Automatic dangerous rock identification method and system based on machine vision system

Abstract

The invention provides an automatic dangerous rock identification method and system based on a machine vision system, and relates to the technical field of engineering mechanics, and the method comprises the steps: obtaining dangerous rock region image data containing disturbance amplitude and frequency, obtaining disturbance stress data, and carrying out the comprehensive analysis to calculate a rockburst failure load; the method comprises the following steps: establishing a relationship among a dynamic disturbance load, a stress gradient, a rockburst failure load and rockburst strength, calculating a stress-strength ratio and establishing a relationship between the stress-strength ratio and rockburst intensity, then extracting deformation and debris granularity characteristics, comprehensively generating a dangerous rock index, setting a safety threshold, and finally comparing the dangerous rock index with the safety threshold and sending out an early warning. According to the method, the image data, the high-resolution imaging equipment, the stress sensor and the numerical simulation method are utilized, automatic dangerous rock identification is achieved, the rockburst risk is effectively predicted, and regional safety is ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of engineering mechanics, and in particular to a dangerous rock automatic identification method and system based on a machine vision system. Background Art

[0002] Automatic identification of dangerous rocks is an emerging technology, which is mainly used in the fields of geological disaster prevention, mine safety, civil engineering and natural disaster monitoring. Based on the development of machine learning and deep learning algorithms, machine vision systems have made significant progress in rock identification, classification and feature extraction, and the application of high-resolution cameras, lidar and drones has further promoted the development of this field. The flexibility of drones and the combination of high-resolution cameras make remote monitoring possible. Dangerous rock identification is applied in many scenarios to ensure the safety of projects. For example, many mines have begun to deploy machine vision systems to monitor the geological structure changes of mines in real time, timely detect potential dangerous rocks, and prevent sudden accidents; monitoring equipment is arranged along highways and railways to monitor landslides and falling rocks to ensure transportation safety. The automatic identification technology of dangerous rocks based on machine vision systems has broad prospects. With the continuous advancement of technology and the deepening of application, it will play an increasingly important role in improving the ability to prevent and control geological disasters and ensuring the safety of production and life.

[0003] In the prior art, publication number CN115861296B discloses a method and system for automatically identifying dangerous rock masses on steep slopes based on UAV point clouds, which collects slope point cloud data and preprocesses it; calculates normal vectors and curvatures in turn based on the preprocessed point cloud data to obtain edge point clouds and clusters them to obtain edge point clouds and overall point clouds of single blocks; performs back wall plane fitting based on the least squares method on the edge point cloud of the single block to obtain a first result; performs overall volume calculation based on the Alphashape algorithm on the overall point cloud of the single block to obtain a second result; performs back wall plane boundary search calculation based on the Alphashape algorithm on the overall point cloud of the single block to obtain a third result; and performs a comprehensive assessment of the dangerous rock masses based on the first, second and third results obtained.

[0004] Insufficient existing technology: First, the preprocessing of point cloud data depends on specific algorithms, which may be sensitive to data noise and incompleteness, resulting in the accuracy of normal vector and curvature calculation being affected, thereby affecting the subsequent edge point cloud extraction and clustering effect. In addition, the plane fitting of the least squares method limits its scope of application due to its assumptions on point cloud distribution, and there may be deviations in the fitting of complex rock blocks.

[0005] Secondly, although the overall volume calculation and boundary search based on the AlphaShape algorithm provide valuable geometric information, the computational efficiency may be low when processing large-scale or highly irregular point clouds, resulting in insufficient real-time monitoring and assessment capabilities. In addition, the current technology lacks a comprehensive analysis of the mechanical properties of rock masses and fails to fully consider the influence of external environmental factors, which may lead to a decrease in the reliability of dangerous rock mass assessment. Overall, the robustness and adaptability of the algorithm need to be further improved to improve the accuracy and efficiency of identification.

[0006] Therefore, it is necessary to provide a dangerous rock automatic identification method and system based on a machine vision system to solve the above problem.

[0007] 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

[0008] The purpose of the present invention is to provide a method and system for automatically identifying dangerous rocks based on a machine vision system to solve the problems raised in the above background technology.

[0009] To achieve the above object, the present invention provides the following technical solutions: A method for automatically identifying dangerous rocks based on a machine vision system, the specific steps comprising: Step 1: Acquire image data of dangerous rock areas, wherein the image data includes disturbance amplitude and disturbance frequency; Step 2: Obtain the disturbance stress data of dangerous rocks, calculate the rockburst failure load by integrating the disturbance stress, disturbance amplitude and disturbance frequency, combine the dynamic disturbance load, stress gradient and rockburst failure load, and establish the quantitative relationship between the dynamic disturbance load, stress gradient, rockburst failure load and rockburst intensity; Step 3: Calculate the stress intensity ratio of dangerous rocks, combine stress gradient, dynamic disturbance load and stress intensity ratio, and establish a quantitative relationship between stress gradient, dynamic disturbance load, stress intensity ratio and rockburst intensity; Step 4: extract dangerous rock features using a feature extraction algorithm, wherein the dangerous rock features include deformation features and debris particle size features, and generate a dangerous rock index by combining the dangerous rock features, rock burst strength and rock burst severity; Step 5: Set up a safety threshold, compare the dangerous rock index with the safety threshold, and determine whether it exceeds the safety threshold. If so, an alarm will be issued immediately.

[0010] Furthermore, the image data of the dangerous rock area is obtained, and the disturbance stress, disturbance amplitude and disturbance frequency of the dangerous rock area are calculated according to the following method: Sensors and high-speed camera equipment, including drones and fixed monitoring equipment, are arranged in the dangerous rock area to continuously shoot and record image data of the dangerous rock area. The sensors record the vibration and displacement data of the dangerous rock area. The sensor data includes displacement sensors, accelerometers and strain gauges, which are used to measure the disturbance amplitude and disturbance frequency. The sensor data is fused with the image data to calculate the disturbance frequency. The formula is as follows: ; ; in, Describes the signal with a time delay of The degree of relevance, It is the expression of the signal in the frequency domain. The frequency corresponding to the detected peak is the disturbance frequency. , represents the image data signal, Indicates time delay, represents the frequency in the frequency domain, represents a complex exponential function; The disturbance amplitude is based on the acceleration data measured by the sensor. The displacement is obtained by integration using the acceleration data, and the amplitude is calculated by taking the maximum value. The formula is: ; in, represents the disturbance amplitude, Represents acceleration data, that is, the acceleration data is integrated once to obtain velocity data, and the velocity data is integrated once to obtain displacement data, and the maximum value of the displacement data is selected as the disturbance amplitude; Measure the initial stress of the dangerous rock, and calculate the disturbance stress based on the calculated disturbance amplitude and disturbance frequency. The formula is: ; in, represents the disturbance stress, represents the initial stress, Indicates the time that the disturbance stress acts on the dangerous rock.

[0011] Furthermore, the disturbance stress data of dangerous rocks are obtained, and the rockburst failure load is calculated by combining the disturbance stress, disturbance amplitude and disturbance frequency. The formula is as follows: ; in, represents the rockburst failure load, , , They represent the disturbance stress in The horizontal axis, Vertical axis, The disturbance stress component in the vertical coordinate direction, , , Respectively represent the disturbance amplitude in The horizontal axis, Vertical axis, The disturbance amplitude component in the vertical coordinate direction, , , Represent the disturbance frequency The horizontal axis, Vertical axis, The disturbance frequency component in the vertical coordinate direction, is a constant coefficient used to adjust the ratio between the terms to match the actual failure load.

[0012] Furthermore, the rockburst intensity is generated and the quantitative relationship between the dynamic disturbance load, stress gradient, rockburst failure load and rockburst intensity is established based on the following method: The dynamic disturbance load is calculated based on the vibration acceleration and the mass of the rock mass, while the stress gradient is calculated based on the spatial rate of change of the rock mass stress distribution, based on the formula: ; ; in, represents the dynamic disturbance load, Indicates the quality of the rock mass, represents the vibration acceleration, represents the stress gradient, represents the stress change, Indicates the change in spatial position; The dynamic disturbance load, stress gradient and rock burst failure load are combined to generate rock burst strength and establish a quantitative relationship based on the formula: ; in, represents the intensity of rock burst, represents the absolute value of the stress gradient, , , They represent the weight coefficients of the corresponding parameters, which are used to characterize the influence weight of each parameter on the rockburst intensity, and .

[0013] Furthermore, the stress intensity ratio of dangerous rock is calculated based on the following method: The stress gradient of the tangential stress is quantified based on the average change rate of the tangential stress inside the dangerous rock, and the maximum tangential stress of the dangerous rock mass under the stress state in three-dimensional space is calculated. According to the stress condition of the dangerous rock mass, the compressive strength or tensile strength is selected. If the rock mass is in a compressive stress state, the compressive strength of the rock mass should be selected; if the rock mass is in a tensile stress state, the tensile strength of the rock mass should be selected. The formula is: ; ; in, is the stress intensity ratio of dangerous rock, represents the maximum tangential stress, Indicates the compressive or tensile strength of rock materials. The strength type is selected according to the specific conditions of the rock mass. Represents the maximum principal stress in three-dimensional space, as the maximum tangential stress.

[0014] Furthermore, the stress gradient, dynamic disturbance load and stress intensity ratio are combined to establish a quantitative relationship between stress gradient, dynamic disturbance load, stress intensity ratio and rockburst intensity, based on the following formula: ; in, Indicates the intensity of rock burst. Indicates the compressive or tensile strength of rock materials. The strength type is selected according to the specific conditions of the rock mass. represents the static load, i.e. the force on the rock mass under static conditions, represents the dangerous rock strength ratio, , , represents the weight coefficient of each corresponding parameter, which is used to characterize the influence weight of each parameter on the rockburst intensity, and .

[0015] Furthermore, the extracted deformation characteristics and debris particle size characteristics are used to generate the dangerous rock index by combining the dangerous rock characteristics, rock burst strength and rock burst severity. The method is as follows: The image and sensor data of the dangerous rock are analyzed using the feature extraction algorithm to extract the key features of the dangerous rock, including deformation features and debris particle size features. The average crack width, length and average expansion speed data of the dangerous rock are obtained to calculate the deformation features. The formula is as follows: ; ; ; ; in, , , They represent the current average crack width, current average crack length, and average expansion speed of dangerous rocks, respectively. Indicates The width of the crack, Indicates The length of the crack, Indicates at the current moment The length of the crack, Indicates that at the previous moment The length of the crack, Before Time to present The time interval between moments, represents the number of cracks, is the index of the crack, Indicates deformation characteristics; The debris area and distribution density are obtained to calculate the debris particle size characteristics, based on the formula: ; ; ; in, , represent the average area and average density of all debris in the observation area, respectively. For the The area of ​​debris, represents the number of debris detected in the observation area, is the observation area, Indicates the particle size characteristics of debris; The calculated deformation characteristics and debris particle size characteristics are used to generate the dangerous rock index by combining the rock burst intensity and rock burst severity. The formula is as follows: ; in, represents the dangerous rock index, , represents the weight coefficient corresponding to each parameter combination, and , represents the intensity of rock burst, Indicates the intensity of rockburst.

[0016] Furthermore, the dangerous rock index is compared with the safety threshold to determine whether it exceeds the safety threshold. The logic formula is as follows: ; in, Indicates the logical value of judging whether the dangerous rock index exceeds the safety threshold. When , it means that the dangerous rock index exceeds the established safety threshold and an early warning signal is issued; when , it means that the dangerous rock index does not exceed the established safety threshold.

[0017] The present invention further provides a dangerous rock automatic identification system based on a machine vision system, the automatic identification system is used to execute the above-mentioned dangerous rock automatic identification method based on a machine vision system, comprising: A data acquisition module, the data acquisition module is used to obtain image data of dangerous rock areas, the image data includes disturbance amplitude and disturbance frequency; A failure load calculation module is used to obtain disturbance stress data of dangerous rocks, calculate rock burst failure load by integrating disturbance stress, disturbance amplitude and disturbance frequency, combine dynamic disturbance load, stress gradient and rock burst failure load, and establish a quantitative relationship between dynamic disturbance load, stress gradient, rock burst failure load and rock burst strength; A stress intensity ratio evaluation module, which is used to calculate the stress intensity ratio of dangerous rocks, combine stress gradient, dynamic disturbance load and stress intensity ratio, and establish a quantitative relationship between stress gradient, dynamic disturbance load, stress intensity ratio and rock burst intensity; A feature extraction module, wherein the feature extraction module uses a feature extraction algorithm to extract dangerous rock features, wherein the dangerous rock features include deformation features and debris particle size features, and generates a dangerous rock index by combining the dangerous rock features, rock burst strength and rock burst severity; The safety monitoring and early warning module is used to set a safety threshold, compare the dangerous rock index with the safety threshold, and determine whether it exceeds the safety threshold. If so, an alarm is immediately issued.

[0018] Compared with the prior art, the present invention has the following beneficial effects: The present invention can obtain more comprehensive images and disturbance data of dangerous rock areas by utilizing high-speed camera equipment and the joint deployment of multiple sensors, thereby improving the measurement accuracy of disturbance amplitude and frequency. This multi-source data fusion method can effectively reduce the impact of noise and ensure the reliability of subsequent calculations; Secondly, the present invention establishes multiple quantitative relationships, comprehensively considering dynamic disturbance loads, stress gradients and rockburst damage loads, and improving the accuracy of rockburst strength and intensity assessment. In addition, the feature extraction module introduces the analysis of deformation characteristics and debris particle size characteristics, providing more detailed identification of dangerous rock features, and then generating a dangerous rock index. This series of improvements enables the present invention to not only more accurately assess the stability of the rock mass, but also achieve real-time monitoring and early warning, effectively improving the safety and efficiency of dangerous rock management; The present invention provides an efficient and accurate method for automatic identification of dangerous rocks by comprehensively utilizing machine vision and multiple sensor data, realizes real-time monitoring and early warning, and significantly improves the accuracy of rock stability assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 It is a schematic diagram of the overall method flow of the present invention.

[0020] Figure 2 It is a schematic diagram of the system module flow of the present invention. DETAILED DESCRIPTION

[0021] 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.

[0022] 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.

[0023] Example: See also Figure 1 , a dangerous rock automatic identification method based on a machine vision system, the specific steps include: Step 1: Acquire image data of dangerous rock areas, wherein the image data includes disturbance amplitude and disturbance frequency; Step 2: Obtain the disturbance stress data of dangerous rocks, calculate the rockburst failure load by integrating the disturbance stress, disturbance amplitude and disturbance frequency, combine the dynamic disturbance load, stress gradient and rockburst failure load, and establish the quantitative relationship between the dynamic disturbance load, stress gradient, rockburst failure load and rockburst intensity; Step 3: Calculate the stress intensity ratio of dangerous rocks, combine stress gradient, dynamic disturbance load and stress intensity ratio, and establish a quantitative relationship between stress gradient, dynamic disturbance load, stress intensity ratio and rockburst intensity; Step 4: extract dangerous rock features using a feature extraction algorithm, wherein the dangerous rock features include deformation features and debris particle size features, and generate a dangerous rock index by combining the dangerous rock features, rock burst strength and rock burst severity; Step 5: Set up a safety threshold, compare the dangerous rock index with the safety threshold, and determine whether it exceeds the safety threshold. If so, an alarm will be issued immediately.

[0024] It should be noted that through image acquisition from drones and fixed monitoring equipment combined with sensor data such as displacement, acceleration and strain, we can fully understand the dynamic changes of dangerous rocks, provide multi-dimensional information, and use autocorrelation functions to analyze disturbance frequencies, identify the vibration mode of the rock mass, and help judge its stability, which is crucial to the accuracy of the early warning system.

[0025] Therefore, it is necessary to obtain image data of dangerous rock areas and calculate the disturbance stress, disturbance amplitude and disturbance frequency of dangerous rock areas. The method is as follows: Sensors and high-speed camera equipment, including drones and fixed monitoring equipment, are arranged in the dangerous rock area to continuously shoot and record image data of the dangerous rock area. The sensors record the vibration and displacement data of the dangerous rock area. The sensor data includes displacement sensors, accelerometers and strain gauges, which are used to measure the disturbance amplitude and disturbance frequency. The sensor data is fused with the image data to calculate the disturbance frequency. The formula is as follows: ; ; in, Describes the signal with a time delay of The degree of relevance, It is the expression of the signal in the frequency domain. The frequency corresponding to the detected peak is the disturbance frequency. , represents the image data signal, Indicates time delay, represents the frequency in the frequency domain, represents a complex exponential function; The disturbance amplitude is based on the acceleration data measured by the sensor. The displacement is obtained by integration using the acceleration data, and the amplitude is calculated by taking the maximum value. The formula is: ; in, represents the disturbance amplitude, Represents acceleration data, that is, the acceleration data is integrated once to obtain velocity data, and the velocity data is integrated once to obtain displacement data, and the maximum value of the displacement data is selected as the disturbance amplitude; Measure the initial stress of the dangerous rock, and calculate the disturbance stress based on the calculated disturbance amplitude and disturbance frequency. The formula is: ; in, represents the disturbance stress, represents the initial stress, Indicates the time that the disturbance stress acts on the dangerous rock.

[0026] It should be noted that by considering the disturbance stress data in three directions at the same time, we can fully understand the stress state of the rock mass in different directions. The calculated failure load is directly related to the stability of the rock mass, and can quantify the risk of rock burst under specific stress conditions, providing a basis for subsequent preventive measures. In addition, the use of autocorrelation function can effectively extract the vibration characteristics in different directions, reflecting the dynamic response of the rock mass under disturbance.

[0027] Therefore, it is necessary to obtain the disturbance stress data of dangerous rocks, and calculate the rock burst failure load by combining the disturbance stress, disturbance amplitude and disturbance frequency. The formula is as follows: ; in, represents the rockburst failure load, , , They represent the disturbance stress in The horizontal axis, Vertical axis, The disturbance stress component in the vertical coordinate direction, , , Respectively represent the disturbance amplitude in The horizontal axis, Vertical axis, The disturbance amplitude component in the vertical coordinate direction, , , Represent the disturbance frequency The horizontal axis, Vertical axis, The disturbance frequency component in the vertical coordinate direction, is a constant coefficient used to adjust the ratio between the terms to match the actual failure load.

[0028] It should be noted that establishing a quantitative relationship can comprehensively reflect the main variables affecting rockburst intensity, provide a more accurate assessment, and help engineers identify potential rockburst risks in advance, providing a scientific basis for formulating protective measures. By analyzing dynamic disturbance loads and stress gradients, it is possible to understand the response mechanism of rock mass under dynamic conditions, providing basic data for subsequent research.

[0029] Therefore, it is necessary to generate rockburst strength and establish the quantitative relationship between dynamic disturbance load, stress gradient, rockburst failure load and rockburst strength. The method is based on: The dynamic disturbance load is calculated based on the vibration acceleration and the mass of the rock mass, while the stress gradient is calculated based on the spatial rate of change of the rock mass stress distribution, based on the formula: ; ; in, represents the dynamic disturbance load, Indicates the quality of the rock mass, represents the vibration acceleration, represents the stress gradient, represents the stress change, Indicates the change in spatial position; The dynamic disturbance load, stress gradient and rock burst failure load are combined to generate rock burst strength and establish a quantitative relationship based on the formula: ; in, represents the intensity of rock burst, represents the absolute value of the stress gradient, , , They represent the weight coefficients of the corresponding parameters, which are used to characterize the influence weight of each parameter on the rockburst intensity, and ; In the above formula, the weight coefficient is set to This is because the rockburst failure load is directly related to the critical condition of rock mass failure. Rockburst usually occurs when the rock mass is subjected to a load that exceeds its bearing capacity. Therefore, the impact of the failure load is the most significant. Once the load borne by the rock mass exceeds this critical value, it may cause instantaneous and violent destruction, which is highly destructive and irreversible. Therefore, the maximum weight coefficient corresponding to the rockburst failure load is given. The dynamic disturbance load reflects the immediate impact of external vibration or dynamic conditions on the rock mass. Although it is less important than the destructive load, under dynamic conditions, the enhanced vibration can significantly increase the stress state of the rock mass, thereby causing rock bursts. In many cases, especially in dynamic environments such as mining or earthquakes, changes in dynamic loads will directly affect the stability of the rock mass and need to be given a relatively high weight. ; The stress gradient reflects the change of stress in space, and its influence on rock burst is relatively indirect. It is usually a problem that may be caused by local stress concentration under other conditions. Although the stress gradient is important, its direct influence on rock burst strength is often lower than that of dynamic disturbance and destructive load. Therefore, it is given the lowest weight coefficient in weight setting. .

[0030] It should be noted that the calculation of dangerous rock stress intensity is because it can effectively evaluate the stability of rock mass under specific stress conditions, help predict potential rock burst risks, and accurately reflect the stress conditions of rock mass in three-dimensional space by using the maximum tangential stress and the compressive or tensile strength of rock mass. The appropriate strength type can be selected according to the actual stress, thereby ensuring the reliability and practicality of the evaluation results. This method provides a scientific basis for engineering practice and helps to formulate corresponding safety measures and design optimization.

[0031] Therefore, it is necessary to calculate the stress intensity ratio of dangerous rock, based on the following method: The stress gradient of the tangential stress is quantified based on the average change rate of the tangential stress inside the dangerous rock, and the maximum tangential stress of the dangerous rock mass under the stress state in three-dimensional space is calculated. According to the stress condition of the dangerous rock mass, the compressive strength or tensile strength is selected. If the rock mass is in a compressive stress state, the compressive strength of the rock mass should be selected; if the rock mass is in a tensile stress state, the tensile strength of the rock mass should be selected. The formula is: ; ; in, is the stress intensity ratio of dangerous rock, represents the maximum tangential stress, Indicates the compressive or tensile strength of rock materials. The strength type is selected according to the specific conditions of the rock mass. Represents the maximum principal stress in three-dimensional space, as the maximum tangential stress.

[0032] It should be noted that in the calculation The formula can comprehensively consider a variety of influencing factors, thereby more comprehensively evaluating the risk of rockburst, and by setting weight coefficients, it can emphasize the relative importance of each parameter to the intensity of rockburst, ensuring accurate prediction of rock mass stability in a dynamic environment. This method not only improves the accuracy of risk assessment, but also provides a practical early warning mechanism and decision-making basis for actual projects.

[0033] Therefore, it is necessary to combine stress gradient, dynamic disturbance load and stress intensity ratio to establish a quantitative relationship between stress gradient, dynamic disturbance load, stress intensity ratio and rockburst intensity. The formula is as follows: ; in, Indicates the intensity of rock burst. Indicates the compressive or tensile strength of rock materials. The strength type is selected according to the specific conditions of the rock mass. represents the static load, i.e. the force on the rock mass under static conditions, represents the dangerous rock strength ratio, , , represents the weight coefficient of each corresponding parameter, which is used to characterize the influence weight of each parameter on the rockburst intensity, and ; In the above formula, the reason why the weight coefficients are arranged in order is Because the strength of dangerous rock is The impact on rockburst intensity is the most significant. The dangerous rock strength ratio directly reflects the safety of the rock mass when it is subjected to external loads. It indicates that the rock mass has a stronger ability to resist damage, thus reducing the possibility of rock burst, so it is given the highest weight. ; Dynamic disturbance load Although important, it is less important than the inherent strength of the rock mass and is therefore given a medium weight. ; Finally, the stress gradient It also plays a role in the formation of rockbursts, but mainly affects the intensity of rockbursts, and has little effect on the severity of rockbursts, so it is given the lowest weight. .

[0034] It should be noted that by comprehensively analyzing deformation characteristics and debris particle size characteristics, the stability and potential risks of dangerous rocks can be more comprehensively evaluated. By using extracted features such as average crack width, length and expansion speed and the area and density of debris, the microscopic changes of the rock mass and its response to external loads can be accurately reflected, thereby predicting the probability of rock bursts and setting equal weight coefficients. The reason is that the intensity of rock burst and rockburst intensity They are considered equally important in influencing the dangerous rock index, reflecting the importance of both to the stability of dangerous rock. Through this comprehensive approach, it is possible to ensure that the model not only considers local deformation, but also takes into account the dynamic response of the entire rock mass, ultimately providing a more accurate basis for the risk assessment of dangerous rock.

[0035] Therefore, it is necessary to use the extracted deformation characteristics and debris particle size characteristics, and integrate the dangerous rock characteristics, rock burst strength and rock burst intensity to generate a dangerous rock index. The method is based on: The image and sensor data of the dangerous rock are analyzed using the feature extraction algorithm to extract the key features of the dangerous rock, including deformation features and debris particle size features. The average crack width, length and average expansion speed data of the dangerous rock are obtained to calculate the deformation features. The formula is as follows: ; ; ; ; in, , , They represent the current average crack width, current average crack length, and average expansion speed of dangerous rocks, respectively. Indicates The width of the crack, Indicates The length of the crack, Indicates at the current moment The length of the crack, Indicates that at the previous moment The length of the crack, Before Time to present The time interval between moments, represents the number of cracks, is the index of the crack, Indicates deformation characteristics; The debris area and distribution density are obtained to calculate the debris particle size characteristics, based on the formula: ; ; ; in, , represent the average area and average density of all debris in the observation area, respectively. For the The area of ​​debris, represents the number of debris detected in the observation area, is the observation area, Indicates the particle size characteristics of debris; The calculated deformation characteristics and debris particle size characteristics are used to generate the dangerous rock index by combining the rock burst intensity and rock burst severity. The formula is as follows: ; in, represents the dangerous rock index, , represents the weight coefficient corresponding to each parameter combination, and , represents the intensity of rock burst, Indicates the intensity of rockburst.

[0036] It should be noted that comparing the dangerous rock index with the safety threshold is an important step to ensure engineering safety. By judging the logical value, it is possible to quickly identify whether there is a potential risk in dangerous rocks, so that early warning signals can be issued in time. This early warning mechanism is crucial for preventing emergencies such as rock bursts, helping to protect personnel safety and reduce property losses, while providing a scientific basis for subsequent monitoring and emergency measures.

[0037] Therefore, it is necessary to compare the dangerous rock index with the safety threshold to determine whether it exceeds the safety threshold. The logical formula is: ; in, Indicates the logical value of judging whether the dangerous rock index exceeds the safety threshold. When , it means that the dangerous rock index exceeds the established safety threshold and an early warning signal is issued; when , it means that the dangerous rock index does not exceed the established safety threshold.

[0038] See also Figure 2 The present invention further provides a dangerous rock automatic identification system based on a machine vision system, the automatic identification system is used to execute the above-mentioned dangerous rock automatic identification method based on a machine vision system, comprising: A data acquisition module, the data acquisition module is used to obtain image data of dangerous rock areas, the image data includes disturbance amplitude and disturbance frequency; A failure load calculation module is used to obtain disturbance stress data of dangerous rocks, calculate rock burst failure load by integrating disturbance stress, disturbance amplitude and disturbance frequency, combine dynamic disturbance load, stress gradient and rock burst failure load, and establish a quantitative relationship between dynamic disturbance load, stress gradient, rock burst failure load and rock burst strength; A stress intensity ratio evaluation module, which is used to calculate the stress intensity ratio of dangerous rocks, combine stress gradient, dynamic disturbance load and stress intensity ratio, and establish a quantitative relationship between stress gradient, dynamic disturbance load, stress intensity ratio and rock burst intensity; A feature extraction module, wherein the feature extraction module uses a feature extraction algorithm to extract dangerous rock features, wherein the dangerous rock features include deformation features and debris particle size features, and generates a dangerous rock index by combining the dangerous rock features, rock burst strength and rock burst severity; The safety monitoring and early warning module is used to set a safety threshold, compare the dangerous rock index with the safety threshold, and determine whether it exceeds the safety threshold. If so, an alarm is immediately issued.

[0039] 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.

[0040] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by 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 electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0041] 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.

[0042] 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 automatically identifying dangerous rocks based on a machine vision system, characterized in that: The specific steps include: Step 1: Acquire image data of dangerous rock areas, wherein the image data includes disturbance amplitude and disturbance frequency; Step 2: Obtain the disturbance stress data of dangerous rocks, calculate the rockburst failure load by integrating the disturbance stress, disturbance amplitude and disturbance frequency, combine the dynamic disturbance load, stress gradient and rockburst failure load, and establish the quantitative relationship between the dynamic disturbance load, stress gradient, rockburst failure load and rockburst intensity; Step 3: Calculate the stress intensity ratio of dangerous rocks, combine stress gradient, dynamic disturbance load and stress intensity ratio, and establish a quantitative relationship between stress gradient, dynamic disturbance load, stress intensity ratio and rockburst intensity; Step 4: extract dangerous rock features using a feature extraction algorithm, wherein the dangerous rock features include deformation features and debris particle size features, and generate a dangerous rock index by combining the dangerous rock features, rock burst strength and rock burst severity; Step 5: Set up a safety threshold, compare the dangerous rock index with the safety threshold, and determine whether it exceeds the safety threshold. If so, an alarm will be issued immediately.

2. The method for automatically identifying dangerous rocks based on a machine vision system according to claim 1, characterized in that: The image data of the dangerous rock area is obtained, and the disturbance stress, disturbance amplitude and disturbance frequency of the dangerous rock area are calculated according to the following method: Sensors and high-speed camera equipment, including drones and fixed monitoring equipment, are arranged in the dangerous rock area to continuously shoot and record image data of the dangerous rock area. The sensors record the vibration and displacement data of the dangerous rock area. The sensor data includes displacement sensors, accelerometers and strain gauges, which are used to measure the disturbance amplitude and disturbance frequency. The sensor data is fused with the image data to calculate the disturbance frequency. The formula is as follows: ； ； in, Describes the signal with a time delay of The degree of relevance when It is the expression of the signal in the frequency domain. The frequency corresponding to the detected peak is the disturbance frequency. , represents the image data signal, Indicates time delay, represents the frequency in the frequency domain, represents a complex exponential function; The disturbance amplitude is based on the acceleration data measured by the sensor. The displacement is obtained by integration using the acceleration data, and the amplitude is calculated by taking the maximum value. The formula is: ； in, represents the disturbance amplitude, Represents acceleration data, that is, the acceleration data is integrated once to obtain velocity data, and the velocity data is integrated once to obtain displacement data, and the maximum value of the displacement data is selected as the disturbance amplitude; Measure the initial stress of the dangerous rock, and calculate the disturbance stress based on the calculated disturbance amplitude and disturbance frequency. The formula is: ； in, represents the disturbance stress, represents the initial stress, Indicates the time that the disturbance stress acts on the dangerous rock.

3. The method for automatically identifying dangerous rocks based on a machine vision system according to claim 2 is characterized in that: Obtain the disturbance stress data of dangerous rocks, and calculate the rockburst failure load by combining the disturbance stress, disturbance amplitude and disturbance frequency. The formula is: ； in, represents the rockburst failure load, , , They represent the disturbance stress in The horizontal axis, Vertical axis, The disturbance stress component in the vertical coordinate direction, , , Respectively represent the disturbance amplitude in The horizontal axis, Vertical axis, The disturbance amplitude component in the vertical coordinate direction, , , Represents the disturbance frequency The horizontal axis, Vertical axis, The disturbance frequency component in the vertical coordinate direction, is a constant coefficient used to adjust the ratio between the terms to match the actual failure load.

4. The method for automatically identifying dangerous rocks based on a machine vision system according to claim 3 is characterized in that: Generate rockburst strength and establish the quantitative relationship between dynamic disturbance load, stress gradient, rockburst failure load and rockburst strength based on the following method: The dynamic disturbance load is calculated based on the vibration acceleration and the mass of the rock mass, while the stress gradient is calculated based on the spatial rate of change of the rock mass stress distribution, based on the formula: ； ； in, represents the dynamic disturbance load, Indicates the quality of the rock mass, represents the vibration acceleration, represents the stress gradient, represents the stress change, Indicates the change in spatial position; The dynamic disturbance load, stress gradient and rock burst failure load are combined to generate rock burst strength and establish a quantitative relationship based on the formula: ； in, represents the intensity of rock burst, represents the absolute value of the stress gradient, , , They represent the weight coefficients of the corresponding parameters, which are used to characterize the influence weight of each parameter on the rockburst intensity, and .

5. The method for automatically identifying dangerous rocks based on a machine vision system according to claim 1, characterized in that: The method for calculating the stress intensity ratio of dangerous rock is based on: The stress gradient of the tangential stress is quantified based on the average change rate of the tangential stress inside the dangerous rock, and the maximum tangential stress of the dangerous rock mass under the stress state in three-dimensional space is calculated. According to the stress condition of the dangerous rock mass, the compressive strength or tensile strength is selected. If the rock mass is in a compressive stress state, the compressive strength of the rock mass should be selected; if the rock mass is in a tensile stress state, the tensile strength of the rock mass should be selected. The formula is: ； ； in, is the stress intensity ratio of dangerous rock, represents the maximum tangential stress, Indicates the compressive or tensile strength of rock materials. The strength type is selected according to the specific conditions of the rock mass. Represents the maximum principal stress in three-dimensional space, as the maximum tangential stress.

6. The method for automatically identifying dangerous rocks based on a machine vision system according to claim 4, characterized in that: Combining stress gradient, dynamic disturbance load and stress intensity ratio, a quantitative relationship between stress gradient, dynamic disturbance load, stress intensity ratio and rockburst intensity is established, based on the following formula: ； in, Indicates the intensity of rock burst. Indicates the compressive or tensile strength of rock materials. The strength type is selected according to the specific conditions of the rock mass. represents the static load, i.e. the force on the rock mass under static conditions, represents the dangerous rock strength ratio, , , represents the weight coefficient of each corresponding parameter, which is used to characterize the influence weight of each parameter on the rockburst intensity, and .

7. The method for automatically identifying dangerous rocks based on a machine vision system according to claim 1, characterized in that: The dangerous rock index is generated by using the extracted deformation characteristics and debris particle size characteristics, and combining the dangerous rock characteristics, rock burst strength and rock burst intensity. The method is based on: The image and sensor data of the dangerous rock are analyzed using the feature extraction algorithm to extract the key features of the dangerous rock, including deformation features and debris particle size features. The average crack width, length and average expansion speed data of the dangerous rock are obtained to calculate the deformation features. The formula is as follows: ； ； ； ； in, , , They represent the current average crack width, current average crack length, and average expansion speed of dangerous rocks, respectively. Indicates The width of the crack, Indicates The length of the crack, Indicates at the current moment The length of the crack, Indicates that at the previous moment The length of the crack, Before Time to present The time interval between moments, represents the number of cracks, is the index of the crack, Indicates deformation characteristics; The debris area and distribution density are obtained to calculate the debris particle size characteristics, based on the formula: ； ； ； in, , represent the average area and average density of all debris in the observation area, respectively. For the The area of ​​debris, represents the number of debris detected in the observation area, is the observation area, Indicates the particle size characteristics of debris; The calculated deformation characteristics and debris particle size characteristics are used to generate the dangerous rock index by combining the rock burst strength and rock burst intensity. The formula is as follows: ； in, represents the dangerous rock index, , represents the weight coefficient corresponding to each parameter combination, and , represents the intensity of rock burst, Indicates the intensity of rockburst.

8. The method for automatically identifying dangerous rocks based on a machine vision system according to claim 7, characterized in that: Compare the dangerous rock index with the safety threshold to determine whether it exceeds the safety threshold. The logical formula is: ； in, Indicates the logical value of judging whether the dangerous rock index exceeds the safety threshold. When , it means that the dangerous rock index exceeds the established safety threshold and an early warning signal is issued; when , it means that the dangerous rock index does not exceed the established safety threshold.

9. An automatic identification system for dangerous rocks based on a machine vision system, characterized in that: The automatic identification system is used to execute the dangerous rock automatic identification method based on a machine vision system according to any one of claims 1 to 8, comprising: A data acquisition module, the data acquisition module is used to obtain image data of dangerous rock areas, the image data includes disturbance amplitude and disturbance frequency; A failure load calculation module is used to obtain disturbance stress data of dangerous rocks, calculate rock burst failure load by integrating disturbance stress, disturbance amplitude and disturbance frequency, combine dynamic disturbance load, stress gradient and rock burst failure load, and establish a quantitative relationship between dynamic disturbance load, stress gradient, rock burst failure load and rock burst strength; A stress intensity ratio evaluation module, which is used to calculate the stress intensity ratio of dangerous rocks, combine stress gradient, dynamic disturbance load and stress intensity ratio, and establish a quantitative relationship between stress gradient, dynamic disturbance load, stress intensity ratio and rock burst intensity; A feature extraction module, wherein the feature extraction module uses a feature extraction algorithm to extract dangerous rock features, wherein the dangerous rock features include deformation features and debris particle size features, and generates a dangerous rock index by combining the dangerous rock features, rock burst strength and rock burst severity; The safety monitoring and early warning module is used to set a safety threshold, compare the dangerous rock index with the safety threshold, and determine whether it exceeds the safety threshold. If so, an alarm is immediately issued.

Images