Image recognition method and system based on intelligent perception analysis

Abstract

The invention relates to the technical field of image recognition, and particularly discloses an image recognition method and system based on intelligent perception analysis. The method comprises the steps of firstly obtaining rock stratum image information of a mine, then performing feature extraction on the rock stratum image information based on intelligent perception analysis to obtain crack features, rock stratum deformation features and rock stratum structure features, then obtaining a first risk index according to the crack features in sequence, and then obtaining a second risk index according to the rock stratum deformation features. Then obtaining a third risk index according to the rock stratum structure characteristics, obtaining a comprehensive risk index according to the first risk index, the second risk index and the third risk index, finally judging whether the comprehensive risk index is greater than a preset threshold value, and if the comprehensive risk index is greater than the preset threshold value, determining that the rock stratum is damaged. In this way, the high-risk area can be found in time by judging the rock stratum risk in advance, and meanwhile, the problem of mine disasters caused by identification time difference can be avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of image recognition, and in particular to an image recognition method and system based on intelligent perception analysis. Background Art

[0002] With the rapid development of information technology and the advent of the industrial age, digital mines have become the only way for the sustainable development of the mining industry. Digital mines have achieved all-round, digital management of mine production and operation activities by integrating digital information, databases, computer networks, and simulation technologies to reduce the risks brought by digital mines. At present, the risk management of rock formations in mines is usually carried out by taking photos regularly by cameras for comparison, and problems are discovered by comparing photos. However, through this image recognition and comparative analysis, the results of the comparative analysis usually need to be identified by professional personnel, which leads to a certain time difference in identification. The existing time difference cannot remind the staff in time, which leads to mining accidents. Therefore, an image recognition method based on intelligent perception analysis is needed to solve the above problems. Summary of the invention

[0003] The purpose of the present invention is to provide an image recognition method and system based on intelligent perception analysis to solve the technical problems raised in the above-mentioned background technology.

[0004] To achieve the above object, the present invention provides the following technical solutions: An image recognition method based on intelligent perception analysis, comprising: Obtain rock formation image information of the mine; Extracting features of the rock formation image information based on intelligent perception analysis to obtain crack features, rock formation deformation features, and rock formation structure features; Acquire crack width, crack length and crack density according to the crack characteristics, and acquire crack trend risk index according to the crack width and crack length; Obtaining a rock formation fragility index according to the fracture density, and obtaining a first risk index according to the rock formation fragility index and the fracture trend risk index; Acquire rock formation fold information and rock formation bending information according to the rock formation deformation characteristics, and acquire a second risk index according to the rock formation fold information and rock formation bending information; Obtaining porosity and permeability according to the structural characteristics of the rock formation, and obtaining rock formation osmotic pressure according to the porosity and the permeability, and using the rock formation osmotic pressure as a third risk index; Obtaining a comprehensive risk index according to the first risk index, the second risk index and the third risk index; Determining whether the comprehensive risk index is greater than a preset threshold; If the comprehensive risk index is greater than a preset threshold, it is determined that there is a rock formation risk in the area corresponding to the rock formation image information of the mine.

[0005] Preferably, the step of extracting features from the rock formation image information based on intelligent perception analysis to obtain crack features, rock formation deformation features and rock formation structure features comprises: Grayscale processing is performed on the rock formation image information based on intelligent perception analysis to obtain black-and-white rock formation image information, a plurality of black stripes are screened out according to the black-and-white surface image information, coordinates are respectively set at both ends of each individual black stripe in the plurality of black stripes to obtain a plurality of first coordinates and a plurality of second coordinates, and crack lengths of the plurality of black stripes are sequentially calculated according to the plurality of first coordinates and the plurality of second coordinates, wherein the intelligent perception analysis includes image preprocessing; Obtaining the maximum diameter of each individual black stripe among the plurality of black stripes, and taking the maximum diameter of the black stripe as the crack width; Obtain edge information of the rock layer image information of the mine based on an edge detection algorithm, and obtain the corresponding edge region area according to the edge information; Acquire the corresponding number of black stripes according to the plurality of black stripes, and calculate the crack density according to the number of black stripes and the area of ​​the edge region; The overall information corresponding to the crack length, crack width and crack density is used as crack characteristics; Acquire initial preset image information, and select the same point position of the initial preset image information and the rock layer image information of the mine; Acquire a first pixel point and a second pixel point at the same point, and acquire pixel displacement information according to the first pixel point and the second pixel point, and use the pixel displacement information as a rock formation deformation feature; Acquire multiple pixel pairs of rock formation image information, wherein the pixel pairs are adjacent pixel points; Based on the gray level co-occurrence matrix, multiple pixel pairs are traversed to obtain multiple adjacent pixel directions and multiple adjacent pixel distances, and the linear texture structure of the rock formation is obtained according to the multiple adjacent pixel directions and multiple adjacent pixel distances, and the linear texture structure of the rock formation is used as the rock formation structure feature.

[0006] Preferably, the step of obtaining the crack trend risk index according to the crack width and crack length comprises: Acquire a first expansion rate within a preset time according to the crack width and a preset initial crack width; Acquire a second expansion rate within a preset time according to the crack length and a preset initial crack length; Performing weighted calculation on the first expansion rate and the second expansion rate to obtain a comprehensive crack expansion rate; Obtaining a rock formation change rate within a preset time according to the rock formation deformation characteristics; Obtaining the thickness of the rock layer, and obtaining a crack width ratio value according to the crack width and the rock layer thickness, and obtaining a corresponding first weight value according to the crack width ratio value; Obtain a fracture length ratio value according to the fracture length and the rock layer thickness, and a second weight value corresponding to the fracture length ratio value; Obtaining a corresponding third weight value according to the comprehensive crack expansion rate; The fracture trend risk index is calculated according to the fracture width proportion value, the fracture length proportion value, the fracture comprehensive expansion rate, the rock formation change rate, the first weight value, the second weight value and the third weight value, wherein the calculation formula is:

[0007] in, Indicates the crack direction risk index, represents the crack width ratio, represents the crack length ratio, represents the comprehensive crack expansion rate, represents the rock formation change rate, a represents the first weight value, b represents the second weight value, and c represents the third weight value.

[0008] Preferably, the step of obtaining a rock formation fragility index according to the fracture density, and obtaining a first risk index according to the rock formation fragility index and the fracture trend risk index comprises: Obtaining a first area of ​​the rock layer in a preset area, dividing the first area into unit areas to obtain a plurality of unit areas; Acquire a plurality of corresponding crack densities according to the plurality of unit areas, and calculate an average crack density according to the plurality of crack densities; The standard crack density is calculated according to the plurality of crack densities and the average crack density, wherein the calculation formula is: ; in, represents the standard crack density, represents the lth crack density, represents the number of crack density, l represents the serial number of crack density, represents the average crack density; The crack density index is calculated according to the standard crack density and the average crack density, wherein the calculation formula is: ; in, represents the crack density index, represents the average crack density, represents the standard crack density; Calculating a density-area ratio value according to the ratio of the fracture density index to the first area, and using the density-area ratio value as a rock formation fragility index; The rock formation fragility index and the fracture trend risk index are weighted and calculated to obtain a first risk index.

[0009] Preferably, the step of obtaining a second risk index according to the rock formation fold information and the rock formation bending information comprises: Acquire a rock formation fold wave crest and a rock formation fold wave trough according to the rock formation fold information, and acquire a rock formation fold amplitude according to the rock formation fold wave crest and the rock formation fold wave trough; Calculating a rock formation fold amplitude difference according to the rock formation fold amplitude and a preset rock formation fold amplitude; Acquire a rock formation bending angle according to the rock formation bending information, and calculate a rock formation bending angle difference according to the rock formation bending angle and a preset rock formation bending angle; The rock formation fold amplitude difference and the rock formation bending angle difference are weightedly calculated to obtain a second risk index.

[0010] Preferably, the step of obtaining the formation osmotic pressure according to the porosity and the permeability comprises: Get fluid viscosity; The pore fluid diffusion coefficient is calculated according to the porosity, the fluid viscosity and the permeability, wherein the calculation formula is: ; in, represents the pore fluid diffusion coefficient, represents the permeability, represents the porosity, Indicates fluid viscosity; The preset hydrostatic pressure of the rock is obtained, and the formation permeability pressure is calculated according to the ratio of the preset hydrostatic pressure to the diffusion coefficient of the pore fluid.

[0011] Preferably, the step of obtaining a comprehensive risk index according to the first risk index, the second risk index and the third risk index comprises: Obtaining a corresponding first risk index weight value according to the first risk index; Obtaining a corresponding second risk index weight value according to the second risk index; The comprehensive risk index is calculated according to the first risk index, the second risk index and the third risk index, the first risk index weight value and the second risk index weight value, wherein the calculation formula is:

[0012] in, represents the comprehensive risk index, represents the first risk index, represents the second risk index, represents the third risk index, d represents the weight value of the first risk index, and e represents the weight value of the second risk index.

[0013] The present application also provides an image recognition system based on intelligent perception analysis, comprising: A first acquisition module is used to acquire rock layer image information of the mine; A first extraction module is used to extract features from the rock formation image information based on intelligent perception analysis to obtain crack features, rock formation deformation features and rock formation structure features; A second acquisition module is used to acquire crack width, crack length and crack density according to the crack characteristics, and acquire a crack trend risk index according to the crack width and crack length; A third acquisition module is used to acquire a rock formation fragility index according to the fracture density, and to acquire a first risk index according to the rock formation fragility index and the fracture trend risk index; A fourth acquisition module, configured to acquire rock formation fold information and rock formation bending information according to the rock formation deformation characteristics, and to acquire a second risk index according to the rock formation fold information and rock formation bending information; a fifth acquisition module, configured to acquire porosity and permeability according to the structural characteristics of the rock formation, and acquire rock formation osmotic pressure according to the porosity and the permeability, and use the rock formation osmotic pressure as a third risk index; A sixth acquisition module, configured to acquire a comprehensive risk index according to the first risk index, the second risk index and the third risk index; A first judgment module is used to judge whether the comprehensive risk index is greater than a preset threshold; If the comprehensive risk index is greater than a preset threshold, it is determined that there is a rock formation risk in the area corresponding to the rock formation image information of the mine.

[0014] The present application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0015] The present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above method are implemented.

[0016] The beneficial effects of the present application are as follows: the present application first obtains the rock formation image information of the mine, and then extracts features of the rock formation image information based on intelligent perception analysis to obtain crack features, rock formation deformation features and rock formation structure features, and then obtains the first risk index according to the crack features in turn, and then obtains the second risk index according to the rock formation deformation features, and then obtains the third risk index according to the rock formation structure features, and then obtains the comprehensive risk index according to the first risk index, the second risk index and the third risk index, and finally, determines whether the comprehensive risk index is greater than a preset threshold value. If the comprehensive risk index is greater than the preset threshold value, it is determined that there is a rock formation risk in the area corresponding to the rock formation image information of the mine. In this way, through the early judgment of rock formation risks, high-risk areas can be discovered in time, early warnings can be given and corresponding protective measures can be taken to avoid the occurrence of geological disasters. Compared with traditional manual judgments, the automated risk assessment system can respond faster and more accurately, improve the safety of mine operations, and at the same time avoid the problem of mine accidents caused by identification time differences. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The figure is a schematic diagram of a method flow of an embodiment of the present application.

[0018] Figure 2 A schematic diagram of the system structure of an embodiment of the present application.

[0019] Figure 3 A schematic diagram of the internal structure of a computer device according to an embodiment of the present application.

[0020] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0022] like Figure 1-3 As shown, the present application provides an image recognition method based on intelligent perception analysis, comprising: S1. Obtain rock formation image information of the mine; S2. Extracting features from the rock formation image information based on intelligent perception analysis to obtain crack features, rock formation deformation features, and rock formation structure features; S3. Obtaining crack width, crack length and crack density according to the crack characteristics, and obtaining a crack trend risk index according to the crack width and crack length; S4. Obtaining a rock formation fragility index according to the fracture density, and obtaining a first risk index according to the rock formation fragility index and the fracture trend risk index; S5. Obtaining rock formation fold information and rock formation bending information according to the rock formation deformation characteristics, and obtaining a second risk index according to the rock formation fold information and rock formation bending information; S6. Obtaining porosity and permeability according to the structural characteristics of the rock formation, and obtaining rock formation osmotic pressure according to the porosity and the permeability, and using the rock formation osmotic pressure as a third risk index; S7. Obtain a comprehensive risk index according to the first risk index, the second risk index and the third risk index; S8. Determine whether the comprehensive risk index is greater than a preset threshold; If the comprehensive risk index is greater than a preset threshold, it is determined that there is a rock formation risk in the area corresponding to the rock formation image information of the mine.

[0023] As described in the above steps S1-S8, since the current rock formation risk management in the mine is usually carried out by regularly taking photos with a camera for comparison, and discovering problems by comparing the photos, and through this image recognition and comparative analysis, the results of the comparative analysis usually need to be identified by professional personnel, which will lead to a certain time difference in identification, and the existing time difference cannot remind the staff in time, which will lead to the occurrence of mining accidents. Therefore, the present invention first obtains the rock formation image information of the mine, so that the rock formation image information provides a basis for subsequent feature extraction and risk assessment, and then extracts features from the rock formation image information based on intelligent perception analysis to obtain crack features, rock formation deformation features and rock formation structure features, wherein the intelligent perception analysis includes image Preprocessing, edge detection and grayscale processing, etc., and through the crack characteristics, rock deformation characteristics and rock structure characteristics, the rock formation risk can be analyzed from multiple angles, which can avoid accidental factors caused by single data, and at the same time, the complex image information is converted into specific geometric and physical characteristics, which provides quantitative data support for subsequent risk assessment. Intelligent perception analysis can automatically identify and quantify subtle changes in the rock formation, avoiding the time difference and subjectivity of manual identification. Since cracks are one of the main causes of rock formation instability, the crack width, crack length and crack density are obtained according to the crack characteristics, and the crack direction risk index is obtained according to the crack width and crack length. In this way, the crack direction risk index can help predict whether the crack will expand further, thereby providing Measures are taken in advance to prevent rock rupture or landslide, and then the rock fragility index is obtained according to the crack density, and the first risk index is obtained according to the rock fragility index and the crack trend risk index, and the rock fragility index reflects the overall fragility of the rock under the condition of crack development, and the first risk index is a preliminary risk assessment result that comprehensively considers the crack characteristics and rock structure, providing a basis for further risk analysis, and then the rock fold information and rock fold bending information are obtained according to the rock deformation characteristics, and the second risk index is obtained according to the rock fold information and rock bending information, and the deformation characteristics of the rock directly affect its stress distribution and bearing capacity. The second risk index can evaluate the possibility of instability of the rock under deformation and help identify potential landslides or collapses. Collapse area, then, obtain porosity and permeability according to the structural characteristics of the rock formation, and obtain rock formation osmotic pressure according to the porosity and the permeability, and use the rock formation osmotic pressure as the third risk index. The change of osmotic pressure will affect the effective stress of the rock formation and reduce its bearing capacity. The greater the change of osmotic pressure, the greater the risk of rock formation instability. The third risk index can evaluate the stability of the rock formation under the action of fluid pressure, and obtain a comprehensive risk index according to the first risk index, the second risk index and the third risk index. In this way, the comprehensive risk index can comprehensively evaluate the overall risk of the rock formation, and then more accurately predict the possibility of rock formation instability, providing a scientific basis for mine safety management. Finally, it is determined whether the comprehensive risk index is greater than the preset threshold.If the comprehensive risk index is greater than the preset threshold, it is determined that the area corresponding to the rock formation image information of the mine has rock formation risk. In this way, through the early judgment of rock formation risk, high-risk areas can be discovered in time, early warnings can be given and corresponding protective measures can be taken to avoid the occurrence of geological disasters. Compared with traditional manual judgment, the automated risk assessment system can respond faster and more accurately, improve the safety of mine operations, and avoid the problem of mine accidents caused by identification time differences.

[0024] In one embodiment, the step S2 of extracting features from the rock formation image information based on intelligent perception analysis to obtain crack features, rock formation deformation features, and rock formation structure features includes: S201, grayscale processing is performed on the rock formation image information based on intelligent perception analysis to obtain black-and-white rock formation image information, a plurality of black stripes are screened out according to the black-and-white surface image information, coordinates are respectively set at both ends of each individual black stripe in the plurality of black stripes to obtain a plurality of first coordinates and a plurality of second coordinates, and crack lengths of the plurality of black stripes are sequentially calculated according to the plurality of first coordinates and the plurality of second coordinates, wherein the intelligent perception analysis includes image preprocessing; S202, obtaining a maximum diameter of each individual black stripe among the plurality of black stripes, and taking the maximum diameter of the black stripe as the crack width; S203, obtaining edge information of the rock layer image information of the mine based on an edge detection algorithm, and obtaining a corresponding edge region area according to the edge information; S204, obtaining the corresponding number of black stripes according to the plurality of black stripes, and calculating the crack density according to the number of black stripes and the area of ​​the edge region; S205, taking the overall information corresponding to the crack length, crack width and crack density as crack characteristics; S206, obtaining initial preset image information, and selecting the same point position of the initial preset image information and the rock layer image information of the mine; S207, obtaining a first pixel point and a second pixel point at the same point, and obtaining pixel displacement information according to the first pixel point and the second pixel point, and using the pixel displacement information as a rock formation deformation feature; S208, obtaining multiple pixel pairs of rock formation image information, wherein the pixel pairs are adjacent pixel points; S209, traversing the plurality of pixel pairs based on the gray level co-occurrence matrix to obtain a plurality of adjacent pixel directions and a plurality of adjacent pixel distances, and acquiring a linear texture structure of the rock formation according to the plurality of adjacent pixel directions and the plurality of adjacent pixel distances, and using the linear texture structure of the rock formation as a structural feature of the rock formation.

[0025] As described in the above steps S201-S209, the present invention first performs grayscale processing on the rock formation image information based on intelligent perception analysis to obtain black-and-white rock formation image information, and selects a plurality of black stripes according to the black-and-white surface image information, and respectively sets coordinates at both ends of each individual black stripe in the plurality of black stripes to obtain a plurality of first coordinates and a plurality of second coordinates, and sequentially calculates the crack lengths of the plurality of black stripes according to the plurality of first coordinates and the plurality of second coordinates, wherein the intelligent perception analysis includes image preprocessing, so that after the grayscale processing simplifies the image data, the calculation complexity can be reduced, and at the same time, the contrast between the crack and the background is enhanced, so that the subsequent crack detection is more accurate, and the screening of black stripes is the first step in crack identification, ensuring the subsequent The object of analysis is the real crack rather than other noise or texture. Since the crack width is one of the important parameters for evaluating the severity of the crack, wider cracks usually mean greater geological risks. Then the maximum diameter of each individual black stripe in the multiple black stripes is obtained, and the maximum diameter of the black stripe is used as the crack width. In this way, accurate measurement of the crack width helps to quantify the harmfulness of the crack. Then, the edge information of the rock formation image information of the mine is obtained based on the edge detection algorithm, and the corresponding edge region area is obtained according to the edge information. The calculation of the edge region area provides a basis for the subsequent crack density analysis. Since the crack density reflects the number of cracks per unit area, it is an important indicator for evaluating the overall fragility of the rock formation. Therefore , according to the plurality of black stripes, the corresponding number of black stripes is obtained, and the crack density is calculated according to the number of black stripes and the area of ​​the edge region. Such high-density cracks mean that the rock formation may be more prone to rupture or landslide, so the calculation of crack density is crucial for risk assessment, and then the coordinated information corresponding to the crack length, crack width and crack density is used as the crack feature. By using the coordinated information as the information of the crack feature, it is possible to avoid risk misjudgment caused by a single information. Here, the coordinated information is used to comprehensively classify the crack length, crack width and crack density parameters into crack features, and then the initial preset image information is obtained, and the initial preset image information and the rock formation image information of the mine are selected at the same point to obtain the same point. The first pixel point and the second pixel point at the position are determined, and pixel displacement information is obtained according to the first pixel point and the second pixel point, and the pixel displacement information is used as the rock formation deformation feature. In this way, the subtle displacement and deformation of the rock formation can be detected through the rock formation deformation feature, which is very critical for early detection of signs of rock formation instability, especially when the rock formation is slowly deformed. Timely capture of these changes can provide early warning of potential risks. Since the texture of the rock surface is determined by its internal mineral composition, cracks, bedding and other microstructures, these microstructures appear as different grayscale change patterns in the image. The grayscale co-occurrence matrix can effectively capture the characteristics of these microstructures by counting the co-occurrence frequency of pixels with different grayscale values ​​in a specific direction and distance. On this basis,Obtain multiple pixel pairs of rock layer image information, wherein the pixel pairs are adjacent pixel points, and traverse the multiple pixel pairs based on the gray level co-occurrence matrix to obtain multiple adjacent pixel directions and multiple adjacent pixel distances, wherein the gray level co-occurrence matrix is ​​a commonly used texture analysis tool that can capture the linear structural characteristics of the rock layer surface (rock layer linear texture), help the system identify the microscopic texture changes of the rock layer, and obtain the linear texture structure of the rock layer according to the multiple adjacent pixel directions and multiple adjacent pixel distances, and use the linear texture structure of the rock layer as the structural feature of the rock layer, and the linear texture structure of the rock layer reflects the structural characteristics of the rock layer, such as cracks, bedding, etc. These structural characteristics are of great significance for evaluating the stability of the rock layer and predicting geological disasters. The extraction of linear texture structure can help the system understand the internal structure of the rock layer more deeply and improve the accuracy of risk assessment. ,

[0026] In one embodiment, the step S3 of obtaining the crack direction risk index according to the crack width and crack length includes: S301, obtaining a first expansion rate within a preset time according to the crack width and a preset initial crack width; S302, obtaining a second expansion rate within a preset time according to the crack length and a preset initial crack length; S303, performing weighted calculation on the first expansion rate and the second expansion rate to obtain a comprehensive crack expansion rate; S304, obtaining a rock formation change rate within a preset time according to the rock formation deformation characteristics; S305, obtaining the thickness of the rock layer, and obtaining a crack width ratio value according to the crack width and the rock layer thickness, and obtaining a corresponding first weight value according to the crack width ratio value; S306, obtaining a crack length ratio value according to the crack length and the rock layer thickness, and obtaining a second weight value corresponding to the crack length ratio value; S307, obtaining a corresponding third weight value according to the comprehensive crack expansion rate; S308, calculating the fracture trend risk index according to the fracture width proportion value, the fracture length proportion value, the fracture comprehensive expansion rate, the rock formation change rate, the first weight value, the second weight value and the third weight value, wherein the calculation formula is:

[0027] in, Indicates the crack direction risk index, represents the crack width ratio, represents the crack length ratio, represents the comprehensive crack expansion rate, represents the rock formation change rate, a represents the first weight value, b represents the second weight value, and c represents the third weight value.

[0028] As described in the above steps S301-S308, since the expansion rate of the crack width reflects the growth rate of the crack over time and is an important parameter for evaluating the possibility of crack expansion, a higher expansion rate means that the crack may expand rapidly in a short period of time, increasing the risk of rock formation instability. The present invention first obtains a first expansion rate within a preset time based on the crack width and the preset initial crack width, and then obtains a second expansion rate within a preset time based on the crack length and the preset initial crack length. Then, the first expansion rate and the second expansion rate are weighted to obtain a comprehensive crack expansion rate. The comprehensive crack expansion rate comprehensively considers the growth of the crack in the width and length directions, and can more comprehensively reflect the overall expansion trend of the crack. Through weighted calculation, the system can obtain the first expansion rate within a preset time based on the crack width and the preset initial crack length. The influence degree of expansion rate in different directions on risk is adjusted to ensure that the assessment result is more accurate. Then, the rock formation change rate within the preset time is obtained according to the rock formation deformation characteristics. The rock formation change rate reflects the overall deformation of the rock formation, including crack expansion, folding, bending, etc. The faster the rock formation change rate, the more unstable the stress state of the rock formation and the higher the risk of instability. Therefore, the rock formation change rate is one of the key parameters for evaluating the overall stability of the rock formation. Then, the rock formation thickness is obtained, and the crack width ratio is obtained according to the crack width and the rock formation thickness. The crack length ratio reflects the extension degree of the crack in the rock formation. Longer cracks may run through the entire rock formation, causing rock formation fracture or landslide, and the crack width ratio reflects the relative size of the crack relative to the rock formation thickness. A larger crack width ratio means that the cracks cause more serious damage to the rock structure, increasing the possibility of rock instability. A corresponding first weight value is obtained according to the crack width ratio, a crack length ratio is obtained according to the crack length and the rock thickness, a second weight value corresponding to the crack length ratio is obtained, and a corresponding third weight value is obtained according to the comprehensive crack expansion rate. The crack direction risk index is calculated according to the crack width ratio, the crack length ratio, the comprehensive crack expansion rate, the rock change rate, the first weight value, the second weight value and the third weight value. In this way, the crack direction risk index (CORI) provides an intuitive risk assessment tool for mine managers, which can warn of potential geological risks in advance. By monitoring the changes in CORI, managers can take preventive measures in time, such as reinforcing slopes, adjusting mining plans, etc., to avoid major accidents.

[0029] In one embodiment, the step S4 of obtaining a rock formation fragility index according to the fracture density, and obtaining a first risk index according to the rock formation fragility index and the fracture trend risk index comprises: S401, obtaining a first area of ​​a rock layer in a preset area, and dividing the first area into unit areas to obtain a plurality of unit areas; S402, acquiring a plurality of corresponding crack densities according to the plurality of unit areas, and calculating an average crack density according to the plurality of crack densities; S403, calculating a standard crack density according to the plurality of crack densities and the average crack density, wherein the calculation formula is: ; in, represents the standard crack density, represents the lth crack density, represents the number of crack density, l represents the serial number of crack density, represents the average crack density; S404, calculating a crack density index according to the standard crack density and the average crack density, wherein the calculation formula is: ; in, represents the crack density index, represents the average crack density, represents the standard crack density; S405, calculating a density-area ratio value according to the ratio of the fracture density index to the first area, and using the density-area ratio value as a rock formation fragility index; S406: Perform weighted calculation on the rock formation fragility index and the fracture trend risk index to obtain a first risk index.

[0030] As described in the above steps S401-S406, the present invention first obtains a first area of ​​the rock formation in a preset area, divides the first area according to the unit area, and obtains multiple unit areas. By dividing a large area into multiple unit areas, the crack density distribution at different locations can be analyzed more finely to avoid the risk of the overall average value covering up the local high-density crack area. This block processing method is helpful to identify potential high-risk areas, and then obtain the corresponding multiple crack densities based on the multiple unit areas. By calculating the crack density of each unit area, the system can accurately quantify the crack development in different areas, provide data support for subsequent vulnerability assessments, and calculate the average crack density based on the multiple crack densities. Then, the standard crack density is calculated based on the multiple crack densities and the average crack density. The standard crack density reflects the degree of discreteness of the crack density, that is, the difference in crack density between different areas. A larger standard crack density indicates that the crack density distribution is uneven, and high-density cracks may exist in some areas, increasing the instability. Risk, by introducing the standard crack density, the system can more accurately evaluate the overall fragility of the rock formation, and then calculate the crack density index based on the standard crack density and the average crack density. The crack density index comprehensively considers the average level and dispersion of the crack density, and can more comprehensively reflect the development of cracks in the rock formation. A higher crack density index value means that the cracks in the rock formation are more seriously developed and the fragility is higher. By calculating the crack density index, the fragility of the rock formation can be quantified, providing a basis for subsequent risk assessment. Finally, the density-area ratio value is calculated based on the ratio of the crack density index to the first area, and the density-area ratio value is used as the rock formation fragility index. In this way, the rock formation fragility index not only considers the crack density, but also the influence of the unit area, and can more accurately reflect the fragility of the rock formation. A higher rock formation fragility index indicates that the cracks in the rock formation are more seriously developed and there is a higher risk of instability. By calculating the rock formation fragility index, the fragility of the rock formation can be comprehensively evaluated, and a basis for the subsequent calculation of the comprehensive risk index can be provided.

[0031] In one embodiment, the step S5 of acquiring the second risk index according to the rock formation fold information and the rock formation bending information comprises: S501, obtaining a rock formation fold wave crest and a rock formation fold wave trough according to the rock formation fold information, and obtaining a rock formation fold amplitude according to the rock formation fold wave crest and the rock formation fold wave trough; S502, calculating a rock formation fold amplitude difference according to the rock formation fold amplitude and a preset rock formation fold amplitude; S503, obtaining a rock formation bending angle according to the rock formation bending information, and calculating a rock formation bending angle difference according to the rock formation bending angle and a preset rock formation bending angle; S504: Perform weighted calculation on the rock fold amplitude difference and the rock stratum bending angle difference to obtain a second risk index.

[0032] As described in the above steps S501-S504, since rock stratum folds are a common phenomenon in geological structures and reflect the deformation of rock strata under long-term geological effects, the positions and height differences of wave crests and wave troughs can directly reflect the fold amplitude of rock strata, which is an important parameter for evaluating the degree of deformation of rock strata. The present invention first obtains rock stratum fold wave crests and rock stratum fold wave troughs according to the rock stratum fold information, and obtains rock stratum fold amplitudes according to the rock stratum fold wave crests and the rock stratum fold wave troughs. In this way, by accurately identifying wave crests and wave troughs, the system can provide accurate data support for subsequent fold amplitude calculations. Then, the rock stratum fold amplitude difference is calculated according to the rock stratum fold amplitude and the preset rock stratum fold amplitude. The rock stratum fold amplitude reflects the degree of deformation of the rock strata in the vertical direction. A larger fold amplitude means that the rock stratum has undergone significant compression or stretching, which may be accompanied by the accumulation of internal stress, increasing the risk of instability. At the same time, by calculating the fold amplitude difference, the degree of rock stratum deformation can be quantified and its changes relative to the normal state can be evaluated, which is helpful to identify abnormal The bending angle of the rock formation reflects the deformation of the rock formation in the horizontal direction, especially the change in the curvature of the rock formation. A larger bending angle means that the rock formation has undergone significant bending deformation, which may be accompanied by redistribution of internal stress, increasing the possibility of rock instability. Therefore, the bending angle of the rock formation is obtained according to the bending information of the rock formation. In this way, the subtle deformation of the rock formation can be captured by measuring the bending angle, which helps to identify potential high-risk areas. The bending angle difference of the rock formation is calculated according to the bending angle of the rock formation and the preset bending angle of the rock formation. By calculating the bending angle difference, the degree of bending deformation of the rock formation can be quantified and its impact on the stability of the rock formation can be evaluated. This helps to identify areas with abnormal deformation and provide early warning of potential geological risks. Finally, the difference in the amplitude of the rock formation fold and the difference in the bending angle of the rock formation are weighted to obtain a second risk index. In this way, the overall risk of the rock formation can be comprehensively evaluated through the second risk index, and a basis for the calculation of the subsequent comprehensive risk index can be provided.

[0033] In one embodiment, the step S6 of obtaining the stratum osmotic pressure according to the porosity and the permeability includes: S601, obtaining fluid viscosity; S602, calculating the pore fluid diffusion coefficient according to the porosity, the fluid viscosity and the permeability, wherein the calculation formula is: ; in, represents the pore fluid diffusion coefficient, represents the permeability, represents the porosity, Indicates fluid viscosity; The preset hydrostatic pressure of the rock is obtained, and the formation permeability pressure is calculated according to the ratio of the preset hydrostatic pressure to the diffusion coefficient of the pore fluid.

[0034] As described in the above steps S601-S602, the present application first obtains the fluid viscosity, and then calculates the porous fluid diffusion coefficient according to the porosity, the fluid viscosity and the permeability, wherein the fluid viscosity reflects the internal friction resistance of the fluid during flow, and the greater the viscosity, the worse the fluidity of the fluid, and the slower the diffusion rate of the fluid in the rock formation; conversely, the smaller the viscosity, the better the fluidity of the fluid and the faster the diffusion rate, and the porous fluid diffusion coefficient is one of the important parameters for calculating the osmotic pressure of the rock formation, and the change of osmotic pressure will affect the effective stress of the rock formation, reduce the bearing capacity of the rock formation, and increase the risk of instability. By calculating the diffusion coefficient, the fluid pressure distribution in the rock formation can be more accurately evaluated, and the potential geological disaster risk can be predicted. Then, the preset hydrostatic pressure of the rock is obtained, and the osmotic pressure of the rock formation is calculated according to the ratio of the preset hydrostatic pressure and the porous fluid diffusion coefficient, and the change of osmotic pressure will affect the effective stress of the rock formation, reduce the bearing capacity of the rock formation, and increase the risk of instability. Knowing the osmotic pressure in advance can help managers take preventive measures in advance to avoid major accidents.

[0035] In one embodiment, the step S7 of obtaining a comprehensive risk index according to the first risk index, the second risk index and the third risk index includes: S701. Obtain a corresponding first risk index weight value according to the first risk index; S702: Obtain a corresponding second risk index weight value according to the second risk index; S703: Calculate a comprehensive risk index according to the first risk index, the second risk index, the third risk index, the first risk index weight value, and the second risk index weight value, wherein the calculation formula is:

[0036] in, represents the comprehensive risk index, represents the first risk index, represents the second risk index, represents the third risk index, d represents the weight value of the first risk index, and e represents the weight value of the second risk index.

[0037] As described in the above steps S701-S703, the present application first obtains the corresponding first risk index weight value according to the first risk index, and then obtains the corresponding second risk index weight value according to the second risk index, and then calculates the comprehensive risk index according to the first risk index, the second risk index and the third risk index, the first risk index weight value and the second risk index weight value. The comprehensive risk index combines the first risk index, the second risk index and the third risk index to form a comprehensive risk assessment indicator. The comprehensive risk index not only takes into account the crack characteristics and deformation characteristics of the rock formation, but also takes into account the influence of fluid pressure, and can more comprehensively assess the overall risk of the rock formation. In this way, through the early judgment of the rock formation risk, high-risk areas can be discovered in time, early warnings can be given and corresponding protective measures can be taken to avoid the occurrence of geological disasters. Compared with traditional manual judgment, the automated risk assessment system can respond faster and more accurately, improve the safety of mining operations, and at the same time avoid the problem of mining accidents caused by identification time differences.

[0038] The present application also provides an image recognition system based on intelligent perception analysis, comprising: The first acquisition module 1 is used to acquire rock layer image information of the mine; A first extraction module 2 is used to extract features from the rock formation image information based on intelligent perception analysis to obtain crack features, rock formation deformation features and rock formation structure features; A second acquisition module 3 is used to acquire a crack width, a crack length and a crack density according to the crack characteristics, and to acquire a crack trend risk index according to the crack width and the crack length; A third acquisition module 4 is used to acquire a rock formation fragility index according to the fracture density, and to acquire a first risk index according to the rock formation fragility index and the fracture trend risk index; A fourth acquisition module 5 is used to acquire rock formation fold information and rock formation bending information according to the rock formation deformation characteristics, and acquire a second risk index according to the rock formation fold information and rock formation bending information; A fifth acquisition module 6, configured to acquire porosity and permeability according to the structural characteristics of the rock formation, and acquire rock formation osmotic pressure according to the porosity and the permeability, and use the rock formation osmotic pressure as a third risk index; A sixth acquisition module 7, configured to acquire a comprehensive risk index according to the first risk index, the second risk index and the third risk index; A first judgment module 8 is used to judge whether the comprehensive risk index is greater than a preset threshold; If the comprehensive risk index is greater than a preset threshold, it is determined that there is a rock formation risk in the area corresponding to the rock formation image information of the mine.

[0039] The present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the above-mentioned image recognition method based on intelligent perception analysis are implemented.

[0040] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media provided in this application and used in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0041] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also other elements not explicitly listed, or includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0042] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An image recognition method based on intelligent perception analysis, characterized in that: include: Obtain rock formation image information of the mine; Extracting features of the rock formation image information based on intelligent perception analysis to obtain crack features, rock formation deformation features, and rock formation structure features; Acquire crack width, crack length and crack density according to the crack characteristics, and acquire crack trend risk index according to the crack width and crack length; Obtaining a rock formation fragility index according to the fracture density, and obtaining a first risk index according to the rock formation fragility index and the fracture trend risk index; Acquire rock formation fold information and rock formation bending information according to the rock formation deformation characteristics, and acquire a second risk index according to the rock formation fold information and rock formation bending information; Obtaining porosity and permeability according to the structural characteristics of the rock formation, and obtaining rock formation osmotic pressure according to the porosity and the permeability, and using the rock formation osmotic pressure as a third risk index; Obtaining a comprehensive risk index according to the first risk index, the second risk index and the third risk index; Determining whether the comprehensive risk index is greater than a preset threshold; If the comprehensive risk index is greater than a preset threshold, it is determined that there is a rock formation risk in the area corresponding to the rock formation image information of the mine.

2. The image recognition method based on intelligent perception analysis according to claim 1 is characterized in that: The step of extracting features from the rock formation image information based on intelligent perception analysis to obtain crack features, rock formation deformation features, and rock formation structure features includes: Grayscale processing is performed on the rock formation image information based on intelligent perception analysis to obtain black-and-white rock formation image information, a plurality of black stripes are screened out according to the black-and-white surface image information, coordinates are respectively set at both ends of each individual black stripe in the plurality of black stripes to obtain a plurality of first coordinates and a plurality of second coordinates, and crack lengths of the plurality of black stripes are sequentially calculated according to the plurality of first coordinates and the plurality of second coordinates, wherein the intelligent perception analysis includes image preprocessing; Obtaining the maximum diameter of each individual black stripe among the plurality of black stripes, and taking the maximum diameter of the black stripe as the crack width; Obtain edge information of the rock layer image information of the mine based on an edge detection algorithm, and obtain the corresponding edge region area according to the edge information; Acquire the corresponding number of black stripes according to the plurality of black stripes, and calculate the crack density according to the number of black stripes and the area of ​​the edge region; The overall information corresponding to the crack length, crack width and crack density is used as crack characteristics; Acquire initial preset image information, and select the same point position of the initial preset image information and the rock layer image information of the mine; Acquire a first pixel point and a second pixel point at the same point, and acquire pixel displacement information according to the first pixel point and the second pixel point, and use the pixel displacement information as a rock formation deformation feature; Acquire multiple pixel pairs of rock formation image information, wherein the pixel pairs are adjacent pixel points; Based on the gray level co-occurrence matrix, multiple pixel pairs are traversed to obtain multiple adjacent pixel directions and multiple adjacent pixel distances, and the linear texture structure of the rock formation is obtained according to the multiple adjacent pixel directions and multiple adjacent pixel distances, and the linear texture structure of the rock formation is used as the rock formation structure feature.

3. The image recognition method based on intelligent perception analysis according to claim 1 is characterized in that: The step of obtaining the crack trend risk index according to the crack width and crack length includes: Acquire a first expansion rate within a preset time according to the crack width and a preset initial crack width; Acquire a second expansion rate within a preset time according to the crack length and a preset initial crack length; Performing weighted calculation on the first expansion rate and the second expansion rate to obtain a comprehensive crack expansion rate; Obtaining a rock formation change rate within a preset time according to the rock formation deformation characteristics; Obtaining the thickness of the rock layer, and obtaining a crack width ratio value according to the crack width and the rock layer thickness, and obtaining a corresponding first weight value according to the crack width ratio value; Obtain a fracture length ratio value according to the fracture length and the rock layer thickness, and a second weight value corresponding to the fracture length ratio value; Obtaining a corresponding third weight value according to the comprehensive crack expansion rate; The fracture trend risk index is calculated according to the fracture width proportion value, the fracture length proportion value, the fracture comprehensive expansion rate, the rock formation change rate, the first weight value, the second weight value and the third weight value, wherein the calculation formula is: in, Indicates the crack direction risk index, represents the crack width ratio, represents the crack length ratio, represents the comprehensive crack expansion rate, represents the rock formation change rate, a represents the first weight value, b represents the second weight value, and c represents the third weight value.

4. The image recognition method based on intelligent perception analysis according to claim 1 is characterized in that: The step of obtaining a rock formation fragility index according to the fracture density, and obtaining a first risk index according to the rock formation fragility index and the fracture trend risk index comprises: Obtaining a first area of ​​the rock layer in a preset area, dividing the first area into unit areas to obtain a plurality of unit areas; Acquire a plurality of corresponding crack densities according to the plurality of unit areas, and calculate an average crack density according to the plurality of crack densities; The standard crack density is calculated according to the plurality of crack densities and the average crack density, wherein the calculation formula is: ； in, represents the standard crack density, represents the lth crack density, represents the number of crack density, l represents the serial number of crack density, represents the average crack density; The crack density index is calculated according to the standard crack density and the average crack density, wherein the calculation formula is: ； in, represents the crack density index, represents the average crack density, represents the standard crack density; Calculating a density-area ratio value according to the ratio of the fracture density index to the first area, and using the density-area ratio value as a rock formation fragility index; The rock formation fragility index and the fracture trend risk index are weighted and calculated to obtain a first risk index.

5. The image recognition method based on intelligent perception analysis according to claim 1, characterized in that: The step of obtaining a second risk index according to the rock formation fold information and the rock formation bending information comprises: Acquire a rock formation fold wave crest and a rock formation fold wave trough according to the rock formation fold information, and acquire a rock formation fold amplitude according to the rock formation fold wave crest and the rock formation fold wave trough; Calculating a rock formation fold amplitude difference according to the rock formation fold amplitude and a preset rock formation fold amplitude; Acquire a rock formation bending angle according to the rock formation bending information, and calculate a rock formation bending angle difference according to the rock formation bending angle and a preset rock formation bending angle; The rock formation fold amplitude difference and the rock formation bending angle difference are weightedly calculated to obtain a second risk index.

6. The image recognition method based on intelligent perception analysis according to claim 1, characterized in that: The step of obtaining the stratum osmotic pressure according to the porosity and the permeability comprises: Get fluid viscosity; The pore fluid diffusion coefficient is calculated according to the porosity, the fluid viscosity and the permeability, wherein the calculation formula is: ； in, represents the pore fluid diffusion coefficient, represents the permeability, represents the porosity, Indicates fluid viscosity; The preset hydrostatic pressure of the rock is obtained, and the formation permeability pressure is calculated according to the ratio of the preset hydrostatic pressure to the diffusion coefficient of the pore fluid.

7. The image recognition method based on intelligent perception analysis according to claim 1, characterized in that: The step of obtaining a comprehensive risk index according to the first risk index, the second risk index and the third risk index comprises: Obtaining a corresponding first risk index weight value according to the first risk index; Obtaining a corresponding second risk index weight value according to the second risk index; The comprehensive risk index is calculated according to the first risk index, the second risk index and the third risk index, the first risk index weight value and the second risk index weight value, wherein the calculation formula is: in, represents the comprehensive risk index, represents the first risk index, represents the second risk index, represents the third risk index, d represents the weight value of the first risk index, and e represents the weight value of the second risk index.

8. An image recognition system based on intelligent perception analysis, characterized in that: include: A first acquisition module is used to acquire rock layer image information of the mine; A first extraction module is used to extract features from the rock formation image information based on intelligent perception analysis to obtain crack features, rock formation deformation features and rock formation structure features; A second acquisition module is used to acquire crack width, crack length and crack density according to the crack characteristics, and acquire a crack trend risk index according to the crack width and crack length; A third acquisition module is used to acquire a rock formation fragility index according to the fracture density, and to acquire a first risk index according to the rock formation fragility index and the fracture trend risk index; A fourth acquisition module, configured to acquire rock formation fold information and rock formation bending information according to the rock formation deformation characteristics, and to acquire a second risk index according to the rock formation fold information and rock formation bending information; a fifth acquisition module, configured to acquire porosity and permeability according to the structural characteristics of the rock formation, and acquire rock formation osmotic pressure according to the porosity and the permeability, and use the rock formation osmotic pressure as a third risk index; A sixth acquisition module, configured to acquire a comprehensive risk index according to the first risk index, the second risk index and the third risk index; A first judgment module is used to judge whether the comprehensive risk index is greater than a preset threshold; If the comprehensive risk index is greater than a preset threshold, it is determined that there is a rock formation risk in the area corresponding to the rock formation image information of the mine.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images