Rock mass fracture quantitative identification method based on image processing technology

Abstract

The invention provides a rock mass fracture quantitative identification method based on an image processing technology, and relates to the technical field of rock mass fracture identification, and the specific steps are as follows: obtaining a rock mass surface image and carrying out graying processing, identifying a fracture part through threshold segmentation, and carrying out connected domain marking to extract geometric parameters of a fracture; meanwhile, the load and mechanical property parameters of the rock mass are collected, the maximum value of the fracture length, the maximum value of the fracture width and the maximum value of the fracture depth are screened out, a fracture evaluation model is constructed, calculation and correlation analysis of a fracture grade coefficient, a rock mass rock burst coefficient and a damage coefficient are conducted, a comprehensive evaluation coefficient is generated, and the fracture risk grade of the rock mass is evaluated according to the comprehensive evaluation coefficient. The method not only considers the number of fractures, but also combines the physical and mechanical properties of the rock mass, can better perform quantitative evaluation of the rock mass fractures, improves the accuracy of rock mass stability evaluation through depth analysis of key parameters, and can more effectively identify potential risks.

Description

Technical Field

[0001] The invention relates to the technical field of rock mass crack identification, and in particular to a rock mass crack quantitative identification method based on image processing technology. Background Art

[0002] The quantitative identification method of rock mass cracks based on image processing technology is a technical means to efficiently and accurately identify and quantitatively analyze cracks on the rock surface using modern computer vision and image analysis technology. This method can evaluate the stability of rock mass in multiple application scenarios, such as mines, tunnels, dams and other geotechnical engineering projects.

[0003] In the prior art, the method for determining the number of cracks in a fractured rock mass based on machine learning, the method for evaluating the quality of a fractured rock mass, an electronic device and a storage medium provided by publication number CN117152060A include the following steps: collecting thermal infrared images of the rock mass surface in the test area of ​​the in-situ fractured rock mass and the number of cracks in the shallow surface layer of the rock mass; processing the temperature sequence matrix, and constructing a thermal infrared image data set in combination with the collected number of cracks data; extracting crack features from the thermal infrared image data set; performing feature recognition on the convolution feature vector through the full convolution layer to obtain a crack number prediction probability model; training the deep learning model until convergence; obtaining the temperature sequence matrix corresponding to the thermal infrared image of the rock mass surface in the target area of ​​the in-situ fractured rock mass, importing the prediction probability model that meets the requirements, and determining the number of cracks in the shallow surface layer of the rock mass in the target area. The determination method provided by the method can quickly and efficiently determine the number of rock mass cracks.

[0004] However, there are still the following deficiencies. As can be seen from the above statements, the existing technology mainly focuses on identifying the number of cracks through thermal infrared images, and lacks in-depth analysis of the geometric characteristics of the cracks (such as length, width, and depth). This leads to the lack of important spatial and geometric parameters when evaluating the impact of cracks on rock stability, and is unable to fully reflect the actual impact of cracks; when evaluating rock stability, it mainly relies on the prediction of the number of cracks, without considering the mechanical properties and load parameters of the rock. This single evaluation method cannot fully reflect the stability of the rock under actual working conditions.

[0005] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0006] The purpose of the present invention is to provide a method for quantitatively identifying rock mass fractures based on image processing technology to solve the problems raised in the above background technology.

[0007] To achieve the above object, the present invention provides the following technical solutions: A method for quantitatively identifying rock mass fractures based on image processing technology, the specific steps include: S1. Obtain the surface image of the rock mass at the current moment, grayscale the surface image of the rock mass, perform threshold segmentation on the grayscaled surface image of the rock mass, identify the crack part and the background part, mark the connected domain of the crack part, identify all crack areas, extract features of the crack areas, obtain the geometric parameters of the cracks, and collect the load parameters and mechanical property parameters of the rock mass at the current moment. The geometric parameters of the cracks include the length, width and depth of the cracks, the load parameters include the stress concentration factor, the load frequency, the load amplitude and the stress intensity ratio, and the mechanical property parameters include the density, porosity and compressive strength of the rock mass; S2. based on the crack lengths, crack widths and crack depths of all crack regions, screening the maximum crack length, maximum crack width and maximum crack depth in all crack regions; S3. Construct a crack assessment model, take the maximum crack length, maximum crack width and maximum crack depth at the historical moment as input, and output the crack grade coefficient at the historical moment as a label, determine the crack grade coefficient through the expert group scoring method, train the crack assessment model, input the maximum crack length, maximum crack width and maximum crack depth at the current moment into the trained model, and output the crack grade coefficient at the current moment; S4. Perform data processing and correlation analysis on the stress concentration factor, load frequency, load amplitude and stress intensity ratio of the rock mass at the current moment to generate a rock mass rock burst coefficient for evaluating the risk of rock mass rock burst; perform data processing and correlation analysis on the density, porosity and compressive strength of the rock mass at the current moment to generate a rock mass destruction coefficient for evaluating the risk of rock mass destruction; perform data processing and correlation analysis on the fracture grade coefficient, rock mass rock burst coefficient and rock mass destruction coefficient to generate a comprehensive evaluation coefficient for comprehensively evaluating the stability of the rock mass; S5. Evaluate the risk level of rock mass fractures based on the comprehensive evaluation coefficient.

[0008] Furthermore, the surface image of the rock mass is grayed, and the grayed surface image of the rock mass is threshold segmented to identify the crack part and the background part. The specific process is as follows: A drone equipped with a high-definition camera is used to capture panoramic images of the rock mass from the air, and the surface images of the rock mass collected are converted into grayscale images according to the following formula: ; in, is a grayscale image, , , are the pixel values ​​of the red, green, and blue channels of the surface image of the rock mass, respectively; The grayscale image is divided into multiple regions of equal size. For each region, the median of the grayscale value of the pixels in each region is calculated as the local threshold according to the grayscale value of the pixels in the region. The pixels in the region are compared with the corresponding local threshold and divided using the grid method. Assume that the entire image is , the segmented region image is ,in represents the index of the region image, , is the total number of regions; For each area , calculate the median of the grayscale values ​​of the pixels as the local threshold. The calculation formula for the median is: ; in, For the Local threshold of the region image; The Sobel operator is used to calculate the gradient of the image in each area in the horizontal and vertical directions. The process of obtaining the gradient amplitude is as follows: The grayscale value of the pixel and its adjacent pixel points are multiplied with the horizontal gradient template of the Sobel operator, and all the product results are added to obtain the horizontal gradient value of the pixel. The grayscale value of the pixel and its adjacent pixel points are multiplied with the vertical gradient template of the Sobel operator, and all the product results are added to obtain the vertical gradient value of the pixel. The formula for calculating the horizontal gradient value and the vertical gradient value is: ; ; in, , Respectively Row, No. The horizontal and vertical gradient values ​​of the column pixels. For the Row, No. The gray value of the pixel. is the row index, is the column index; The formula for generating the gradient amplitude of a pixel is: ; in, For the Row, No. The gradient magnitude of the column pixel; The gradient amplitude of each pixel in the regional image is compared with the local threshold The comparison process is as follows: when , then the pixel is considered to belong to the crack part and is retained; when , then the pixel is considered to belong to the background and is discarded.

[0009] Furthermore, the crack assessment model is composed of a deep neural network based on a multilayer perceptron, and the deep neural network of the multilayer perceptron includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, and the first hidden layer, the second hidden layer and the third hidden layer each have at least two neurons, and each uses ReLU as an activation function; The process of training the fracture assessment model is as follows: The maximum crack length, maximum crack width and maximum crack depth at the historical moment are used as input, and the crack grade coefficient is determined by the expert group scoring method. The crack grade coefficient is scored between 1 and 9, and the larger the crack grade coefficient, the higher the severity of the crack. The mean square error is used as the loss function. When the mean square error is When the crack assessment model is within the range, the training of the crack assessment model is completed; The maximum value of the crack length, the maximum value of the crack width and the maximum value of the crack depth at the current moment are input into the trained model, and the crack grade coefficient at the current moment is output.

[0010] Furthermore, the stress concentration factor, load frequency, load amplitude and stress intensity ratio of the rock mass at the current moment are processed and correlated to generate a rock burst coefficient for assessing the risk of rock burst in the rock mass, based on the following formula: ; in, is the rockburst coefficient of the rock mass at the current moment, is the stress concentration factor, is the load amplitude, is the load frequency, is the stress intensity ratio, is the weight coefficient of the combination of stress concentration factor and load amplitude, is the weight coefficient of the combination of load frequency and stress intensity ratio, is the weight coefficient of stress intensity ratio, On this basis, .

[0011] Furthermore, the density, porosity and compressive strength of the rock mass at the current moment are processed and correlated to generate a rock mass failure coefficient for assessing the risk of rock mass failure, based on the following formula: ; in, is the rock mass failure coefficient at the current moment, is the density of the rock mass, is the porosity of the rock mass, is the compressive strength of the rock mass, is the weight coefficient of rock mass density, is the weight coefficient of rock mass porosity, is the weight coefficient of rock mass compressive strength. On this basis, .

[0012] Furthermore, the fracture grade coefficient, rock burst coefficient and rock destruction coefficient at the current moment are subjected to data processing and correlation analysis to generate a comprehensive evaluation coefficient for comprehensive evaluation of rock stability, based on the following formula: ; in, is the comprehensive evaluation coefficient at the current moment. The comprehensive evaluation coefficient is used to combine the rock burst coefficient and the rock destruction coefficient to comprehensively evaluate the stability of the rock mass. is the crack level coefficient at the current moment, is the weight coefficient of rock burst coefficient, is the weight coefficient of rock mass failure coefficient, is the weight coefficient of the crack grade coefficient, and , , The specific value of is determined by the hierarchical analysis method.

[0013] Furthermore, the fracture risk level of the rock mass is evaluated according to the comprehensive evaluation coefficient at the current moment. The specific process is as follows: when , the fracture risk level is low risk, indicating that the fracture impact is small and the rock mass stability is good; when , the fracture risk level is medium risk, indicating that the fracture influence is gradually significant and the rock mass stability is medium; when , the fracture risk level is high risk, indicating that the fracture has a large impact and the rock mass stability is poor; Indicates the critical value at which the fracture risk level changes from low risk to medium risk. Indicates the critical value at which the fracture risk level changes from medium risk to high risk.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The present invention collects the length, width and depth of the crack, screens the maximum crack length, the maximum crack width and the maximum crack depth in all crack areas, inputs the maximum crack length, the maximum crack width and the maximum crack depth at the current moment into the crack assessment model, outputs the crack grade coefficient at the current moment, processes the stress concentration coefficient, load frequency, load amplitude and stress intensity ratio of the rock mass at the current moment to generate the rock burst coefficient, processes the density, porosity and compressive strength of the rock mass at the current moment to generate the rock destruction coefficient, processes the crack grade coefficient, the rock burst coefficient and the rock destruction coefficient to generate a comprehensive evaluation coefficient. This comprehensive assessment not only takes into account the number of cracks, but also combines the physical and mechanical properties of the rock mass, so as to better perform quantitative assessment of rock mass cracks, and improves the accuracy of rock mass stability assessment through in-depth analysis of key parameters, so as to more effectively identify potential risks. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a schematic diagram of the overall method flow of the present invention. DETAILED DESCRIPTION

[0016] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0017] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0018] Embodiment 1: See also Figure 1 , the present invention provides a technical solution: A method for quantitatively identifying rock mass fractures based on image processing technology, the specific steps include: S1. Obtain the surface image of the rock mass at the current moment, grayscale the surface image of the rock mass, perform threshold segmentation on the grayscaled surface image of the rock mass, identify the crack part and the background part, mark the connected domain of the crack part, identify all crack areas, extract features of the crack areas, obtain the geometric parameters of the cracks, and collect the load parameters and mechanical property parameters of the rock mass at the current moment. The geometric parameters of the cracks include the length, width and depth of the cracks, the load parameters include the stress concentration factor, the load frequency, the load amplitude and the stress intensity ratio, and the mechanical property parameters include the density, porosity and compressive strength of the rock mass; S2. based on the crack lengths, crack widths and crack depths of all crack regions, screening the maximum crack length, maximum crack width and maximum crack depth in all crack regions; S3. Construct a crack assessment model, take the maximum crack length, maximum crack width and maximum crack depth at the historical moment as input, and output the crack grade coefficient at the historical moment as a label, determine the crack grade coefficient through the expert group scoring method, train the crack assessment model, input the maximum crack length, maximum crack width and maximum crack depth at the current moment into the trained model, and output the crack grade coefficient at the current moment; S4. Perform data processing and correlation analysis on the stress concentration factor, load frequency, load amplitude and stress intensity ratio of the rock mass at the current moment to generate a rock mass rock burst coefficient for evaluating the risk of rock mass rock burst; perform data processing and correlation analysis on the density, porosity and compressive strength of the rock mass at the current moment to generate a rock mass destruction coefficient for evaluating the risk of rock mass destruction; perform data processing and correlation analysis on the fracture grade coefficient, rock mass rock burst coefficient and rock mass destruction coefficient to generate a comprehensive evaluation coefficient for comprehensively evaluating the stability of the rock mass; S5. Evaluate the risk level of rock mass fractures based on the comprehensive evaluation coefficient.

[0019] On the basis of the above embodiment, the surface image of the rock mass is grayed, and the grayed surface image of the rock mass is threshold segmented to identify the crack part and the background part. The specific process is as follows: A drone equipped with a high-definition camera is used to capture panoramic images of the rock mass from the air, and the surface images of the rock mass collected are converted into grayscale images according to the following formula: ; in, is a grayscale image, , , are the pixel values ​​of the red, green, and blue channels of the surface image of the rock mass, respectively; The grayscaled surface image of the rock mass is subjected to threshold segmentation to identify the crack part and the background part. The specific process is as follows: The grayscale image is divided into multiple regions of equal size. For each region, the median of the grayscale value of the pixels in each region is calculated as the local threshold according to the grayscale value of the pixels in the region. The pixels in the region are compared with the corresponding local threshold and divided using the grid method. Assume that the entire image is , the segmented region image is ,in represents the index of the region image, , is the total number of regions; For each area , calculate the median of the grayscale values ​​of the pixels as the local threshold, and the calculation formula for the median is: ; in, For the Local threshold of the region image; The Sobel operator is used to calculate the gradient of the image in each area in the horizontal and vertical directions. The process of obtaining the gradient amplitude is as follows: The grayscale value of the pixel and its adjacent pixel points are multiplied with the horizontal gradient template of the Sobel operator, and all the product results are added to obtain the horizontal gradient value of the pixel. The grayscale value of the pixel and its adjacent pixel points are multiplied with the vertical gradient template of the Sobel operator, and all the product results are added to obtain the vertical gradient value of the pixel. The formula for calculating the horizontal gradient value and the vertical gradient value is: ; ; in, , Respectively Row, No. The horizontal and vertical gradient values ​​of the column pixels. For the Row, No. The gray value of the pixel. is the row index, is the column index; The formula for generating the gradient amplitude of a pixel is: ; in, For the Row, No. The gradient magnitude of the column pixel; The gradient amplitude of each pixel in the regional image is compared with the local threshold The comparison process is as follows: when , then the pixel is considered to belong to the crack part and is retained; when , then the pixel is considered to belong to the background and is discarded.

[0020] On the basis of the above embodiment, the crack part is marked as a connected domain to identify the crack area. The specific process is as follows: Use the connected domain labeling algorithm (such as the 8-neighborhood or 4-neighborhood algorithm) to process each regional image, identify and label each crack region. The specific steps are as follows: Create a marker image of the same size as the region image, with an initial value of 0; Traverse each pixel, if the gradient amplitude of the pixel point in the regional image is greater than or equal to the local threshold and is not marked, it is marked as a new connected domain; Use depth-first search or breadth-first search to mark all connected pixels, assign the same label value, and then obtain the crack area.

[0021] On the basis of the above embodiment, the maximum crack length, the maximum crack width and the maximum crack depth in all crack regions are screened, and the specific process is as follows: Use the minimum circumscribed rectangle to extract the main axis of the crack, calculate the length of each crack, and record the maximum value; Measure the width through the outline of the crack area and count the maximum width of all cracks; Through laser measurement, the depth of all cracks is accurately extracted and the maximum depth of all cracks is obtained.

[0022] Based on the above embodiments, the collection method and calculation formula of stress concentration factor, load frequency, load amplitude, stress intensity ratio, density, porosity and compressive strength of rock mass are as follows: The stress concentration factor is a dimensionless value that measures the degree of stress increase at the location of geometric changes (cracks) in the rock mass. Its calculation formula is: ; ; in, is the stress concentration factor, is the maximum stress measured in the stress concentration area (crack), is the nominal stress, which is usually calculated during design. is the external load acting on the structure, is the cross-sectional area under load; The load cycle is the whole process from the initial load state, loading to the maximum load, and then unloading back to the initial state. The load frequency is the inverse of the load cycle. The calculation formula for the load amplitude is: ; in, is the load amplitude, is the maximum load during the load cycle, is the minimum load within the load cycle; The stress intensity ratio is defined as the ratio between the stress intensity of the rock mass and the compressive strength of the rock mass, and is calculated as: ; in, is the stress intensity ratio, is the stress intensity (the strength of the rock mass under tension or bending), is the compressive strength of the rock mass; The density of the rock mass is obtained by testing the mass and volume of the sample; The pore volume of the sample is measured using a gas pycnometer or liquid drainage method, and then the porosity of the rock mass is obtained using the pore volume of the sample. The calculation formula is: ; in, is the porosity of the rock mass, is the pore volume of the sample, is the total volume of the sample; The calculation formula of the compressive strength of rock mass is as follows: ; in, is the compressive strength of the rock mass, is the maximum load at which the rock mass fails.

[0023] On the basis of the above embodiment, the crack assessment model is composed of a deep neural network based on a multilayer perceptron, and the deep neural network of the multilayer perceptron includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, and the first hidden layer, the second hidden layer and the third hidden layer each have at least two neurons, and each uses ReLU as an activation function; In this embodiment, the input features of the deep learning network of the multilayer perceptron include: maximum crack length, maximum crack width and maximum crack depth, three features; The structure of the deep learning network of multi-layer perceptron is: Input layer: receives input of 3 features; The first hidden layer has 64 neurons and uses ReLU as the activation function. The second hidden layer has 32 neurons and also uses the ReLU activation function. The third hidden layer has 16 neurons and uses the ReLU activation function. Output layer: has a single neuron and a crack level coefficient; The process of training the fracture assessment model is as follows: The maximum crack length, maximum crack width and maximum crack depth at the historical moment are used as input, and the crack grade coefficient is determined by the expert group scoring method. The crack grade coefficient is scored between 1 and 9, and the larger the crack grade coefficient, the higher the severity of the crack. The mean square error is used as the loss function. When the mean square error is When the crack assessment model is within the range, the training of the crack assessment model is completed; The maximum value of the crack length, the maximum value of the crack width and the maximum value of the crack depth at the current moment are input into the trained model, and the crack grade coefficient at the current moment is output.

[0024] Based on the above embodiments, the correlation between the stress concentration factor, load frequency, load amplitude, stress intensity ratio and rock burst risk of the rock mass is as follows: The stress concentration factor is usually related to the increase of local stress. When the stress concentration factor increases, the stress in the local area is significantly higher than the average stress, which may make the rock mass more susceptible to damage and increase the risk of rock burst. Therefore, a higher stress concentration factor is usually positively correlated with a higher risk of rock burst.

[0025] The increase in load frequency leads to more frequent cyclic stress in the rock mass. This high-frequency load accelerates the occurrence of fatigue damage, increases the risk of brittle failure of the rock mass, and thus increases the risk of rock burst. Therefore, there is generally a positive correlation between load frequency and rock burst risk.

[0026] An increase in load amplitude means greater forces applied to the rock mass, which generally leads to higher stress concentrations and a greater risk of failure. Therefore, there is usually a positive correlation between load amplitude and rockburst risk.

[0027] The stress intensity ratio refers to the ratio of the tensile strength to the compressive strength of the rock mass. A lower stress intensity ratio usually means that the rock mass is more likely to fail under tension, thereby increasing the risk of rock burst. Therefore, there is a negative correlation between the stress intensity ratio and the risk of rock burst.

[0028] According to the correlation between the stress concentration factor, load frequency, load amplitude, stress intensity ratio and rock burst risk of the rock mass, the stress concentration factor, load frequency, load amplitude and stress intensity ratio of the rock mass at the current moment are processed and correlated to generate the rock burst coefficient for assessing the risk of rock burst in the rock mass, based on the following formula: ; in, is the rockburst coefficient of the rock mass at the current moment, is the stress concentration factor, is the load amplitude, is the load frequency, is the stress intensity ratio; It should be noted that the stress concentration factor The larger the rock burst coefficient is, the The larger the load, the greater the risk of rock burst. The larger the rockburst coefficient is, the The larger the load, the greater the risk of rock burst. The larger the rock burst coefficient is, the The larger the value, the greater the risk of rock burst. The larger the rockburst coefficient is, the The smaller it is, the smaller the risk of rock burst.

[0029] The reason why the above function form is used to express the functional relationship between the rockburst coefficient and the stress concentration factor, load frequency, load amplitude, and stress intensity ratio of the rock mass is as follows: First, the stress concentration factor reflects the degree of stress on the rock mass at a specific point or area, and the load amplitude represents the maximum external pressure applied to the rock mass. The product of the two is It can effectively reflect the stress state and potential damage risk of rock mass under external loads, so it is combined to reflect the effect on rock mass; Second, when the stress concentration factor and the load amplitude act together, their influence is not a simple linear superposition. Local stress concentration will cause the rock mass to produce different responses when subjected to force, and changes in the load amplitude will change this stress state. This interaction makes the combined effect of the two complicated on the response of the rock mass, usually showing a nonlinear relationship.

[0030] Third, the increase in load frequency leads to more frequent cyclic stress inside the rock mass. This high-frequency load accelerates the occurrence of fatigue damage and increases the risk of brittle failure of the rock mass. Therefore, the load frequency is set on the numerator of the fractional function. Fourth, introduce it in the denominator , can better reflect the marginal effect when the load frequency increases. When the load frequency is low, the influence of the denominator is relatively small. As the load frequency increases, the increase in the denominator will effectively reduce the value of the entire formula, thus reflecting the characteristic that the increase in carrying capacity is gradually slowing down. Adding 1 in the middle is to avoid the situation where the denominator is 0; fifth, and In the formula, as the load frequency increase, The increase is greater than Therefore, overall, with the increase in load frequency The increase in rock burst coefficient Increase, can reflect the rock burst coefficient of rock mass and load frequency positive correlation; Sixth, a lower stress intensity ratio usually means that the rock mass is more likely to fail under tension, thus increasing the risk of rock burst, so the stress intensity ratio is set in the denominator of the fractional function; Seventh, different load parameters have different influences on rock mass, so it is necessary to set the weight coefficient of load parameters. , , , which can more accurately reflect the influence of load parameters on rock mass and emphasize the interaction between various factors; In summary, the rockburst coefficient of the above form is set The calculation formula for .

[0031] In the formula, is the weight coefficient of the combination of stress concentration factor and load amplitude, is the weight coefficient of the load frequency, is the weight coefficient of stress intensity ratio; Stress concentration usually occurs at geometric discontinuities of structures, such as cracks, holes or other defects. These areas will be subjected to higher stress than the surrounding areas, resulting in an increased risk of local damage. The load amplitude is directly related to the external force applied to the rock mass. High-amplitude loads can significantly enhance the stress concentration effect, greatly reducing the bearing capacity of the rock mass. When stress concentration and high load amplitude work together, the probability of rock mass damage will increase significantly. Therefore, in design and analysis, the weights of stress concentration factor and load amplitude are Usually larger; Load frequency refers to the rate of change of the load applied to a structure or rock mass. Under dynamic load conditions, an increase in frequency can lead to an increase in the dynamic response of the rock mass, causing fatigue and microcrack expansion. The influence of frequency usually causes cumulative damage to the rock mass when it undergoes repeated loads, thereby affecting its stability. The stress intensity ratio is used to describe the performance of the rock mass under dynamic conditions. It is related to the tensile or compressive capacity of the material. Although it is also important in some cases, its impact is usually not as direct as stress concentration and load amplitude. Therefore, under static or low-frequency conditions, the material's performance is more stable, and the change in frequency has a relatively small impact on the overall bearing capacity, resulting in ; The stress intensity ratio has a negative correlation with the rockburst risk, that is, its increase will reduce the rockburst risk. It plays a role in reducing risk in the denominator and has the smallest weight; In summary, On the basis of .

[0032] As an implementation method, The value range is 0.4-0.5. The value range is 0.25-0.3, The value range is 0.1-0.2. The specific value is set by the technicians according to the actual situation and is not limited here.

[0033] Based on the above examples, the correlation between the density, porosity, compressive strength of the rock mass and the risk of rock mass failure is as follows: Higher density generally means that the rock mass is denser, which makes it more resistant to compression and thus reduces the risk of failure. Therefore, the density and risk of failure of a rock mass are usually negatively correlated.

[0034] Porosity refers to the volume fraction of pores in a rock. A higher porosity usually means that the rock mass structure is looser, which reduces its ability to resist external forces and increases the risk of damage. Therefore, porosity and damage risk are usually positively correlated.

[0035] Compressive strength is a measure of the ability of a rock mass to resist compression. Higher compressive strength usually means that the rock mass can withstand greater external pressure without failure. Therefore, it is negatively correlated with the risk of failure.

[0036] According to the correlation between the density, porosity, compressive strength and the risk of rock mass damage, the density, porosity and compressive strength of the rock mass at the current moment are processed and correlated to generate a rock mass damage coefficient for assessing the risk of rock mass damage. The formula is as follows: ; in, is the rock mass failure coefficient at the current moment, is the density of the rock mass, is the porosity of the rock mass, is the compressive strength of the rock mass; It should be noted that, as can be seen from the above description, the density of the rock mass The larger the rock mass failure coefficient The smaller the rock mass is, the smaller the risk of rock mass damage is. The larger the rock mass failure coefficient The larger the rock mass, the greater the risk of rock mass damage. The larger the rock mass failure coefficient The smaller it is, the lower the risk of rock damage.

[0037] The reasons why the above functional form is used to express the functional relationship between the rock mass failure coefficient and the density, porosity and compressive strength of the rock mass are as follows: First, placing porosity in the numerator of the formula directly reflects its positive correlation with the risk of damage. The higher the porosity, the greater the rock mass damage coefficient ( ), indicating that the risk of rock mass failure is higher.

[0038] Second, density is negatively correlated with the risk of damage, because higher density usually means that the rock mass is denser and more solid, and can better resist damage. Therefore, density is placed in the denominator to reduce the rock mass damage coefficient ( ).

[0039] Third, compressive strength is also negatively correlated with the risk of damage, because the rock with higher compressive strength can withstand greater external pressure. Putting compressive strength in the denominator can reflect its inhibitory effect on the risk of rock damage.

[0040] Fourth, the influence of each factor on the risk of rock mass failure may be different, so the weight coefficients (β1, β2, β3) are introduced to balance the effects of different variables.

[0041] Fifth, the density and compressive strength in the denominator appear in the form of addition and are adjusted through the weight coefficient to reflect the superimposed inhibitory effect of these factors on the risk of damage; the porosity in the numerator is amplified alone and directly acts on the damage coefficient, reflecting its more direct and significant impact on the risk of rock damage.

[0042] In the formula, is the weight coefficient of rock mass density, is the weight coefficient of rock mass porosity, is the weight coefficient of rock mass compressive strength; In the assessment of rock mass stability and failure risk, rock mass density It is usually a very important indicator. Rocks with higher density usually have higher strength and stability, so its weight coefficient should be set to a larger value, which means that in the failure risk analysis, the rock density The influence of and rock mass failure coefficient The correlation between them is the largest, so the largest weight is set ; Compressive strength It is the ability of rock mass to resist external loads. It is one of the core parameters for assessing the risk of rock mass damage. The higher the compressive strength, the greater the load that the rock mass can withstand and the lower the risk of damage. Although compressive strength is important, it is affected by many factors, including the composition, structure and external environment of the material. Therefore, although its weight is large, compared with the directness of density, the weight of compressive strength is Lower than rock mass density weight ; In the risk assessment of rock mass failure, density is a more direct indicator because it reflects the overall material composition of the rock mass and is generally considered to be the primary factor affecting the strength of the rock mass. Porosity Although important, its direct effect is often overshadowed by density and compressive strength and is therefore relative to rock mass density. , compressive strength , the porosity weight lowest; In summary, On the basis of .

[0043] As an implementation method, The value range is 0.4-0.5. The value range is 0.2-0.25, The value range is 0.3-0.35. The specific value is set by the technicians according to the actual situation and is not limited here.

[0044] On the basis of the above embodiment, the rockburst coefficient and rock destruction coefficient at the current moment are subjected to data processing and correlation analysis to generate a comprehensive evaluation coefficient for comprehensively evaluating the stability of the rock mass, based on the following formula: ; in, is the comprehensive evaluation coefficient at the current moment. The comprehensive evaluation coefficient is used to combine the rock burst coefficient and the rock destruction coefficient to comprehensively evaluate the stability of the rock mass. The larger it is, the lower the stability of the rock mass. is the crack level coefficient at the current moment; It should be noted that, as can be seen from the above description, the rockburst coefficient The larger the value, the greater the risk of rock burst and the rock mass damage coefficient The larger the value, the greater the risk of rock mass damage. Therefore, the comprehensive evaluation coefficient Rockburst coefficient , rock mass failure coefficient are positively correlated, so the comprehensive evaluation coefficient in the above weighted sum form is set The calculation formula of In the formula, is the weight coefficient of rock burst coefficient, is the weight coefficient of rock mass failure coefficient, is the weight coefficient of the crack grade coefficient, and , , The specific value of is determined by the hierarchical analysis method, and the specific logic is as follows: Rockburst coefficient , rock mass failure coefficient , crack grade coefficient , three indicators are marked, and the relative importance values ​​between each other are determined by the nine-scale method to construct a judgment matrix, in which the index of rock burst coefficient is marked as 1, the index of rock destruction coefficient is marked as 2, and the index of crack grade coefficient is marked as 3. The constructed judgment matrix for: ; in, , denotes the index of the coefficient, and , , indicating that the index is The importance of the coefficient of index v to the comprehensive evaluation coefficient is The specific value is determined by relevant experts using a 1-9 scoring method. Indicates that the index is The coefficient of is extremely important for the comprehensive evaluation coefficient compared to the coefficient with index v. Indicates that the index is The coefficient of is extremely unimportant to the comprehensive evaluation coefficient compared to the coefficient with index v; Divide each element value in the judgment matrix by the sum of its columns to obtain a normalized judgment matrix, calculate the mean of each row of element values ​​in the normalized judgment matrix, and take the mean of the first row of element values ​​as the rockburst coefficient The proportional coefficient is the mean of the second row of element values ​​as the rock mass failure coefficient The proportional coefficient is the mean of the third row of elements as the crack level coefficient The proportional coefficients are scaled in equal proportions with the constraint that the sum of the scaled values ​​is 1, and the values ​​obtained after scaling are used as the weights of the corresponding coefficients.

[0045] On the basis of the above embodiment, the fracture risk level of the rock mass is evaluated according to the comprehensive evaluation coefficient at the current moment. The specific process is as follows: when , the fracture risk level is low risk, indicating that the fracture impact is small and the rock mass stability is good; when , the fracture risk level is medium risk, indicating that the fracture influence is gradually significant and the rock mass stability is medium; when , the fracture risk level is high, indicating that the fracture has a large impact and the rock mass stability is poor.

[0046] Indicates the critical value at which the fracture risk level changes from low risk to medium risk. Indicates the critical value at which the fracture risk level changes from medium risk to high risk. and Finite element analysis software can be used to simulate the response of the rock mass under different values, evaluate the stability, and identify the influence of cracks on the stability of the rock mass through simulation results, so as to set the corresponding threshold value. and .

[0047] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0048] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented using software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by computer software, electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software depends on the specific application and design constraints of the technical solution.

[0049] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0050] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for quantitatively identifying rock mass fractures based on image processing technology, characterized in that: The specific steps include: S1. Obtain the surface image of the rock mass at the current moment, grayscale the surface image of the rock mass, perform threshold segmentation on the grayscaled surface image of the rock mass, identify the crack part and the background part, mark the connected domain of the crack part, identify all crack areas, extract features of the crack areas, obtain the geometric parameters of the cracks, and collect the load parameters and mechanical property parameters of the rock mass at the current moment. The geometric parameters of the cracks include the length, width and depth of the cracks, the load parameters include the stress concentration factor, the load frequency, the load amplitude and the stress intensity ratio, and the mechanical property parameters include the density, porosity and compressive strength of the rock mass; S2. based on the crack lengths, crack widths and crack depths of all crack regions, screening the maximum crack length, maximum crack width and maximum crack depth in all crack regions; S3. Construct a crack assessment model, take the maximum crack length, maximum crack width and maximum crack depth at the historical moment as input, and output the crack grade coefficient at the historical moment as a label, determine the crack grade coefficient through the expert group scoring method, train the crack assessment model, input the maximum crack length, maximum crack width and maximum crack depth at the current moment into the trained model, and output the crack grade coefficient at the current moment; S4. Perform data processing and correlation analysis on the stress concentration factor, load frequency, load amplitude and stress intensity ratio of the rock mass at the current moment to generate a rock mass rock burst coefficient for evaluating the risk of rock mass rock burst; perform data processing and correlation analysis on the density, porosity and compressive strength of the rock mass at the current moment to generate a rock mass destruction coefficient for evaluating the risk of rock mass destruction; perform data processing and correlation analysis on the fracture grade coefficient, rock mass rock burst coefficient and rock mass destruction coefficient to generate a comprehensive evaluation coefficient for comprehensively evaluating the stability of the rock mass; S5. Evaluate the risk level of rock mass fractures based on the comprehensive evaluation coefficient.

2. The method for quantitatively identifying rock mass fractures based on image processing technology according to claim 1 is characterized in that: The surface image of the rock mass is grayed, and the grayed surface image of the rock mass is threshold segmented to identify the crack part and the background part. The specific process is as follows: A drone equipped with a high-definition camera is used to capture panoramic images of the rock mass from the air, and the surface images of the rock mass collected are converted into grayscale images according to the following formula: ； in, is a grayscale image, , , are the pixel values ​​of the red, green, and blue channels of the surface image of the rock mass, respectively; The grayscale image is divided into multiple regions of equal size. For each region, the median of the grayscale value of the pixels in each region is calculated as the local threshold according to the grayscale value of the pixels in the region. The pixels in the region are compared with the corresponding local threshold and divided using the grid method. Assume that the entire image is , the segmented region image is ,in represents the index of the region image, , is the total number of regions; For each area , calculate the median of the grayscale values ​​of the pixels as the local threshold. The calculation formula for the median is: ； in, For the Local threshold of the region image; The Sobel operator is used to calculate the gradient of the image in each area in the horizontal and vertical directions. The process of obtaining the gradient amplitude is as follows: The grayscale value of the pixel and its adjacent pixel points are multiplied with the horizontal gradient template of the Sobel operator, and all the product results are added to obtain the horizontal gradient value of the pixel. The grayscale value of the pixel and its adjacent pixel points are multiplied with the vertical gradient template of the Sobel operator, and all the product results are added to obtain the vertical gradient value of the pixel. The formula for calculating the horizontal gradient value and the vertical gradient value is: ； ； in, , Respectively Row, No. The horizontal and vertical gradient values ​​of the column pixels. For the Row, No. The gray value of the pixel. is the row index, is the column index; The formula for generating the gradient amplitude of a pixel is: ； in, For the Row, No. The gradient magnitude of the column pixel; The gradient amplitude of each pixel in the regional image is compared with the local threshold The comparison process is as follows: when , then the pixel is considered to belong to the crack part and is retained; when , then the pixel is considered to belong to the background and is discarded.

3. The method for quantitatively identifying rock mass fractures based on image processing technology according to claim 2 is characterized in that: The crack assessment model is composed of a deep neural network based on a multilayer perceptron, and the deep neural network of the multilayer perceptron includes an input layer, a first hidden layer, a second hidden layer, a third hidden layer and an output layer, and the first hidden layer, the second hidden layer and the third hidden layer each have at least two neurons, and each uses ReLU as an activation function; The process of training the fracture assessment model is as follows: The maximum crack length, maximum crack width and maximum crack depth at the historical moment are used as input, and the crack grade coefficient is determined by the expert group scoring method. The crack grade coefficient is scored between 1 and 9, and the larger the crack grade coefficient, the higher the severity of the crack. The mean square error is used as the loss function. When the mean square error is When the crack assessment model is within the range, the training of the crack assessment model is completed; The maximum value of the crack length, the maximum value of the crack width and the maximum value of the crack depth at the current moment are input into the trained model, and the crack grade coefficient at the current moment is output.

4. The method for quantitatively identifying rock mass fractures based on image processing technology according to claim 3 is characterized in that: The stress concentration factor, load frequency, load amplitude and stress intensity ratio of the rock mass at the current moment are processed and correlated to generate the rock burst coefficient for assessing the risk of rock burst in the rock mass. The formula is as follows: ； in, is the rockburst coefficient of the rock mass at the current moment, is the stress concentration factor, is the load amplitude, is the load frequency, is the stress intensity ratio, is the weight coefficient of the combination of stress concentration factor and load amplitude, is the weight coefficient of the load frequency, is the weight coefficient of stress intensity ratio, On the basis of .

5. The method for quantitatively identifying rock mass fractures based on image processing technology according to claim 4 is characterized in that: The density, porosity and compressive strength of the rock mass at the current moment are processed and correlated to generate a rock mass failure coefficient for assessing the risk of rock mass failure. The formula is as follows: ； in, is the rock mass failure coefficient at the current moment, is the density of the rock mass, is the porosity of the rock mass, is the compressive strength of the rock mass, is the weight coefficient of rock mass density, is the weight coefficient of rock mass porosity, is the weight coefficient of rock mass compressive strength. On the basis of .

6. The method for quantitatively identifying rock mass fractures based on image processing technology according to claim 5 is characterized in that: The fracture grade coefficient, rock burst coefficient and rock destruction coefficient at the current moment are processed and correlated to generate a comprehensive evaluation coefficient for comprehensive evaluation of rock stability, based on the following formula: ； in, is the comprehensive evaluation coefficient at the current moment. The comprehensive evaluation coefficient is used to combine the rock burst coefficient and the rock destruction coefficient to comprehensively evaluate the stability of the rock mass. is the crack level coefficient at the current moment, is the weight coefficient of rock mass rockburst coefficient, is the weight coefficient of rock mass failure coefficient, is the weight coefficient of the crack grade coefficient, and , , The specific value of is determined by the hierarchical analysis method.

7. The method for quantitatively identifying rock mass fractures based on image processing technology according to claim 6 is characterized in that: According to the comprehensive evaluation coefficient at the current moment, the fracture risk level of the rock mass is evaluated. The specific process is as follows: when , the fracture risk level is low risk, indicating that the fracture impact is small and the rock mass stability is good; when , the fracture risk level is medium risk, indicating that the fracture influence is gradually significant and the rock mass stability is medium; when , the fracture risk level is high risk, indicating that the fracture has a large impact and the rock mass stability is poor; Indicates the critical value at which the fracture risk level changes from low risk to medium risk. Indicates the critical value at which the fracture risk level changes from medium risk to high risk.

Images