A mine safety monitoring method and system for an open pit mining area

Abstract

The present application relates to the technical field of image processing, in particular to a kind of open pit mining area's mine safety monitoring method and system.The present application is first by constructing different sliding window on wedge-shaped slope image sliding, using similarity evaluation index extracts periodic characteristics of slope surface, overall assesses the similarity of each region of image;Second, based on similarity evaluation index determines the period of image in vertical and horizontal direction, and this optimizes the block grid parameter, so that algorithm processing scale and image itself periodic characteristics match, to enhance the pertinence and detail performance of image enhancement;Finally, using the image after enhancement calculates crack risk evaluation index, significantly improves the accuracy and reliability of crack risk evaluation, enhances the precision and robustness of slope landslide disaster monitoring.

Description

Technical Field

[0001] This invention relates to the field of image processing technology, specifically to a method and system for monitoring mine safety in open-pit mining areas. Background Technology

[0002] Landslides occur along with the formation of cracked areas on slopes, with wedge-shaped slopes being the most common type of slope in open-pit mining areas. The formation mechanism of wedge-shaped slopes is due to the mutual cutting of two or more rock strata, resulting in cracks in the cutting area. As the cutting surfaces move, the cracks widen, eventually leading to wedge-shaped slope landslides. Therefore, monitoring cracked areas is equivalent to monitoring landslide hazards.

[0003] In acquiring wedge-shaped slope images, considering factors such as localized areas being too dark or too bright, and details being difficult to discern, the CLAHE algorithm can be used to enhance the wedge-shaped slope images before using existing technologies for slope landslide monitoring. Based on this enhanced image, subsequent crack risk assessment can be performed. However, in real-world mining areas, the surface of wedge-shaped slopes is complex and rich in detail, containing various real-world factors such as rock textures, fine cracks, and turf. Therefore, using only empirical parameters in the CLAHE algorithm cannot achieve the optimal image enhancement effect for wedge-shaped slope images, thus affecting the accuracy and stability of crack area analysis. Summary of the Invention

[0004] To address the technical problem that existing technologies using the empirical parameter-based CLAHE algorithm cannot achieve optimal image enhancement, thus affecting the accurate monitoring of crack areas, the present invention aims to provide a mine safety monitoring method and system for open-pit mining areas. The specific technical solution adopted is as follows:

[0005] This invention proposes a method for monitoring mine safety in open-pit mining areas, the method comprising:

[0006] Images of wedge-shaped slopes in the mining area are acquired, and sliding windows are constructed in the vertical and horizontal directions respectively. Based on the pixel grayscale difference and structural similarity between the images in the sliding windows before and after sliding in the vertical and horizontal directions, similarity evaluation indicators under different sliding conditions in the vertical and horizontal directions are obtained.

[0007] The similarity evaluation index under different sliding conditions in the vertical and horizontal directions is curve-fitted with the sliding amount under different sliding conditions in the corresponding directions to obtain the extreme points on the fitted curve; the periodicity analysis of the wedge slope image is performed based on the position of the extreme points to obtain the periodicity of the wedge slope image in the vertical and horizontal directions.

[0008] Based on the period of the wedge slope image in the vertical and horizontal directions, the block grid size parameters of the CLAHE algorithm in the vertical and horizontal directions are optimized, and image enhancement is performed to obtain an enhanced wedge slope image.

[0009] Obtain crack risk assessment indicators for cracked areas in enhanced wedge slope images, and monitor landslide hazards based on crack risk assessment indicators.

[0010] Furthermore, the method for obtaining the pixel grayscale difference between the images within the sliding window before and after different sliding in the vertical and horizontal directions includes:

[0011] Slide a sliding window in any direction on the wedge slope image along the corresponding direction. Compare the absolute value of the difference between the gray value of each pixel in the image at each position in all sliding windows after sliding and the corresponding pixel in the initial sliding window. Determine the absolute value of the gray value difference. Take the average value of the absolute value of the gray value difference in each sliding window as the pixel gray value difference under that sliding. Obtain the pixel gray value difference under all sliding in the corresponding direction.

[0012] Furthermore, the method for obtaining the similarity evaluation index under different sliding conditions in the vertical and horizontal directions includes:

[0013] The sum of the pixel grayscale difference under different sliding directions in any direction and a preset first constant is used as the denominator, and the structural similarity is used as the numerator. The resulting ratio is used as the similarity evaluation index under different sliding directions in the corresponding direction.

[0014] Furthermore, the curve fitting method includes:

[0015] Using the sliding amount in any direction as the abscissa and the similarity evaluation index corresponding to the sliding amount as the ordinate, curve fitting is performed using the least squares method to obtain the fitted curve in the corresponding direction; the sliding amount is the actual displacement step value between the position after the sliding window slides and the initial position.

[0016] Furthermore, methods for obtaining extreme points on the fitted curve include:

[0017] Obtain the curvature at the corresponding point on the fitted curve for each slip amount. Perform a first binary classification on all points based on the curvature, and retain the points of the class with the largest average curvature. Perform a second binary classification on the retained points based on the similarity evaluation index corresponding to the points to obtain the extreme points on the fitted curve.

[0018] Furthermore, methods for obtaining the period include:

[0019] Calculate the difference in slip between any two adjacent extreme points in any direction to determine the set of differences; calculate the mean of all differences in the set of differences as the period in the corresponding direction.

[0020] Furthermore, the optimized CLAHE algorithm includes the following block mesh size parameters in the vertical and horizontal directions:

[0021] For any direction, substitute the period in the corresponding direction into the design formula for the block mesh size parameter to obtain the block mesh size parameter for each direction;

[0022] The formula for designing the block grid size parameters is:

[0023]

[0024] In the formula:

[0025] : Block grid size parameters;

[0026] (): Function to find the maximum value;

[0027] (): Minimum function;

[0028] ROU(): Takes the smallest integer power of 2 that is greater than the input value;

[0029] Y: Dimension in any direction of the wedge slope image;

[0030] X: Preset second constant;

[0031] : The period in any direction of the wedge slope image.

[0032] Furthermore, the crack risk assessment indicators include three categories of indicators: crack width growth rate, main crack length, and branch crack number.

[0033] Furthermore, landslide hazards are monitored based on crack risk assessment indicators, including:

[0034] If any one of the crack risk assessment indicators is greater than the preset safety threshold under the corresponding dimension, the crack area is judged to be a level 1 high risk.

[0035] If any two of the crack risk assessment indicators are greater than the preset safety threshold in the corresponding dimension, the crack area is judged to be a level 2 high risk.

[0036] If three of the crack risk assessment indicators are greater than the preset safety threshold in the corresponding dimension, the crack area is judged to be a level three high-risk area.

[0037] The present invention also proposes a mine safety monitoring system for open-pit mining areas, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of any one of the mine safety monitoring methods for open-pit mining areas.

[0038] The present invention has the following beneficial effects:

[0039] This invention considers that wedge-shaped slopes are formed by multiple rock facets cutting into each other, and these facets are periodically arranged on the rock face. Therefore, wedge-shaped slopes have periodic characteristics in both the vertical and horizontal directions. This scheme first constructs different sliding windows that slide on the wedge-shaped slope image to obtain similarity evaluation indices under different sliding conditions. This allows for multi-scale acquisition of the periodic characteristics of the wedge-shaped slope surface, thus comprehensively evaluating the similarity of different regions in the wedge-shaped slope image. Then, through the similarity evaluation indices, the periodicity of the wedge-shaped slope image in the vertical and horizontal directions is obtained. Optimizing the block grid parameters based on the periodicity matches the processing scale of the algorithm with the periodic characteristics of the wedge-shaped slope image itself, effectively improving the targeting of image enhancement and the detail of the enhanced image. Finally, using the enhanced wedge-shaped slope image to obtain crack risk assessment indices significantly improves the accuracy and reliability of crack risk assessment indices, thereby enhancing the accuracy and reliability of slope landslide disaster monitoring. Attached Figure Description

[0040] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0041] Figure 1 This is a flowchart of a mine safety monitoring method for open-pit mining areas, provided as an embodiment of the present invention. Detailed Implementation

[0042] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a mine safety monitoring method and system for open-pit mining areas proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0043] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0044] The following description, in conjunction with the accompanying drawings, details the specific scheme of a mine safety monitoring method and system for open-pit mining areas provided by the present invention.

[0045] This invention proposes a method for monitoring mine safety in open-pit mining areas. Please refer to [link / reference]. Figure 1 The diagram illustrates a flowchart of a mine safety monitoring method for open-pit mining areas according to an embodiment of the present invention. The method includes:

[0046] Step S101: Obtain wedge-shaped slope images of the mining area, construct sliding windows in the vertical and horizontal directions respectively, and obtain similarity evaluation indicators under different sliding conditions in the vertical and horizontal directions based on the pixel grayscale difference and structural similarity between the images in the sliding windows before and after sliding in the vertical and horizontal directions.

[0047] To analyze the cracked areas on the surface of the wedge slope, it is necessary to first obtain images of the wedge slope in the mining area. However, drones cannot capture all the images of the wedge slope at once. Therefore, drone timed photogrammetry is used to obtain multiple local wedge slope surface images, and all local wedge slope surface images are stitched together along the drone's flight path to form a single wedge slope image.

[0048] The formation mechanism of wedge-shaped slopes involves the mutual cutting of two or more rock facets. Under the action of mutual cutting and compression, each facet exhibits a wedge shape. These wedge-shaped facets are periodically arranged on the rock face, resulting in a wedge-shaped facet appearing at regular intervals in both the vertical and horizontal directions. Therefore, similar local features appear on the surface of the wedge-shaped slope at regular intervals in both the vertical and horizontal directions. This invention considers that the crack area is darker than the rock face, and the pixel grayscale difference between the crack area and the rock face area is greater, while the pixel grayscale difference between crack areas is smaller within the same mining area. Therefore, to analyze the similarity between different areas of the wedge-shaped slope image, this invention uses the same sliding window to continuously slide across the wedge-shaped slope image in the vertical or horizontal direction, obtaining the pixel grayscale difference and structural similarity between all pixels within each sliding window, and constructing a similarity evaluation index. The higher the similarity between the images in two windows, the smaller the pixel grayscale difference between the images in the two sliding windows, and the higher the structural similarity. Using a sliding window with multiple sliding values ​​can ensure the acquisition of periodic features of the wedge slope surface at multiple scales, and comprehensively evaluate the similarity of different regions of the wedge slope image.

[0049] In one specific implementation of this invention, the UAV delayed photogrammetry adopts an S-shaped flight path mode of shooting layer by layer from high to low, with the forward overlap rate set to 80%, the lateral overlap rate to 70%, and the UAV tilt angle set to the average slope of the wedge slope.

[0050] In one specific implementation of this invention, the width of the vertical sliding window is set to the width of the wedge-shaped slope image, and the height is set to the height of the camera's capture size; the width of the horizontal sliding window is set to the width of the camera's capture size, and the height is set to the height of the wedge-shaped slope image. The initial position of the vertical sliding window is set at the top of the wedge-shaped slope image, and the window slides from top to bottom; the initial position of the horizontal sliding window is set at the left side of the wedge-shaped slope image, and the window slides from left to right. The sliding window in either direction is slid along the corresponding direction, sliding one pixel at a time until it reaches the bottom or right side of the wedge-shaped slope image. The sliding step size is set to 1 pixel.

[0051] Step S102: The similarity evaluation index under different sliding in the vertical and horizontal directions is curve-fitted with the sliding amount under different sliding in the corresponding directions to obtain the fitted curve; the periodic analysis of the wedge slope image is performed according to the position of the extreme point to obtain the period of the wedge slope image in the vertical and horizontal directions.

[0052] Within the same sliding window, the similarity evaluation index is at its maximum. As the sliding window continues to slide vertically or horizontally, the similarity evaluation index between the image within the sliding window and the image within the unslided window shows a trend of first decreasing, then increasing, then decreasing again, and then increasing again, repeating continuously. Multiple minimum points where the similarity evaluation index increases from small to large and multiple maximum points where the similarity evaluation index decreases from large to small appear on the fitted curve. The amount of sliding between each pair of adjacent maximum or minimum points represents a possible period of the wedge slope image in the vertical or horizontal direction. Therefore, periodic analysis of all maximum or minimum points on the fitted curve yields the period of the wedge slope image in the vertical and horizontal directions.

[0053] Step S103: Based on the period of the wedge slope image in the vertical and horizontal directions, optimize the block grid size parameters of the CLAHE algorithm in the vertical and horizontal directions, and perform image enhancement to obtain an enhanced wedge slope image.

[0054] The CLAHE algorithm's grid size parameter determines the spatial scale of the enhanced image. If the grid size parameter is smaller than the period in the corresponding direction, each grid contains only a portion of the crack or only the rock wall background, resulting in a discontinuous enhanced image. If the grid size parameter is larger than the period in the corresponding direction, too many cracks are mixed in one grid, causing the contrast differences between different cracks to be averaged out. Therefore, optimizing the grid size parameter using the inherent period of the wedge slope can enhance the crack area while reducing the influence of the background area. It should be noted that the specific image enhancement steps of the CLAHE algorithm are well-known techniques to those skilled in the art and will not be elaborated here.

[0055] Step S104: Obtain crack risk assessment indicators for cracked areas in the enhanced wedge slope image, and monitor landslide hazards based on crack risk assessment indicators.

[0056] Landslides typically begin in cracked areas on slopes. The greater the safety risk posed by a crack, the more likely the surrounding rock face is to detach and cause a landslide. Therefore, detecting the safety risk of cracks is equivalent to monitoring the safety risk of landslides. This embodiment of the scheme considers the morphological characteristics of the cracks themselves and uses a crack safety risk assessment index to monitor landslides.

[0057] In summary, this invention considers that wedge slopes are formed by multiple rock facets cutting into each other, and these facets are periodically arranged on the rock face. Therefore, wedge slopes exhibit periodic characteristics in both the vertical and horizontal directions. This scheme first constructs different sliding windows that slide on the wedge slope image to obtain similarity evaluation indices under different sliding conditions. This allows for multi-scale acquisition of the periodic characteristics of the wedge slope surface, comprehensively evaluating the similarity of different regions of the wedge slope image. Then, through the similarity evaluation indices, the periodicity of the wedge slope image in the vertical and horizontal directions is obtained. Based on this periodicity, the block grid parameters are optimized, matching the algorithm's processing scale with the periodic characteristics of the wedge slope image itself, effectively improving the targeting of image enhancement and the detail of the enhanced image. Finally, using the enhanced wedge slope image to obtain crack risk assessment indices significantly improves the accuracy and reliability of crack risk assessment indices, thereby enhancing the accuracy and reliability of slope landslide disaster monitoring.

[0058] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the pixel grayscale difference between the images within the sliding window before and after different sliding in the vertical and horizontal directions includes:

[0059] A sliding window in any direction is slid across the wedge slope image along the corresponding direction. The absolute value of the difference between the grayscale value of each pixel in all sliding windows and the corresponding pixel in the initial sliding window is compared to determine the absolute value of the grayscale difference. The average of the absolute values ​​of the grayscale difference in each sliding window is taken as the pixel grayscale difference under that sliding. The pixel grayscale differences under all sliding in the corresponding direction are obtained. By sliding the window one pixel at a time and performing multiple sliding operations, the difference in grayscale values ​​can be compared pixel by pixel, which can accurately identify the high-contrast area between the crack and the rock wall.

[0060] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the similarity evaluation index under different sliding conditions in the vertical and horizontal directions includes:

[0061] The sum of the pixel grayscale difference under different sliding directions in any direction and a preset first constant is used as the denominator, and the structural similarity is used as the numerator. The resulting ratio is used as the similarity evaluation index under different sliding directions in the corresponding direction.

[0062] The smaller the pixel grayscale difference and the greater the structural similarity, the greater the similarity between the images before and after the sliding. Using the sum of the pixel grayscale difference and a preset constant as the denominator can prevent the image from being identical to the image before sliding, thus avoiding a situation where the denominator is zero. The similarity evaluation index can reflect the similarity of wedge slope images under different sliding amounts in the vertical or horizontal direction. In a specific implementation of this invention, the preset first constant is set to 1.

[0063] Preferably, in some possible implementations of the embodiments of the present invention, the curve fitting method includes:

[0064] Using the sliding amount in any direction as the x-axis and the corresponding similarity evaluation index as the y-axis, curve fitting is performed using the least squares method to obtain the fitted curve for that direction. The sliding amount is the actual displacement step size between the sliding window's position after sliding and its initial position. The fitted curve reflects the changing trend of the similarity evaluation index under different sliding amounts. An upward-rising fitted curve indicates that the similarity of images within all windows increases under that sliding amount; a downward-rising fitted curve indicates that the similarity of images within all windows decreases under that sliding amount.

[0065] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the extreme points on the fitted curve includes:

[0066] Obtain the curvature at the corresponding point on the fitted curve for each slip amount. Perform a first binary classification on all points based on the curvature, and retain the points of the class with the largest average curvature. Perform a second binary classification on the retained points based on the similarity evaluation index corresponding to the points to obtain the extreme points on the fitted curve.

[0067] Extreme points include maxima and minima. The monotonicity changes at extreme points, so the curvature value is the largest at these points. Primary binary classification can retain the maxima and minima with the largest average curvature. Since the similarity evaluation index for minima is smaller than that for maxima, secondary binary classification can distinguish between minima and maxima.

[0068] In one specific implementation of this invention, the binary classification method uses the K-means algorithm. Specifically, the number of clusters in the K-means algorithm used for both binary classifications is set to two.

[0069] Preferably, in some possible implementations of the embodiments of the present invention, the method for obtaining the period includes:

[0070] Calculate the difference in slip between any two adjacent zero-crossing curvature points among all extreme points in any direction to determine the set of differences; calculate the mean of all differences in the set of differences as the period in the corresponding direction.

[0071] By statistically analyzing the difference in slip volume between any two adjacent extreme points, these differences can reflect the interval between repeated occurrences of similar patterns. By averaging all the differences in slip volume as the inherent period of the pattern, a comprehensive period that simultaneously covers all repetitive features can be obtained, avoiding the deviation of manually setting the period.

[0072] In one specific implementation of this invention, the extreme point is selected as the maximum point.

[0073] Preferably, in some possible implementations of the embodiments of the present invention, optimizing the block grid size parameters of the CLAHE algorithm in the vertical and horizontal directions includes:

[0074] For any direction, substitute the period in the corresponding direction into the design formula for the block mesh size parameter to obtain the block mesh size parameter for each direction;

[0075] The formula for designing the block grid size parameters is:

[0076]

[0077] In the formula:

[0078] : Block grid size parameters;

[0079] (): Function to find the maximum value;

[0080] (): Minimum function;

[0081] ROU(): Takes the smallest integer power of 2 that is greater than the input value;

[0082] Y: Dimension in any direction of the wedge slope image;

[0083] X: Preset second constant;

[0084] : The period in any direction of the wedge slope image.

[0085] The constants 8 and 64 are used to ensure that the minimum size parameter of the block grid is not less than 8. 8, maximum not exceeding 64 64; If the block grid size parameter is less than 8 8. This results in an excessively small mesh size, which amplifies noise. If the patch mesh size parameter is greater than 64... The value of 64 results in an excessively large mesh size, which fails to effectively enhance the small crack areas in the wedge slope image.

[0086] The preset second constant X reflects the proportional relationship between the grid size and the period of the wedge slope image, that is, each grid contains Each cycle, with different X values, can adapt to the image enhancement needs of wedge slopes in mines of different sizes.

[0087] In one specific implementation of this invention, the second constant X is preset to a value of 2.

[0088] Preferably, in some possible implementations of the embodiments of the present invention, the crack risk assessment index includes three types of assessment indexes: crack width growth rate, main crack length, and branch crack number.

[0089] The crack width growth rate refers to the growth rate of the width of the main crack per unit time. In a specific implementation of this invention, the crack width growth rate is calculated using the following formula:

[0090]

[0091]

[0092] In the formula, A represents the... The rate of increase in the width of the crack skeleton over time; They represent , The average width of the crack skeleton at any given moment; This indicates the time interval between acquiring two images of the wedge slope; I represents the width of the i-th crack skeleton pixel in the wedge slope image skeleton map acquired at time t; I represents the total number of pixels in the crack skeleton.

[0093] Wherein, the length of the main crack is the average length of the main crack per unit time. In a specific implementation of this invention, the crack width growth rate is calculated using the main crack, and the formula for calculating the crack width growth rate is:

[0094]

[0095]

[0096] In the formula, express The average length of the main crack skeleton within a given time period; , They represent The length of the main crack skeleton at time t; This represents the Euclidean distance between the j-th pixel and the (j+1)-th pixel on the main crack skeleton at time t; , Let x and y represent the x-coordinate and y-coordinate of the (j+1)th and jth pixels on the main fracture skeleton at time t, respectively; J represents the total number of pixels on the main fracture skeleton.

[0097] The number of branch cracks refers to the total number of branch cracks. The main crack skeleton includes two endpoints, and one end of the branch crack skeleton is connected to the main crack, so it only contains one endpoint. The total number of branch cracks is obtained by subtracting 2 from the total number of endpoints in the crack skeleton. In a specific implementation of this invention, determining whether a pixel belongs to an endpoint pixel involves traversing each pixel in the crack skeleton and obtaining the number of crack pixels in the eight-neighborhood of that pixel. If there is only one crack pixel in the eight-neighborhood of that pixel, then that pixel belongs to an endpoint pixel.

[0098] By constructing three types of crack risk assessment indicators, the safety risks of cracks can be monitored from multiple dimensions, thereby improving the reliability of safety monitoring.

[0099] Preferably, in some possible implementations of the embodiments of the present invention, monitoring landslide disasters based on crack risk assessment indicators includes:

[0100] Landslides typically begin in cracked areas on slopes. The wider, longer, and more numerous the cracks, the more likely the surrounding rock face is to detach, leading to a landslide. Therefore, this invention aims to quantify the safety risk level of cracked areas from multiple dimensions by obtaining three crack risk assessment indicators. Based on the number of crack risk assessment indicators exceeding preset thresholds for the corresponding dimensions, different levels of safety risk monitoring can be conducted on the cracked areas.

[0101] If any one of the crack risk assessment indicators is greater than the preset safety threshold under the corresponding dimension, the crack area is judged to be a level 1 high risk.

[0102] If any two of the crack risk assessment indicators are greater than the preset safety threshold in the corresponding dimension, the crack area is judged to be a level 2 high risk.

[0103] If three of the crack risk assessment indicators are greater than the preset safety threshold in the corresponding dimension, the crack area is judged to be a level three high-risk area.

[0104] In one specific implementation of this invention, the preset safety threshold for each type of crack risk assessment index is obtained using the natural breakpoint method. Fifty index values ​​for any type of crack risk assessment index are obtained, and these 50 index values ​​are divided into three categories—high risk, medium risk, and low risk—using the natural breakpoint method. Simultaneously, two thresholds are output, which are the preset safety thresholds for the corresponding dimension of that type of crack risk assessment index. It should be noted that the natural breakpoint method is a well-known technique and will not be elaborated upon in this invention.

[0105] The natural breakpoint method can automatically find the optimal split point between each category by utilizing the distribution characteristics of the data itself, without the need to manually set a fixed threshold. This avoids the problems of poor adaptability and poor generalization ability of manually set fixed thresholds in other mining areas.

[0106] In summary, this invention first obtains a similarity evaluation index in the corresponding direction based on the pixel grayscale difference and structural similarity between images within the sliding window in the vertical and horizontal directions. Then, it performs curve fitting between the similarity evaluation index and the sliding amount in the corresponding direction to obtain the period of the wedge slope image in that direction. Next, it optimizes the block grid parameters of the CLAHE algorithm based on the period, and uses the optimized parameters for image enhancement. Finally, it monitors landslide hazards based on crack risk assessment indicators. This invention optimizes the block grid parameters through periodic analysis of wedge slope images, improving the targeting of image enhancement and the accuracy of landslide hazard monitoring.

[0107] Based on the same inventive concept, the present invention also proposes a mine safety monitoring system for open-pit mining areas, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of any one of the methods for mine safety monitoring in open-pit mining areas.

[0108] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0109] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for monitoring mine safety in open-pit mining areas, characterized in that, The method includes: Images of wedge-shaped slopes in the mining area are acquired, and sliding windows are constructed in the vertical and horizontal directions respectively. Based on the pixel grayscale difference and structural similarity between the images in the sliding windows before and after sliding in the vertical and horizontal directions, similarity evaluation indicators under different sliding conditions in the vertical and horizontal directions are obtained. The similarity evaluation index under different sliding conditions in the vertical and horizontal directions is curve-fitted with the sliding amount under different sliding conditions in the corresponding directions to obtain the extreme points on the fitted curve; the periodicity analysis of the wedge slope image is performed based on the position of the extreme points to obtain the periodicity of the wedge slope image in the vertical and horizontal directions. Based on the period of the wedge slope image in the vertical and horizontal directions, the block grid size parameters of the CLAHE algorithm in the vertical and horizontal directions are optimized, and image enhancement is performed to obtain an enhanced wedge slope image. Obtain crack risk assessment indicators for cracked areas in enhanced wedge slope images, and monitor landslide hazards based on crack risk assessment indicators; The optimized CLAHE algorithm includes the following block grid size parameters in the vertical and horizontal directions: For any direction, substitute the period in the corresponding direction into the design formula for the block mesh size parameter to obtain the block mesh size parameter for each direction; The formula for designing the block grid size parameters is: In the formula: : Block grid size parameters; (): Function to find the maximum value; (): Minimum function; ROU(): Takes the smallest integer power of 2 that is greater than the input value; Y: Dimension in any direction of the wedge slope image; X: Preset second constant; : The period in any direction of the wedge slope image.

2. The method for monitoring mine safety in an open-pit mining area according to claim 1, characterized in that, The method for obtaining the pixel grayscale difference between the images within the sliding window before and after different sliding in the vertical and horizontal directions includes: Slide a sliding window in any direction on the wedge slope image along the corresponding direction. Compare the absolute value of the difference between the gray value of each pixel in the image at each position in all sliding windows after sliding and the corresponding pixel in the initial sliding window. Determine the absolute value of the gray value difference. Take the average value of the absolute value of the gray value difference in each sliding window as the pixel gray value difference under that sliding. Obtain the pixel gray value difference under all sliding in the corresponding direction.

3. The method for monitoring mine safety in an open-pit mining area according to claim 1, characterized in that, The methods for obtaining the similarity evaluation indicators under different sliding conditions in the vertical and horizontal directions include: The sum of the pixel grayscale difference under different sliding directions in any direction and a preset first constant is used as the denominator, and the structural similarity is used as the numerator. The resulting ratio is used as the similarity evaluation index under different sliding directions in the corresponding direction.

4. The method for monitoring mine safety in an open-pit mining area according to claim 1, characterized in that, The curve fitting method includes: Using the sliding amount in any direction as the abscissa and the similarity evaluation index corresponding to the sliding amount as the ordinate, curve fitting is performed using the least squares method to obtain the fitted curve in the corresponding direction; the sliding amount is the actual displacement step value between the position after the sliding window slides and the initial position.

5. A method for monitoring mine safety in an open-pit mining area according to claim 4, characterized in that, Methods for obtaining extreme points on a fitted curve include: Obtain the curvature at the corresponding point on the fitted curve for each slip amount. Perform a first binary classification on all points based on the curvature, and retain the points of the class with the largest average curvature. Perform a second binary classification on the retained points based on the similarity evaluation index corresponding to the points to obtain the extreme points on the fitted curve.

6. The method for monitoring mine safety in an open-pit mining area according to claim 5, characterized in that, Methods for obtaining the period include: Calculate the difference in slip between any two adjacent extreme points in any direction to determine the set of differences; calculate the mean of all differences in the set of differences as the period in the corresponding direction.

7. A method for monitoring mine safety in an open-pit mining area according to claim 1, characterized in that, The crack risk assessment indicators include three categories: crack width growth rate, main crack length, and branch crack number.

8. A method for monitoring mine safety in an open-pit mining area according to claim 7, characterized in that, Monitoring landslide hazards based on crack risk assessment indicators includes: If any one of the crack risk assessment indicators is greater than the preset safety threshold under the corresponding dimension, the crack area is judged to be a level 1 high risk. If any two of the crack risk assessment indicators are greater than the preset safety threshold in the corresponding dimension, the crack area is judged to be a level 2 high risk. If three of the crack risk assessment indicators are greater than the preset safety threshold in the corresponding dimension, the crack area is judged to be a level three high-risk area.

9. A mine safety monitoring system for an open-pit mining area, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the mine safety monitoring method for open-pit mining areas as described in any one of claims 1 to 8.

Images