data point. Starting from the root node, search recursively down the tree. Compare the coordinate values of the given point with the axis value of the current node's division to determine whether to search in the left subtree or the right subtree. During the search process, it is necessary to keep track of the k nearest neighbor points. Once the k nearest neighbor points of the target point are found, it is necessary to backtrack to the parent node to check if there are closer points. If so, update the set of nearest neighbor points. When backtracking to the root node, the k nearest neighbor points of the given point can be obtained.

[0228] After the KD - tree algorithm, the normal vector of the skeleton line is calculated by singular value decomposition (SVD). Then, find the two crack edge points closest to the normal vector of the skeleton line, and calculate the intersection point of the line segment formed by these two points and the normal vector of the skeleton line. The definition of singular value decomposition (SVD) is as follows:

[0229] Let there be a bilinear function: f(x,y) = x T Ay

[0230] where x, y belong to R n×1 and A belongs to R n×n , by introducing a linear transformation, the above formula can be alternatively expressed as: f(x,y)=(Uξ) T A(Vη)=ξ T U T AVη

[0231] Next, let: S = U T AV, then the formula can be written as: f(x,y)=ξ T Sη

[0232] If both U and V are orthogonal matrices, S can be made a diagonal matrix through appropriate U and V, which can be specifically expressed as: S = diag(σ 1 ,σ 2 ,...,σ n )

[0233] Multiply both sides by U and VT, it can be expressed as: A = USV T

[0234] As described above is the definition of singular value decomposition (SVD) given by Beltrami.

[0235] First, convert the geometric representation of the skeleton line into matrix form. For example, represent the endpoint coordinates of the skeleton line as a matrix. Each row represents the coordinates of an endpoint, and each column represents a different dimension. Perform a centering operation on the matrix representing the skeleton line by subtracting the mean value of each dimension from the coordinates of that dimension, so that the center of the data is located at the origin. Then perform singular value decomposition on the centered matrix. The normal vector can usually be obtained from the result of singular value decomposition (SVD). The right singular vector corresponding to the smallest singular value can be used as the normal vector of the skeleton line representing the main direction.

[0236] After skeletonizing the fissures and splicing the fissures, the length of the fissure skeleton line is calculated as the fissure width. The main methods for calculating the crack length include the neighborhood scanning method based on the skeleton curve and the approximate calculation method based on the start and end points. In this paper, the neighborhood scanning method based on the skeleton curve is used. First, the fissures are skeletonized, and then the burr parts on the fissure skeleton line are removed by morphological operations; scan the global image to find the endpoints of the crack, and search for fissure pixels and count the number according to the 8-neighborhood method starting from the endpoints; multiply the obtained number of fissure pixels by the resolution to obtain the true length of the fissure. Using this method, the results are relatively accurate, but the requirements for the processing process are relatively high. In the morphological processing process, the effect of removing the burrs on the fissure skeleton line directly affects the calculation results.

[0237] The calculation of the strike of rock mass fissures is realized based on the extraction of the fissure skeleton line. By marking each point on the fissure surface, then calculating the tangent direction vector of its skeleton line, and then taking the average value of all these tangent direction vectors of the skeleton line, the average vector representing the overall tangent direction of the skeleton line can be obtained. Using the average tangent direction vector, the dip angle of the fissure can be calculated using the inverse trigonometric function. Specifically, the arctan function (or other equivalent functions) can be used to determine the dip angle through the x and y components of the given average tangent direction vector. This dip angle is usually the azimuth angle relative to the horizontal plane, and the specific mathematical expression is as follows:

[0238]

[0239]

[0240] After solving the direction of the average tangent, the angle θ of the fissure can be obtained through the inverse trigonometric function. The specific mathematical expression is as follows:

[0241]

[0242] The calculation of the rock mass fracture area is based on the effect diagram after image denoising operation and combines with the fracture edge detection algorithm. First, the boundary of the fracture is identified, and the pixel area of the fracture is calculated using the image processing algorithm. Finally, the true area of the fracture is calculated through the proportional relationship between the pixel area and the actual size. This method of calculating the fracture area based on the pixel area is applicable to various irregularly shaped fractures and can be realized through digital image processing technology.

[0243] Step (4): The development of the rock mass structural plane is an important factor in judging the quality of engineering rock mass. The fracture distribution in the rock mass is one of the important factors affecting the blasting effect. Therefore, taking the rock mass fractures as the research object, analyzing the rock mass fractures in the blasting holes. After obtaining the fracture information in the holes, the charge structure can be designed according to the fracture distribution and geometric information. In the blasting process of air interval charging, the pressure in the hole can be expressed by the following formula:

[0244]

[0245] In the formula: P is the average pressure acting on the blast hole; Pe is the initial pressure; γ is the gas expansion index; le is the charge section length; la is the air column length.

[0246] The calculation of the blast hole charge has a greater impact on the blasting effect and blasting cost. In the case of continuous charging, the mathematical expression for calculating the charge is generally calculated according to the following empirical formula:

[0247] Q = q × S × L

[0248] In the formula: Q is the total charge for one blasting; q is the unit consumption of blasting explosive; S is the excavation cross-sectional area; L is the average depth of the blast hole.

[0249] For the blast holes with discontinuous charging, the charge Q can be calculated according to the following formula:

[0250]

[0251] In the formula: r is the blast hole radius; q is the unit consumption of blasting explosive; l e is the charge section length; K d is the radial decoupling coefficient, K d = r b / r c , r b 、r c is the blast hole diameter and cartridge diameter. The unit consumption q of the explosive varies according to the different rock properties.

[0252] Blast hole coupled charging means that the radial decoupling coefficient K d = 1, K d = r b / r c , r b 、r c are the hole diameter and the charge diameter of the blast hole. When the charge is coupled, the impact load acting on the rock mass is as follows:

[0253]

[0254]

[0255] From the above two equations, we can get:

[0256]

[0257] In the formula: P is the maximum pressure; P H is the detonation pressure of the explosive; D e is the detonation velocity of the explosive; ρ e , ρ m are the density of the explosive and the density of the rock; Cp is the wave velocity of the rock mass.

[0258] When the charge is uncoupled with air radially, the peak value of the blasting load pressure in the charged section is as follows:

[0259]

[0260] In the formula: P 0 is the peak value of the maximum pressure in the charged section; K d is the decoupling coefficient; n is the pressure amplification factor, n = 8; D e is the detonation velocity of the explosive; ρ e is the density of the explosive.

[0261] Since the time for the air expansion in the hole is very short, it can be considered that the pressure on the hole wall when the air fills the whole blast hole is the peak value of the air section pressure, and the formula is as follows:

[0262]

[0263] In the formula: P 1 is the peak value of the maximum pressure in the air section; γ is the gas expansion index; l e is the length of the charged section; l is the depth of the blast hole.

[0264] The fracture rate of the rock mass can be divided into the volume fracture rate Kt1 and the surface fracture rate Kt2. The volume fracture rate Kt1 is the total fracture area contained in the rock mass per unit volume, and the surface fracture rate Kt2 is the total fracture length contained in the rock mass per unit area. The fracture rate of the rock mass comprehensively reflects the development of the structural plane of the rock mass, and the specific mathematical expression is as follows:

[0265]

[0266]

[0267] Where: A i , l i are respectively the area and trace length of the fracture of the i-th structural plane; V is the volume of the rock mass; A is the area of the cross-section of the rock mass; n is the number of structural planes.

[0268] The integrity of the rock mass is an important index for evaluating the quality of the rock mass. The rock mass quality index RQD is a commonly used method to represent the integrity of the rock mass. The definition of RQD is the ratio of the cumulative length of the fractured core blocks with a length greater than or equal to 10 cm during drilling to the total length of the drilling footage. The specific mathematical expression is as follows:

[0269]

[0270] Where: L i is the length of the i-th core with a length greater than or equal to 10 cm during drilling; L is the total length of the drilling.

[0271] The new index RQD proposed by Azimian et al. in 2016 I considers different types of rock masses, including the scoured area CW, the fractured area Fr, the crushed area Cr, and the karst area K. The joint direction influence coefficient fi is introduced into the index. The mathematical expression is as follows:

[0272]

[0273] Where: TL is the total length of the drilling; CW is the length of the scoured core section; Fr is the length of the core with a length less than 5 cm; Cr is the length of the core in the crushed section; K is the length of the core in the karst area.

[0274] On the basis of RQD, considering the influence of fractures on the evaluation of rock mass quality, the volumetric fracture ratio K t1 of the rock mass fractures and the surface fracture ratio K t2 of the rock mass fractures are selected as two indexes. At the same time, referring to the new index RQD I , the fractured area Fr and the crushed area Cr are retained. Similar data can be obtained by observing the fracture distribution on the borehole wall. Therefore, the new index LRQD based on the RQD method considering the influence of the above indexes is as follows:

[0275]

[0276] Where: TL is the total length of the drilling; Cr is the length of the core in the crushed section; Fr is the length of the core with a length less than 5 cm; K t1 is the volumetric fracture ratio; K t2 is the surface fracture ratio; μ is the correction coefficient of the rock mass fracture ratio; the preliminary determination of the correction coefficient is: when the rock mass fracture ratio is greater than 25, the correction coefficient μ is used to slow down the increasing speed of the rock mass fracture ratio; when the rock mass fracture ratio is less than 25, the correction coefficient μ is 1.

[0277] Through the comparative analysis of the new index LRQD and RQD, referring to the volume fracture rate K t1 and the surface fracture rate K t2 , the fracture area Fr and the broken area Cr, the original RQD method was modified and adjusted, and the mathematical expression of the correction coefficient α was formulated as follows:

[0278]

[0279] The correction coefficient α is mainly used to correct the peak pressure P that the blast hole can actually withstand. By calculating the rock mass fracture rate and combining with on-site data for analysis, determine whether to continuously charge and the ratio of the length of the air column in the blast hole to the length of the charge, so as to realize the charge structure design. According to the grade type of the rock mass, calculate the impact resistance pressure of different grade rock masses, and then correct the impact resistance pressure according to the fracture distribution of the rock mass to achieve the ideal effect. The formula is as follows:

[0280] P d = αP

[0281] In the formula: α is the correction coefficient; P is the pressure.

[0282] Step (4): Use Matlab software to carry out an automatic matching design system for charging based on the identification of fractures in the blast hole. Matlab provides a powerful GUI development toolbox, which can quickly create a custom graphical interface. Users do not need to write code tediously. They can build the GUI only through simple drag-and-drop and configuration operations, such as Figure 4 shown, which is the functional structure diagram of the system. As Figure 5 shown, which is the schematic diagram of the system interface. In Matlab software, packaging the executable program can be completed only by using the built-in functions.

[0283] Step (5): Establish an in-hole imaging device. The hardware part of the in-hole imaging device consists of an in-hole camera, in-hole camera accessories, an industrial all-in-one computer, cables, a power supply and charger, a power connection wire, a converter and an adapter wire, a video capture card, and an equipment box; the software part of the in-hole imaging device consists of a rock mass fracture detection system based on deep learning and an intelligent matching design system for charging based on the identification of fractures in the blast hole. As Figure 6As shown, the connection relationships of the devices are represented by red lines and numbered. The parts within the red dashed boxes are, from top to bottom, the partial view of the in-hole camera, the accessories of the in-hole camera, and the equipment box. The specific connection relationships are as follows: First, connect the cable 3 to the in-hole camera 2. Select a suitable accessory and install it on the in-hole camera 2. Connect the cable 3 and the power supply 5 through the first connecting line 4. Then connect the power supply 5 and the industrial all-in-one computer 1 to supply power to the in-hole camera 2 and the industrial all-in-one computer 1. Connect the second connecting line 6, the converter 7, and the video capture card 8 in sequence to convert the analog signal of the in-hole camera 2 into a digital signal that can be processed by the industrial all-in-one computer 1. Finally, connect the converter 7 and the video capture card 8 to the industrial all-in-one computer 1 through suitable connecting lines. At this time, the connection of the in-hole camera equipment is completed, and the designed program can be used for research activities on the industrial all-in-one computer 1. When in use, both the cable 3 and the in-hole camera 2 are waterproof devices, while the waterproof ability of other parts is relatively weak. Therefore, special attention needs to be paid. After use, turn off the industrial all-in-one computer 1 and disconnect all components. After simple cleaning, put them into the equipment box.

[0284] The above are only the specific implementation manners of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should all be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. A method for identifying cracks in underground engineering blasting holes and intelligent matching of charges, characterized in that: The method comprises the following steps: S10. Obtain rock mass crack images and preprocess the rock mass crack images; establish a data set in a deep learning network model; S20. Build a deep learning network model; use the improved network model to perform image recognition on the processed image; S30. According to the rock mass fracture identification result of step S20, the fracture is segmented from the background and quantitatively expressed by extracting the identified fracture feature area, extracting and splicing the fracture skeleton and obtaining fracture set information; S40. According to the recognition result of step S30, the charge structure in the blasthole is analyzed, and an intelligent charge matching design system based on the recognition of cracks in the blasthole is established; Step S40 includes the following steps: S41. Calculate the borehole pressure; the expression is: Where: P a is the average pressure acting on the blast hole; P e is the initial pressure; γ is the gas expansion index; l e is the length of the charging section; l a is the length of the air column; S42. For the blastholes with continuous charge, calculate the charge amount. The expression is: Q = q × S × L; Where: Q is the total charge for one blast; q is the unit consumption of blasting explosives; S is the excavation cross-sectional area; L is the average depth of the blasthole; S43. For blastholes with discontinuous charge, the charge quantity Q can be calculated according to the following formula: Where: r is the radius of the blast hole; q is the unit consumption of blasting explosives; l e is the length of the charging section; K d is the radial uncoupling coefficient, K d =r b / r c , r b 、r c is the diameter of the blast hole and the diameter of the charge roll; S44. When calculating the coupled charge, the impact load acting on the rock mass is expressed as follows: We can get: Where: P b is the maximum pressure; P H is the detonation pressure of explosive; D e is the detonation velocity of explosive; e , m is the density of explosive and rock; C p is the rock mass wave velocity; When radial air is not coupled with charge, the peak value of blasting load pressure of charge section is as follows: Where: P0 is the maximum pressure peak of the charging section; K d is the uncoupling coefficient; n is the pressure increase factor, n = 8; D e is the detonation velocity of explosive; e is the density of explosive; Since the time of air expansion in the hole is very short, it can be determined that the pressure on the hole wall after the air fills the entire blasthole is the peak pressure of the air section. The formula is as follows: Where: P1 is the maximum pressure peak of the air segment; γ is the gas expansion index; l e is the length of the charging section; l is the depth of the blasthole; S45. Calculate K t1 The surface crack ratio K t2 , the expression is: Where: A i , l i are the area and trace length of the crack of the i-th structural plane respectively; V is the volume of the rock mass; A is the area of ​​the rock mass section; n is the number of structural planes; S46. Calculation of rock mass quality index and rock mass volume joint number J v , the expression is as follows: Where: RQD is the rock mass quality index, L i is the length of the ith core greater than or equal to 10 cm in the drilling; L is the total length of the drilling; The joint direction influence coefficient f1 is introduced, and the mathematical expression is as follows: Where: RQD I To introduce the rock mass quality index of the joint direction influence coefficient, TL is the total length of drilling; CW is the length of the scouring core section; Fr is the length of the core less than 5 cm; Cr is the length of the broken section core; K is the length of the karst area core; Considering the influence of the above indicators, the new indicator LRQD based on the RQD method is as follows: Where: TL is the total length of drilling; Cr is the length of the core in the crushing section; Fr is the length of the core less than 5 cm; K t1 is the body crack rate; K t2 is the surface crack ratio; μ is the correction coefficient of the rock mass crack ratio; it is preliminarily determined that when the rock mass crack ratio is greater than 25, the correction coefficient μ is used to slow down the increase rate of the rock mass crack ratio; when the rock mass crack ratio is less than 25, the correction coefficient μ is 1; S47. By comparing the new index LRQD and RQD, the body crack rate K t1 , surface crack rate K t2 , fracture area Fr and crushing area Cr, the original RQD method was modified and adjusted, and the mathematical expression of the correction coefficient α was proposed as follows: According to the distribution of rock mass cracks, the impact pressure is corrected to achieve the ideal effect. The formula is as follows: P d =αP 1 Where: α is the correction coefficient; P 1 is the pressure, which is the maximum value obtained by the pressure calculation method under different charging conditions; S50. Establish an in-hole camera system.

2. The method for designing crack identification and intelligent charge matching in underground engineering blasting holes according to claim 1 is characterized in that: Step S30 includes the following steps: S31. Using fuzzy C-means clustering algorithm to perform image segmentation operation on rock mass crack image; S32. Using the connected domain of the least squares method to remove tiny noise points from the image; S33. Using Zhang-Suen thinning algorithm to extract crack skeleton lines and thin the crack image; S34. Use the SURF algorithm to detect and match feature points to stitch the cracks, including feature point detection, descriptor extraction, feature point matching, geometric transformation and image fusion; use the weighted average fusion algorithm to stitch and fuse the two images into a complete image by superposition or mixing; S35. Use Canny algorithm to extract edge features of rock crack images; The crack width is calculated using the orthogonal skeleton algorithm; The crack length is calculated using the neighborhood scanning method based on the skeleton curve; The steps to calculate the crack area are: S351. calibrate the image according to the scale mark or other known size information in the image; S352. Use edge detection algorithm to extract crack contours and mark crack areas; S353. Use the pixel counting function in the image processing software or programming language to calculate the number of pixels inside the crack region; that is, the pixel area of ​​the crack region; S354. Multiply the pixel area by the actual area of ​​each pixel to obtain the actual area of ​​the crack.

3. The method for designing crack identification and intelligent charge matching in underground engineering blasting holes according to claim 2 is characterized in that: Step S31 includes the following steps: S311. First, input the rock mass crack image data into the algorithm, set the number of clustered rock mass crack categories, fuzzy index p and initialize the cluster center; S312. Set the threshold for stopping iteration and the maximum number of iterations; S313. Then update the membership degree and calculate the cluster center; S314. Determine whether the conditions of the standardized measurement function are met. If so, stop the algorithm and output the membership degree and cluster center; S315. Divide the image pixels according to the maximum membership principle to achieve image segmentation of rock mass fissures; Its expression is: Given a data set containing n pixels: x = {x1, x2, ..., xn }, the objective function of the fuzzy C-means (FCM) clustering algorithm is defined as follows: Where: u mn is the pixel xn relative to the mth cluster v m The fuzzy membership degree, p is the fuzzy index, d mn is the nth pixel X n and the mth cluster center v m The grayscale distance between Find the objective function f FCM The minimum value of the objective function is obtained by using the Lagrange multiplier method while satisfying the constraints. Let f FCM v m and u mn The partial derivative of is 0, so we can get the membership matrix u mn , cluster center v m , the specific expression is as follows:

4. The method for designing crack identification and intelligent charge matching in underground engineering blasting holes according to claim 2 is characterized in that: Step 32 includes the following steps: S321. Find all connected domains in the rock mass fracture image and obtain the number of pixels contained in each of their connected domains; S322. Determine the threshold range of the algorithm by using the least squares method; S323. Determine the threshold range of the algorithm according to the middle value of the threshold; S324. Processing the connected domain of the rock mass fracture image according to the threshold range; The expression of the least squares method is: Where: a i is the coordinate information of the pixel point; n is the number of pixels in the image; Step S33 includes the following steps: S331: Scanning pixel by pixel, by scanning the image pixel by pixel, to find candidate pixels to be deleted; judging conditions include the black and white mode around the pixel and the connectivity of the pixel, and for the foreground color, checking whether the following conditions are met: The current pixel is white, there is at least one black pixel in the 8-neighborhood of the current pixel P, and the number of white pixels in the 8-neighborhood of the current pixel is between 2 and 6. If all the above conditions are met, then the current pixel is marked as black; S332: The algorithm is used to traverse each pixel of the image again, and for the foreground color, it is checked whether it meets the following conditions: The current pixel is white, and in the 8-neighborhood of the current pixel, at least one is black, and in the 8-neighborhood of the current pixel, the number of white pixels is between 2 and 6, and the black pixels in the 8-neighborhood of the current pixel are not connected, that is, it is not a connected component; if all the above conditions are met, then the current pixel is marked as black; after repeated iterations, based on the first step, the rules are continuously applied iteratively until no pixels can be deleted; after the iteration, the image after refinement is obtained.

5. The method for identifying cracks in underground engineering blasting holes and intelligent matching design of explosives according to claim 2 is characterized in that: Step S34 includes the following steps: S341. Determine the key point position by detecting the local extreme point in the image. If (x, y) is a pixel point in the image, the image of the point is defined as the sum of the values ​​of all pixels in the rectangular area surrounded by the image origin and the point. The mathematical expression is as follows: S342. Construct the Hessian matrix to generate stable mutation points in the image. The Hessian matrix can be expressed as: Where: σ is the Gaussian scale; (x, y) is the position of the image pixel; L xx (x, y, σ) represents the convolution of the second-order Gaussian derivative at the pixel (x, y) of image I, and the same is true for the other L; S343. Generate a descriptor for each key point to describe the gradient information and distribution of the area around the key point; match the key points in the image and use a similarity metric based on the descriptor to find the corresponding key points in the image; Through the matched key points, different transformation models can be used to estimate the geometric transformation relationship between images; S344. According to the estimated geometric transformation relationship, the two images are spliced ​​and fused into a complete image by superposition or mixing. A weighted average fusion algorithm is adopted, and the weighting factor is u. The mathematical expression of the weighted fusion of the pixel values ​​of the images is as follows: Where: u is the weighting factor; x and y are coordinates; f1 and f2 represent images.

6. The method for designing crack identification and intelligent charge matching in underground engineering blasting holes according to claim 2 is characterized in that: The mathematical expression of the two-dimensional Gaussian function used by the Gaussian filter in the Canny algorithm is as follows: S=G×I Where: G represents the Gaussian filter obtained by operating the two-dimensional Gaussian function; I represents the image to be detected; S represents the image after smoothing; The magnitude and direction of the gradient of each pixel in the smoothed image are calculated. The components of the gradient of the pixel on the edge in the vertical and horizontal directions (x, y) are as follows: Pixel gradient size g θ and direction θ are: According to the given high and low thresholds, the strong and weak edges are divided, and the isolated weak edges are eliminated to obtain the final detection result; two high and low thresholds T1 and T2 are set; the point with a gradient value larger than T1 is a strong edge point, and vice versa is a weak edge point; The orthogonal skeleton method uses the KD tree algorithm, and the expression is as follows: Where: K is the dimension; d is the distance between data points, a i and i is a data point; After using the KD tree algorithm to find the given k nearest neighbor points, the normal vector of the skeleton line is calculated by singular value decomposition, and then the two crack edge line points closest to the skeleton line normal vector are found, and the intersection of the line segment formed by these two points and the skeleton line normal vector is calculated; the expression of singular value decomposition is as follows: Suppose the bilinear function is: f(x,y)=x T Oh Where x, y belong to R n×1 , A belongs to R n×n , by introducing linear changes, the bilinear function can be expressed as: f(x,y)=(Uξ) T A(Vη)=ξ T U T AVh Next command: S=U T OF Then the bilinear function can be written as: f(x,y)=ξ T S If U and V are both orthogonal matrices, S can be made into a diagonal matrix by using appropriate U and V, which can be specifically expressed as: S=diag(σ1,σ2,...,σ n ) Multiplying the left and right sides by U and VT can be expressed as: A=USV T ; The crack length is calculated using the neighborhood scanning method based on the skeleton curve; the steps are as follows: S351. First, the crack is skeletonized, and then the burr part on the crack skeleton line is processed by morphological operation; S352. Scan the global image to find the end point of the crack, and use the end point as the starting point to search for crack pixels according to the 8-neighborhood method and count the number; S353. Multiplying the obtained number of crack pixels by the resolution to obtain the actual length of the crack; The calculation of the rock mass fracture direction is based on the extraction of the fracture skeleton line. By marking each point on the fracture surface, calculating the skeleton tangent direction vector, and then taking the average of all these skeleton tangent direction vectors, the average vector representing the overall skeleton tangent direction can be obtained. The average tangent direction vector is used to determine the inclination angle through the given x and y components of the average tangent direction vector. The mathematical expression is as follows: Where: g1, g2, f1, f2 are coordinate points; After solving the direction of the average tangent, the angle θ of the crack can be obtained through inverse trigonometric functions. The specific mathematical expression is as follows:

7. The method for identifying cracks in underground engineering blasting holes and intelligent matching design of explosives according to claim 1 is characterized in that: Step S10 includes the following steps: S11. Use the weighted average method to grayscale the obtained rock mass crack color image, and perform weighted average distribution on the values ​​of R, G and B. The expression is as follows: R=G=B=W R R+W G G+W B B; Where: R, G and B are the three component values ​​of the image, and W R , W G and W B is the weight value of R, G and B. When W R =0.3,W G =0.59,W B =0.11, the required grayscale image can be obtained; S12. The CA attention mechanism is used to obtain the attention of the image width and height. The input image is globally pooled in the horizontal and vertical directions respectively, and then the feature maps in the horizontal and vertical directions are obtained in turn. The calculation formula is as follows: Where: x c is the input feature vector; h is the image height; w is the image width; S13. The horizontal and vertical feature maps are spliced ​​together in the spatial direction, and then a 1×1 convolution operation is performed to reduce the dimension. The feature map after the BN layer processing is sent to the Sigmoid activation function to obtain the feature map f. The calculation formula is as follows: f=σ(F1([z h ,z w ])) Where: F1 is the result of BN layer operation; After a 1×1 convolution kernel, we get a feature map with the same number of channels as the original one, and decompose f into feature maps f in the spatial dimension. h and f w , and finally the attention weights in the horizontal and vertical directions are obtained through the Sigmoid activation function. The calculation formula is as follows: g h =σ(F h (f h )) g w =σ(F w (f w )); Where: F h and F w f h and f w Transform into a transformation function of the same dimension as the output; S14. Through the multiplication weighted operation on the original feature map, the final feature map with attention weights in both the horizontal and vertical directions will be obtained. The calculation formula is as follows: and c =x c ×g h ×g w ; Where: x c is the input feature map dimension.

8. The method for identifying cracks in underground engineering blasting holes and intelligent matching design of explosives according to claim 1 is characterized in that: In the hole camera system, the connection steps are: S51. Connect the cable (3) to the camera in the hole (2), select suitable accessories to install the camera in the hole (2), and connect the cable (3) and the power supply (5) through the first connecting line (4); S52. Connect the power supply (5) and the industrial integrated machine (1) to power the hole camera (2) and the industrial integrated machine (1), and connect the second connecting line (6), the converter (7) and the video acquisition card (8) in sequence to convert the analog signal of the hole camera (2) into a digital signal that can be processed by the industrial integrated machine (1); S53. Connect the converter (7) and the video capture card (8) to the industrial integrated machine (1) through a suitable connecting cable. At this time, the hole camera device is connected and the designed program can be used on the industrial integrated machine (1) for research activities; when in use, the cable (3) and the hole camera (2) are both waterproof devices; S54. After use, turn off the industrial integrated machine (1) and disconnect all components, clean it briefly and put it in the equipment box.

Images