Method, device, medium and equipment for extracting minimum resistance line of medium-length hole blasting

Abstract

The application discloses a minimum resistance line extraction method, device, medium and equipment for medium-length hole blasting, relates to the technical field of digital mine blasting, and comprises the following steps: preprocessing three-dimensional original point cloud of an open pit mine to obtain geological target point cloud data; determining a slope zenith distance effective interval and roughness of each point cloud point based on the geological target point cloud data, and configuring a region growing constraint rule based on the slope zenith distance effective interval, a roughness determination threshold and a normal vector angle condition; screening seed point cloud points meeting the region growing constraint rule in the geological target point cloud data, performing region growing processing with the seed point cloud points as starting points, obtaining an initial slope cluster, and sequentially performing density clustering and edge detection processing on the initial slope cluster to obtain a reconstructed slope free surface in combination with a boundary constraint; determining an explosive geometric center based on the geological target point cloud data, and calculating a Euclidean distance minimum value between the explosive geometric center and the slope free surface as a minimum resistance line of a target blast hole.

Description

Technical Field

[0001] This application relates to the field of digital mining blasting technology, and in particular to a method, apparatus, medium and equipment for extracting the minimum resistance line in medium-deep hole blasting. Background Technology

[0002] In open-pit mining, medium-deep hole blasting is a core process for rock stripping and mineral resource extraction. Its operational efficiency and blasting effect directly affect the overall benefits, resource utilization, and construction safety of the mine. The minimum resistance line, a key parameter in medium-deep hole blasting design, is defined as the shortest straight-line distance from the geometric center of the explosive to the free surface of the slope. Its accuracy directly determines the distribution of blasted fragments, the range of flyrock control, the energy utilization efficiency of the explosive, and the mining cost. It is a crucial technical foundation for optimizing borehole network parameter design and ensuring the safe and stable operation of blasting.

[0003] However, the terrain of open-pit mine slopes is complex and varied, often with complex working conditions such as protruding rock blocks and accumulation of blasting residue. Moreover, the risk of close-range manual operations in high and steep slope areas is extremely high. Traditional methods for extracting the minimum resistance line mostly rely on manual experience estimation or indirect means such as total station polar coordinate measurement. These methods not only have the drawbacks of long measurement time, low efficiency, and strong subjectivity, but also cannot directly obtain accurate minimum resistance line data, making it difficult to meet the needs of efficient and high-precision blasting design in complex mining scenarios. Summary of the Invention

[0004] In view of this, this application provides a method, apparatus, medium and equipment for extracting the minimum resistance line in medium-deep hole blasting, which can accurately obtain the minimum resistance line data and can adapt to the needs of efficient and high-precision blasting design in complex mining scenarios.

[0005] According to a first aspect of this application, a method for extracting the minimum resistance line in medium-deep hole blasting is provided, comprising:

[0006] The original three-dimensional point cloud of the open-pit mine is preprocessed to obtain geological target point cloud data. The preprocessing includes at least spatial dimensionality reduction processing and non-geological target removal processing.

[0007] Based on the geological target point cloud data, the effective range of the slope zenith distance and the roughness of each point cloud point are determined, and the region growth constraint rules are configured based on the effective range of the slope zenith distance, the judgment threshold of the roughness and the normal vector angle condition.

[0008] Seed point cloud points that satisfy the regional growth constraint rules are selected from the geological target point cloud data. Regional growth processing is performed starting from the seed point cloud points to obtain an initial slope cluster. Density clustering and edge detection processing are performed on the initial slope cluster in sequence. Combined with boundary constraints, the reconstructed slope free surface is obtained.

[0009] Based on the top coordinates and borehole parameters of the target borehole in the geological target point cloud data, the geometric center of the explosive is determined, and the minimum Euclidean distance between the geometric center of the explosive and the free surface of the slope is calculated as the minimum resistance line of the target borehole.

[0010] According to a second aspect of this application, a minimum resistance line extraction device for medium-deep hole blasting is provided, comprising:

[0011] The processing module is used to preprocess the original three-dimensional point cloud of the open-pit mine to obtain geological target point cloud data. The preprocessing includes at least spatial dimensionality reduction processing and non-geological target removal processing.

[0012] The configuration module is used to determine the effective range of the slope zenith distance and the roughness of each point cloud point based on the geological target point cloud data, and to configure the regional growth constraint rules based on the effective range of the slope zenith distance, the judgment threshold of the roughness and the normal vector angle condition.

[0013] The reconstruction module is used to filter seed point cloud points that meet the regional growth constraint rules in the geological target point cloud data, perform regional growth processing with the seed point cloud points as the starting point to obtain an initial slope cluster, and perform density clustering and edge detection processing on the initial slope cluster in sequence, and combine the boundary constraints to obtain the reconstructed slope free surface.

[0014] The calculation module is used to determine the geometric center of the explosive based on the top coordinates and parameters of the target borehole in the geological target point cloud data, and to calculate the minimum Euclidean distance between the geometric center of the explosive and the free surface of the slope, which is used as the minimum resistance line of the target borehole.

[0015] According to a third aspect of this application, a storage medium is provided that stores a computer program thereon, which, when executed by a processor, implements the above-described method for extracting the minimum resistance line in deep-hole blasting.

[0016] According to a fourth aspect of this application, an electronic device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the program to implement the above-described method for extracting the minimum resistance line of deep-hole blasting.

[0017] By employing the aforementioned technical solution, this application provides a method, apparatus, medium, and equipment for extracting the minimum resistance line in medium-deep hole blasting. Through spatial dimensionality reduction and non-geological target removal preprocessing of the original three-dimensional point cloud of an open-pit mine, it effectively reduces the computational load of subsequent data processing, improves overall processing efficiency, removes irrelevant interference points, and ensures the purity and validity of the geological target point cloud data, laying a high-quality data foundation for subsequent accurate analysis. Next, based on the geological target point cloud data, the effective range of the slope zenith distance and the roughness of each point cloud point are determined. Combined with the normal vector angle condition, regional growth constraint rules are configured. Multiple constraints accurately lock down slope features, avoiding the one-sidedness of a single constraint and improving the accuracy and reliability of seed point selection. Subsequently, regional growth is performed starting from the seed point cloud points that satisfy the constraint rules. The initial slope cluster is obtained, and then noise points such as rock protrusions and blasting residues are removed through density clustering. The top and bottom lines of the slope are identified through edge detection. Combined with boundary constraints, the free surface of the slope is reconstructed, which can effectively solve the problem of distortion in free surface extraction caused by noise interference under complex working conditions and achieve accurate depiction of the free surface of the slope. Finally, the geometric center of the explosive is determined based on the top coordinates and parameters of the target blast hole in the geological target point cloud data. The minimum resistance line is obtained by calculating the minimum Euclidean distance between the center and the free surface of the slope. The fully automated operation can avoid the safety risks of manual close-range operation in high and steep slope areas and overcome the defects of traditional methods such as long measurement time, low efficiency and strong subjectivity. It can directly obtain accurate minimum resistance line data, which is perfectly adapted to the high-efficiency and high-precision blasting design requirements in complex mining scenarios.

[0018] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0019] Figure 1 A flowchart illustrating a method for extracting the minimum resistance line in medium-deep hole blasting according to an embodiment of this application is shown.

[0020] Figure 2 A flowchart illustrating a method for extracting the minimum resistance line in medium-deep hole blasting according to another embodiment of this application is shown.

[0021] Figure 3 A flowchart illustrating the principle of minimum resistance line extraction for medium-deep hole blasting according to an embodiment of this application is shown.

[0022] Figure 4 This illustration shows a structural schematic diagram of a minimum resistance line extraction device for medium-deep hole blasting provided in an embodiment of this application. Detailed Implementation

[0023] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.

[0024] Open-pit mine slopes are characterized by complex and varied terrain, often featuring protruding rock blocks and accumulated blasting debris. Furthermore, close-range manual operations in steep slope areas pose extremely high risks. Traditional methods for extracting the minimum resistance line rely heavily on manual experience estimation or indirect means such as total station polar coordinate measurement. These methods not only suffer from drawbacks such as long measurement time, low efficiency, and strong subjectivity, but also fail to directly obtain accurate minimum resistance line data, making it difficult to meet the needs of efficient and high-precision blasting design in complex mining scenarios.

[0025] To overcome the limitations of traditional manual measurement, automated extraction technology based on 3D point cloud data can be developed, among which two core algorithms are widely used. One is the region growing algorithm, which relies on terrain continuity constraints, selects initial seed points by setting growth rules, and then gradually expands to generate slope clusters to extract the free surface of the slope. However, due to the inevitable inclusion of noise points such as rock protrusions and blasting residues in open-pit mine point cloud data, this type of algorithm is prone to oversegmentation due to noise interference, resulting in distortion of the free surface extraction boundary and affecting the accuracy of the minimum resistance line calculation. The other is the DBSCAN clustering algorithm, which uses density threshold judgment to remove noise points and extract target clusters. However, the setting of key parameters such as neighborhood radius and minimum cluster size is highly dependent on human experience. In areas with uneven point cloud density (such as the transition zone between the top and bottom of the slope), undersegmentation is prone to occur, making it difficult to effectively suppress noise while completely preserving the true boundary features of the free surface of the slope. This results in insufficient stability and reliability of the minimum resistance line calculation results, and cannot provide accurate data support for blasting design.

[0026] To address the shortcomings of traditional manual measurement and the inadequacy of single algorithms, this invention provides a method for extracting the minimum resistance line in medium-deep hole blasting that integrates region growth and DBSCAN optimization algorithms. This method enables accurate extraction of free surfaces and automated, high-precision calculation of the minimum resistance line, mitigating risks in slope operations and providing reliable data support for blasting design optimization.

[0027] Accordingly, in order to solve the above-mentioned technical problems, embodiments of the present invention provide a method for extracting the minimum resistance line in medium-deep hole blasting, such as... Figure 1 As shown, the method includes:

[0028] Step 110: Preprocess the original 3D point cloud of the open-pit mine to obtain geological target point cloud data. The preprocessing includes at least spatial dimensionality reduction processing and non-geological target removal processing.

[0029] Among them, the open-pit mine's original 3D point cloud is a set of discrete 3D coordinate points that reflect the topography and surface objects of the open-pit mine, collected through technologies such as 3D laser scanning. It includes geological and topographical information as well as non-geological interference information. The geological target point cloud data is point cloud data that has been preprocessed and retains only geological features such as slope surfaces. Spatial dimensionality reduction processing is a technical operation that reduces the total amount of point cloud data, reduces the computational load of subsequent data processing, and improves processing efficiency without losing key topographical features of the open-pit mine. Non-geological target removal processing is a technical operation that removes interference points (such as workers, engineering equipment, etc.) that are unrelated to geology and topography from the point cloud data through filtering and other technical means, thus purifying the point cloud data.

[0030] In this embodiment of the disclosure, for the original 3D point cloud of an open-pit mine obtained through 3D scanning technology, an appropriate spatial dimensionality reduction algorithm (including but not limited to the octree downsampling algorithm) is first used to perform spatial dimensionality reduction processing on the basis of fully preserving key geological and topographic features such as slope surface and slope top, thereby reducing data redundancy; then, through targeted filtering techniques (including but not limited to morphological filtering), interference removal operation is performed on the compressed point cloud data to remove non-geological target points, and finally, geological target point cloud data containing only geological core features and with pure data quality is obtained, providing reliable data support for subsequent technical steps.

[0031] This preprocessing step, through spatial dimensionality reduction, effectively reduces the computational complexity of subsequent data processing and improves the overall efficiency of the method. At the same time, by removing non-geological targets, irrelevant interference points can be eliminated, ensuring the purity and validity of the geological target point cloud data. The synergistic effect of these two steps not only avoids the impact of redundant data on processing accuracy but also lays a high-quality data foundation for subsequent key steps such as determining the effective range of slope zenith distance and reconstructing the slope free surface, thus ensuring the accuracy and stability of the minimum resistance line extraction from the source.

[0032] Step 120: Determine the effective range of slope zenith distance and the roughness of each point cloud point based on the geological target point cloud data, and configure regional growth constraint rules based on the effective range of slope zenith distance, the judgment threshold of roughness and the angle condition of normal vectors.

[0033] Among them, the effective range of slope zenith distance is the range of zenith distance values ​​that can accurately reflect the slope tilt characteristics, selected by cluster analysis of the zenith distance of each point in the geological target point cloud data; the roughness of each point cloud point is a statistical feature value calculated based on the elevation information of the neighboring points of the point cloud point, used to characterize the undulation of the terrain surface in the area where the point is located, and the core is reflected in the dispersion of the elevation values ​​of the neighboring points; the roughness judgment threshold is a critical value determined based on the roughness statistical characteristics of the initial seed point set with slope characteristics, used to judge the consistency of the growth point and the seed point in roughness characteristics, and is the core quantitative standard of roughness constraint; the normal vector angle condition is a constraint requirement set based on the overall distribution law of the normal vectors of all point cloud points in the geological target point cloud data, which ensures the terrain continuity of the growth area by limiting the angle range between the normal vectors of the growth point and the seed point; the regional growth constraint rule is a comprehensive constraint system formed by integrating the effective range of slope zenith distance, the roughness judgment threshold and the normal vector angle condition, used to accurately select growth points that meet the slope characteristics and guide the orderly progress of the regional growth process.

[0034] In this embodiment of the disclosure, based on the preprocessed geological target point cloud data, the normal vector of each point cloud point is first solved using an adapted algorithm. The zenith distance distribution is obtained by calculating the angle between the normal vector and the positive direction of the Z-axis of the spatial rectangular coordinate system. Then, a cluster analysis method is used to determine the effective zenith distance interval that can characterize the slope features. At the same time, a suitable neighborhood analysis method is used to calculate the roughness of each point cloud point. Based on the roughness statistical characteristics of the initial seed point set with potential slope features, a judgment threshold for judging roughness consistency is determined. In addition, by analyzing the overall distribution characteristics of the normal vectors of all point cloud points, the constraint conditions of the normal vector angle are clarified. Finally, the effective zenith distance interval of the slope, the judgment threshold of roughness, and the normal vector angle conditions are integrated to form a comprehensive constraint rule for guiding regional growth.

[0035] This technical step accurately identifies the effective range of slope zenith distance through cluster analysis, determines the roughness threshold by combining neighborhood statistics, and sets angular constraints based on normal vector distribution. The integrated regional growth constraint rule can precisely control the growth process from three dimensions: terrain tilt characteristics, surface undulation degree, and spatial direction consistency. It effectively eliminates interference from non-slope points and noise points, ensuring that the seed points selected by regional growth and subsequent growth points all have the core features of the slope. It can provide reliable rule support for the subsequent generation of initial slope clusters and accurate reconstruction of slope free surfaces, significantly improving the accuracy and stability of slope feature extraction under complex terrain conditions.

[0036] Step 130: Select seed point cloud points that meet the regional growth constraint rules in the geological target point cloud data, perform regional growth processing starting from the seed point cloud points to obtain the initial slope cluster, perform density clustering and edge detection processing on the initial slope cluster in sequence, and obtain the reconstructed slope free surface by combining the boundary constraints.

[0037] Seed point cloud points are selected from geological target point cloud data and meet the regional growth constraints (zenith distance, roughness, and normal vector angle constraints). They serve as the starting reference points for regional growth, guiding the orderly expansion of slope features. Regional growth processing involves starting with the seed point cloud points, traversing neighboring points and verifying their compliance with the constraints, incorporating valid points into the growth range, and updating them as new seed points. This iterative process expands the slope range. The initial slope cluster is a preliminary slope shape generated through regional growth processing, consisting of multiple discrete point cloud point sets, including potential slope points, pseudo-slope clusters, and small-scale noise clusters. Density Clustering is a clustering algorithm based on the spatial density features of point clouds. By setting a density threshold, low-density noise points are removed, while high-density effective slope clusters are retained, thus achieving noise reduction. Edge detection processing is an operation that extracts slope boundary features (such as the top line and bottom line) through specific techniques to clarify the spatial boundaries of the slope. Boundary constraint is a process that uses the top line and bottom line extracted by edge detection as spatial constraints to filter the coordinates of candidate slope points, ensuring the accuracy of the slope range. The reconstructed slope free surface is a continuous slope surface that accurately reflects the real topographic features of the open-pit mine slope after initial slope cluster generation, noise removal, and boundary constraint filtering.

[0038] In this embodiment of the present disclosure, point cloud points that meet the preset regional growth constraint rules can be selected from the preprocessed geological target point cloud data as seed point cloud points. Using the seed point cloud points as the starting point, regional growth processing is performed by traversing its neighboring points and verifying the compliance of the constraints. The neighboring points that meet the rules are included in the growth range and updated as new seed points. This process is continuously iterated and expanded to form a discretely distributed initial slope cluster. Subsequently, an appropriate density clustering algorithm is used on the initial slope cluster to remove low-density noise points such as rock protrusions and blasting residues, and retain effective point clusters with slope characteristics. Then, the slope top line and slope bottom line of the slope are extracted using appropriate edge detection technology. These two lines are used as spatial boundary constraints to perform coordinate screening on the point clusters purified by density clustering, retaining only the effective slope points within the boundary interval, and finally completing the accurate reconstruction of the free surface of the slope.

[0039] This technical process, by selecting seed point cloud points that conform to the constraint rules, provides a precise starting benchmark for region growth, ensuring the correctness of the slope growth direction. Region growth processing enables rapid expansion of slope features, forming an initial cluster of points covering the potential slope. Density clustering effectively eliminates noise interference under complex conditions, purifying the slope data. The combination of edge detection and boundary constraints clearly defines the spatial extent of the slope, avoiding slope boundary distortion. Through the synergistic effect of these steps, accurate reconstruction of the free surface of slopes under complex terrain conditions can be achieved, providing a reliable slope foundation for subsequent high-precision calculation of the minimum resistance line, and significantly improving the accuracy and stability of slope feature extraction in complex mining scenarios.

[0040] Step 140: Determine the geometric center of the explosive based on the top coordinates and parameters of the target borehole in the geological target point cloud data, and calculate the minimum Euclidean distance between the geometric center of the explosive and the free surface of the slope as the minimum resistance line of the target borehole.

[0041] Among them, the top coordinates of the target blast hole are three-dimensional spatial coordinates extracted from the geological target point cloud data, representing the position of the top of the target blast hole; the blast hole parameters are key parameters describing the structure and size of the blast hole, including but not limited to the blast hole radius, blast hole depth and plugging length, used to accurately reconstruct the actual structure of the blast hole; the geometric center of the explosive is a three-dimensional coordinate point determined by comprehensively considering the blast hole structure and the distribution characteristics of the explosive segment, which is calculated by combining the average elevation of the top and bottom of the explosive segment with the horizontal coordinate of the blast hole axis, representing the central area where the explosive is concentrated; the minimum Euclidean distance is the smallest distance value among the spatial straight-line distances from the geometric center of the explosive to all discrete point cloud points on the free surface of the slope, directly reflecting the shortest path between the two; the minimum resistance line is a core parameter that determines the blasting effect and construction safety in blasting engineering, specifically referring to the shortest spatial distance from the geometric center of the explosive to the free surface of the slope, and its accuracy directly affects the blasting block size, flyrock distance and mining cost.

[0042] In this embodiment of the disclosure, the top coordinates of the target borehole can be extracted based on the geological target point cloud data. Combined with the preset borehole parameters, an adapted three-dimensional modeling method can be used to construct a three-dimensional model of the borehole that is consistent with the actual borehole structure. The distribution range of the explosive section inside the borehole can be accurately located through this model. The average elevation of the top and bottom of the explosive section can be calculated. Then, the three-dimensional coordinates of the geometric center of the explosive can be determined by combining the horizontal coordinates of the borehole axis. Finally, the Euclidean distance from the geometric center to all point cloud points on the reconstructed slope free surface can be calculated one by one. The minimum value among them can be selected as the minimum resistance line of the target borehole.

[0043] This technical process extracts precise coordinates of the top of the blast holes and constructs a 3D model by combining core blast hole parameters. This ensures the accuracy of the explosive's geometric center positioning and avoids errors caused by manual estimation. The method of calculating the minimum resistance line based on the minimum Euclidean distance can directly capture the shortest path from the explosive to the free surface of the slope, perfectly matching the core definition of the minimum resistance line and significantly improving extraction accuracy. The fully automated operation avoids the safety risks of close-range manual operations in steep slope areas and overcomes the shortcomings of traditional methods, such as low efficiency and strong subjectivity. This provides reliable core data support for blasting block size control and borehole parameter design in complex mining scenarios, ensuring the safety and stability of blasting operations.

[0044] In summary, the minimum resistance line extraction method for medium-deep hole blasting provided in this application, through spatial dimensionality reduction and non-geological target removal preprocessing of the original 3D point cloud of the open-pit mine, can effectively reduce the computational load of subsequent data processing and improve the overall processing efficiency. It can also remove irrelevant interference points and ensure the purity and validity of the geological target point cloud data, laying a high-quality data foundation for subsequent accurate analysis. Next, based on the geological target point cloud data, the effective range of the slope zenith distance and the roughness of each point cloud point are determined. Combined with the normal vector angle condition, regional growth constraint rules are configured. Multiple constraints accurately lock down slope features, avoiding the one-sidedness of a single constraint and improving the accuracy and reliability of seed point selection. Subsequently, starting from the seed point cloud points that satisfy the constraint rules, regional growth is performed to obtain the initial slope cluster. Next, density clustering is used to remove noise points such as rock protrusions and blasting residues. Edge detection is used to identify the top and bottom lines of the slope. Combined with boundary constraints, the free surface of the slope is reconstructed, which can effectively solve the problem of distortion in free surface extraction caused by noise interference under complex working conditions and achieve accurate depiction of the free surface of the slope. Finally, the geometric center of the explosive is determined based on the top coordinates and parameters of the target blast hole in the geological target point cloud data. The minimum resistance line is obtained by calculating the minimum Euclidean distance between this center and the free surface of the slope. The fully automated operation can avoid the safety risks of manual close-range operation in high and steep slope areas and overcome the defects of traditional methods such as long measurement time, low efficiency and strong subjectivity. It can directly obtain accurate minimum resistance line data, which is perfectly adapted to the high-efficiency and high-precision blasting design requirements in complex mining scenarios.

[0045] Furthermore, as a refinement and extension of the specific implementation of the above embodiments, and to fully illustrate the implementation of this embodiment, this embodiment also provides another method for extracting the minimum resistance line in medium-deep hole blasting, such as... Figure 2 As shown, the method includes:

[0046] Step 210: Preprocess the original three-dimensional point cloud of the open-pit mine to obtain geological target point cloud data.

[0047] For embodiments of this disclosure, step 210 may include the following steps:

[0048] Step 210-1: Use the octree downsampling algorithm to perform spatial dimensionality reduction on the original 3D point cloud of the open-pit mine to obtain compressed point cloud data.

[0049] Among them, the octree downsampling algorithm is a point cloud simplification algorithm based on three-dimensional spatial hierarchical indexing. It constructs an octree structure by recursively dividing the space from top to bottom, and adaptively adjusts the sampling resolution according to the point cloud distribution density, reducing the number of point clouds while retaining key terrain features. The compressed point cloud data is the simplified point cloud data obtained after processing by the octree downsampling algorithm. It not only eliminates redundant points, but also completely retains key geological features such as slope surface and slope top platform, which can provide efficient and high-quality basic data for subsequent data processing.

[0050] In this embodiment of the disclosure, after obtaining the original three-dimensional point cloud of the open-pit mine, an octree downsampling algorithm can be used to construct a three-dimensional spatial hierarchical index structure. Through top-down spatial recursive segmentation, the differences in point cloud distribution density can be identified in real time. The sampling resolution is increased in dense point cloud areas and decreased in sparse areas. While fully preserving key terrain features such as open-pit mine slopes and tops, the number of redundant point clouds is reduced, achieving spatial dimensionality reduction of point cloud data, and finally obtaining concise and efficient compressed point cloud data.

[0051] This technical step, through spatial recursive segmentation and adaptive resolution adjustment using the octree downsampling algorithm, can effectively compress the scale of the original 3D point cloud of the open-pit mine, significantly reducing the computational load of subsequent steps such as non-geological target removal and parameter calculation, and improving the overall processing efficiency. At the same time, it can accurately preserve key terrain features, avoid terrain distortion caused by data compression, and lay a high-quality data foundation for accurate processing in subsequent stages, ensuring the reliability and accuracy of the entire technical solution.

[0052] Step 210-2: For the compressed point cloud data, perform iterative operations of morphological dilation and erosion using progressive morphological filtering to remove point cloud data containing non-geological targets and obtain point cloud data containing geological targets. Non-geological targets include at least workers, pickup trucks, and forklifts.

[0053] Among them, progressive morphological filtering is a point cloud denoising and target separation algorithm based on morphological operations. Its core is to gradually increase the size of the structuring element and perform morphological combination operations on the data in stages to achieve layered and accurate removal of non-geological targets of different scales. Morphological dilation is the basic operation of morphological filtering. By expanding the spatial range of point cloud targets, it fills the small gaps inside the targets and strengthens the integrity of continuous geological and topographic features, providing a basis for subsequent erosion operations. Morphological erosion is the basic operation of morphological filtering. By shrinking the spatial range of point cloud targets, it removes discrete small noise points and protruding structures, purifies the main geological and topographic data, and complements the dilation operation. Non-geological targets are interference objects in the point cloud data that are unrelated to the geological and topographic features of the open-pit mine, including point clouds of workers, pickup trucks, bulldozers and other human or equipment objects, which will interfere with the accuracy of subsequent topographic feature extraction.

[0054] In this embodiment of the disclosure, for the compressed point cloud data after spatial dimensionality reduction, a progressive morphological filtering technique can be used. By gradually increasing the size of the structural elements and cyclically performing a combination of morphological dilation and erosion operations, small-scale non-geological targets such as workers are first removed, and then larger-scale interference objects such as pickup trucks and bulldozers are gradually removed. Under the premise of preserving the integrity of the core geological and topographic features such as open-pit mine slopes and slopes, the non-geological target point cloud and the geological target point cloud are completely separated, and finally pure geological target point cloud data is obtained.

[0055] This technical approach, through iterative design that progressively increases the size of structural elements, can accurately adapt to the removal requirements of non-geological targets of different scales, avoiding interference residues or damage to terrain features caused by structural elements of a single size. The synergistic operation of morphological expansion and erosion can effectively purify human and equipment interference in point cloud data while completely preserving the continuous features and details of geological terrain. This provides high-quality data support for subsequent key steps such as slope zenith distance calculation and regional growth constraint rule configuration, thereby significantly improving the accuracy and stability of terrain feature extraction in complex mining scenarios.

[0056] Step 220: Determine the effective range of the slope zenith distance and the roughness of each point cloud point based on the geological target point cloud data.

[0057] For embodiments of this disclosure, step 220 may include the following steps:

[0058] Step 220-1: Based on the geological target point cloud data, solve the point cloud normal vector of each point cloud point using the K-nearest neighbor algorithm, calculate the angle between the point cloud normal vector and the positive direction of the Z-axis of the spatial rectangular coordinate system to obtain the zenith distance distribution, and use the elbow rule to perform cluster analysis on the zenith distance distribution to determine the effective range of slope zenith distance.

[0059] Among them, the K-nearest neighbor algorithm is a neighborhood analysis algorithm based on the spatial distribution characteristics of data. It finds a preset number of neighboring points around the target point and solves the relevant parameters of the target point by combining the spatial information of the neighboring points. It is a commonly used method for feature extraction of point cloud data. The point cloud normal vector is a three-dimensional vector that represents the spatial orientation of the point cloud. It can reflect the terrain tilt direction of the local area where the point is located and is the core parameter for describing the spatial attitude of the slope. The positive direction of the Z-axis of the spatial rectangular coordinate system is the coordinate axis direction perpendicular to the horizontal plane and upward in the three-dimensional spatial coordinate system. It serves as the reference direction for calculating the zenith distance. The zenith distance distribution is the set and distribution law of the zenith distance values ​​of all point cloud points in the geological target point cloud data. It can reflect the overall tilt characteristics of the open-pit mine terrain. The elbow rule is a data analysis method used to determine the optimal number of clusters. By plotting the curve of the sum of squares within a cluster group as a function of the number of clusters, the number of clusters corresponding to the inflection point of the curve (resembling an "elbow") is taken as the optimal number of clusters to achieve reasonable classification of data.

[0060] In this embodiment of the disclosure, based on the preprocessed geological target point cloud data, an adapted K-nearest neighbor algorithm can be used to find the nearest points of each point cloud point. The point cloud normal vector of each point cloud point is then obtained by combining the spatial coordinate information of the nearest points, and finally calculated using the zenith distance formula. (In the formula, For point cloud normal vectors, This represents the positive direction of the Z-axis in a spatial rectangular coordinate system. This represents the inverse cosine function, with a range of values ​​of 1 to 2. The angle between the point cloud normal vector and the positive Z-axis of the spatial rectangular coordinate system is calculated to obtain the zenith distance of all point cloud points and form a zenith distance distribution. Finally, the elbow rule is used to calculate the total sum of squares (WCSS) within groups corresponding to different cluster numbers k. The optimal cluster number k=2 is determined by the inflection point of the WCSS curve as a function of cluster number. Cluster analysis is performed on the zenith distance distribution to divide the zenith distance range corresponding to the slope surface and non-slope surface, and finally, the effective range of slope zenith distance that can accurately reflect the slope surface characteristics is determined. .

[0061] This technical step uses the K-nearest neighbor algorithm to solve for the point cloud normal vectors, ensuring the accuracy of the normal vector calculation and laying the foundation for accurate acquisition of zenith distance. By calculating the angle between the normal vector and the positive Z-axis, the zenith distance distribution is obtained, which can fully capture the tilt characteristics of the terrain. Cluster analysis using the elbow rule can accurately delineate the zenith distance boundaries between slopes and non-slopes, effectively eliminating interference from non-slope points such as slope tops, flat areas, etc. The determined effective range of slope zenith distance can provide a reliable basis for the configuration of subsequent regional growth constraint rules, significantly improving the accuracy and targeting of subsequent seed point selection and slope extraction.

[0062] Step 220-2: Based on the geological target point cloud data, the K nearest neighbor method is used to select K neighboring points for each point cloud point, and the standard deviation of the K neighboring points with respect to the elevation value is calculated as the roughness of the corresponding point cloud point.

[0063] In this embodiment of the disclosure, based on the preprocessed geological target point cloud data, the K nearest neighbor method can be used to select the K nearest neighbor points for each point cloud point. This is done first using the formula... Calculate the mean elevation of the K neighboring points (where, Let K be the mean elevation of the neighborhood points of the i-th target point, and K be the number of neighborhood points. (where the elevation value is the j-th neighboring point of the i-th target point), and then the roughness calculation formula is used. Calculate the standard deviation of the elevation values ​​of K neighboring points relative to the mean, and use this standard deviation directly as the roughness of the corresponding target point cloud point. Finally, obtain the roughness quantification result of all point cloud points.

[0064] This technical step uses the K-nearest neighbor method to accurately select neighboring points of the target point, ensuring the targeted extraction of local terrain features. By using the calculation logic of elevation mean and standard deviation, the degree of undulation of the terrain surface is transformed into a quantitative roughness index, which can effectively distinguish points on gentle slopes from points in irregular areas such as rock protrusions and depressions. The obtained roughness data can provide a reliable quantitative basis for the roughness constraints in the subsequent regional growth constraint rules, and can also help to remove noise points in areas of abrupt terrain changes, significantly improving the accuracy and stability of seed point selection and subsequent slope free surface extraction.

[0065] Step 230: Configure region growth constraint rules based on the effective range of slope zenith distance, roughness determination threshold and normal vector angle condition.

[0066] The regional growth constraint rules include zenith distance constraint, roughness constraint, and normal vector angle constraint. The zenith distance constraint uses the effective range of zenith distance on the slope as the criterion to select growth points that meet the slope tilt characteristics, ensuring that regional growth only extends along the slope characteristics. The roughness constraint uses the roughness difference threshold as the core to limit the range of roughness difference between the growth point and the seed point, ensuring the surface smoothness of the growth area. The normal vector angle constraint uses the normal vector angle threshold as the criterion to limit the range of the angle between the normal vectors of the growth point and the seed point, ensuring the continuity of the topographic spatial direction of the growth area.

[0067] For embodiments of this disclosure, step 230 may include the following steps:

[0068] Step 230-1: Configure zenith distance constraints based on the effective range of zenith distance on the slope. The zenith distance constraint is that the zenith distance of the cloud point to be grown is within the effective range of zenith distance on the slope.

[0069] Among them, the point cloud points to be grown are the K nearest neighbors of the seed point cloud points, and the point cloud points that have not yet been verified to meet the regional growth constraint rules and are to be included in the growth range.

[0070] For the embodiments of this disclosure, the effective range of the slope zenith distance can be... To define the zenith distance constraint for the core configuration, the cloud points to be grown are clearly defined. The points to be included in the growth range are the K nearest neighbors of the seed point cloud that have not yet had their constraint compliance verified. These points must satisfy the constraint that their zenith distance is within the effective range of the slope's zenith distance. .

[0071] This technical step, by accurately calculating the zenith distance and combining it with the elbow rule to determine the effective range, provides a scientific and quantitative standard for zenith distance constraints, avoiding the deviation of subjective experience in setting thresholds. By limiting the zenith distance of the points to be grown to be within the effective range of the slope, the zenith distance constraint can efficiently screen out point cloud points that conform to the slope tilt characteristics, effectively eliminating interference from non-slope points such as slope top platforms and flat ground, ensuring that regional growth only extends along the real slope characteristics, which can significantly improve the targeting and accuracy of regional growth, and ensure the purity and integrity of the initial slope cluster.

[0072] Step 230-2: Determine the initial seed point set composed of all point cloud points in the geological target point cloud data. Calculate the roughness standard deviation of the initial seed point set based on the roughness of each point cloud point in the initial seed point set, and use it as the roughness difference threshold. Configure roughness constraints based on the roughness difference threshold. The roughness constraint is that the roughness difference between the point cloud point to be grown and the seed point cloud point does not exceed the roughness difference threshold.

[0073] In this embodiment of the disclosure, all point cloud points in the geological target point cloud data can be first determined as the initial seed point set. The K-nearest neighbor method is then used to select K neighboring points for each point cloud point in this set, and the roughness calculation formula is applied. Calculate the roughness of each point, and then calculate the roughness standard deviation of the initial seed point set based on the roughness of all points. Based on this standard deviation, the roughness difference threshold is determined to be... Finally, roughness constraints are configured to define the roughness of the cloud points to be grown. Roughness of seed point cloud Must meet .

[0074] This technical step, based on an initial seed point set consisting of all points in the geological target point cloud, ensures the comprehensiveness and objectivity of roughness difference threshold calculation, avoiding subjective biases caused by manually setting thresholds. Utilizing a standardized roughness calculation formula, it accurately quantifies the surface undulation characteristics of the area where each point is located. By limiting the roughness difference range between the growth point and the seed point through roughness constraints, it effectively filters out point cloud points with consistent surface undulation characteristics, eliminating noise points in irregular areas such as rock protrusions and depressions. This ensures that regional growth extends only along smooth slopes, providing a reliable guarantee for the pure generation of subsequent initial slope clusters and the accurate reconstruction of free slope surfaces, significantly improving the consistency and stability of slope feature extraction under complex terrain.

[0075] Step 230-3: Based on the geological target point cloud data, solve the normal vector of each point cloud point using the K-nearest neighbor algorithm. By analyzing the overall distribution characteristics of the normal vectors of all point cloud points, determine the normal vector angle threshold. Configure the normal vector angle constraint based on the normal vector angle threshold. The normal vector angle constraint is that the angle between the normal vector of the point cloud point to be grown and the normal vector of the seed point cloud point does not exceed the normal vector angle threshold.

[0076] In this embodiment of the disclosure, based on the preprocessed geological target point cloud data, the K-nearest neighbor algorithm can be used to find the neighboring points of each point cloud point. The normal vector of each point cloud point can be obtained by combining the spatial coordinate information of the neighboring points. By statistically analyzing the spatial orientation distribution characteristics of the normal vectors of all point cloud points (such as the range of the included angle distribution and the degree of clustering), the threshold of the included angle of the normal vector that can ensure the consistency of the slope direction can be determined. Then, the normal vector angle constraint is configured, and the constraint requirement is specified as the normal vector of the point cloud point to be grown. Normal vector of seed point cloud points The angle between satisfy The normal vector of the point cloud to be grown Normal vector of seed point cloud points The angle between Through formula Calculation yields (" " represents the vector dot product, (represents the L2 norm of the normal vector).

[0077] This technical step determines the included angle threshold based on the overall distribution characteristics of all normal vectors, ensuring the objectivity and adaptability of the threshold and avoiding the subjective bias of manually setting the threshold. The included angle constraint of the normal vector effectively eliminates noise points in areas of abrupt changes in terrain direction by limiting the consistency of the normal vector direction between the growth point and the seed point, ensuring that the region growth only extends along the continuous slope direction in space. It forms a synergistic and complementary relationship with the zenith distance constraint and roughness constraint, which can significantly improve the spatial consistency and purity of the initial slope cluster, provide a reliable guarantee for the accurate reconstruction of the subsequent free surface of the slope, and enhance the stability and accuracy of slope feature extraction under complex terrain conditions.

[0078] Step 240: Select seed point cloud points that meet the regional growth constraint rules in the geological target point cloud data, perform regional growth processing starting from the seed point cloud points to obtain the initial slope cluster, perform density clustering and edge detection processing on the initial slope cluster in sequence, and obtain the reconstructed slope free surface by combining the boundary constraints.

[0079] For embodiments of this disclosure, step 240 may include the following steps:

[0080] Step 240-1: Using the selected seed point cloud points as the starting point for growth, traverse the K nearest neighbors of each seed point cloud point, include the K nearest neighbors that satisfy the regional growth constraint rules into the growth range and update them as new seed point cloud points, and continuously iterate to expand the growth boundary to generate discrete initial slope clusters.

[0081] Among them, the growth boundary is the spatial range limit of the effective slope points currently included in the region growth process. It continues to expand outward with the generation of new seed points until there are no more points to be grown that meet the constraints. The initial slope cluster is a slope prototype generated by iterative expansion of the region growth, consisting of multiple discrete point cloud point sets. It contains potential point clusters with slope characteristics and is the basic data for subsequent optimization processing.

[0082] In this embodiment of the disclosure, seed point cloud points that are pre-selected and satisfy the constraints of zenith distance, roughness, and normal vector angle can be used as the starting point for growth. The K nearest neighbors of each seed point cloud point are traversed through spatial retrieval technology, and each nearest neighbor point is verified to meet the above-mentioned regional growth constraint rules. The K nearest neighbors that simultaneously meet all the constraints are included in the growth range and updated as new seed point cloud points. The growth boundary is continuously expanded through iteration until there are no more points to be grown that meet the constraint rules, and finally an initial slope cluster composed of multiple discrete point cloud point sets is generated.

[0083] Step 240-2: Based on the total number of point clouds in the geological target point cloud data, set a small cluster threshold, screen the initial slope clusters one by one, remove pseudo slope clusters and small-scale noise clusters with a point cloud count less than the small cluster threshold, and retain potential slope clusters with slope characteristics.

[0084] Among them, the pseudo-slope cluster is a set of point cloud points in the initial slope cluster that does not have the characteristics of a real slope surface. It is mostly formed by the erroneous growth of non-slope points and usually has a small number of points. The small-scale noise cluster is a set of points in the initial slope cluster that consists of a small number of discrete noise points, such as small rock debris and local interference points, and does not have the characteristics of slope continuity. The potential slope cluster is a set of point cloud points that is retained after small cluster threshold screening, which meets the threshold requirements and has the characteristics of a real slope surface. It is the basic data for subsequent density clustering optimization.

[0085] For the embodiments of this disclosure, the total number of point cloud data of the preprocessed geological target point cloud data can be counted first. Combined with the proportional coefficient for cloud density adapted to open-pit mine sites (Value range is 0.001~0.01, dynamically adjusted according to cloud density at open-pit mine sites) Dynamically set small cluster threshold. Then, the discrete initial slope clusters generated by the regional growth are screened one by one to determine whether the number of points in each slope cluster reaches the small cluster threshold. Pseudo slope clusters and small-scale noise clusters with fewer points than the threshold are removed, and finally potential slope clusters with the required number of points and real slope characteristics are retained.

[0086] This technical step sets a small cluster threshold based on the total number of points in the geological target point cloud, combined with a dynamically adjusted scaling factor. This ensures the objectivity and scene adaptability of the small cluster threshold, avoiding screening bias caused by a fixed threshold. By removing pseudo-slope clusters and small-scale noise clusters with fewer points, the initial slope cluster data can be effectively purified, improving the overall purity and effectiveness of the slope clusters. This lays a high-quality data foundation for subsequent DBSCAN clustering to remove large-scale noise points such as rock protrusions and blasting residues, significantly enhancing the accuracy and stability of subsequent slope free surface reconstruction.

[0087] Step 240-3: Use the DBSCAN clustering algorithm to perform density clustering on potential slope clusters, remove noise points corresponding to rock protrusions and blasting residues, and obtain denoised slope clusters.

[0088] In this embodiment of the present disclosure, a preset number of sample point cloud points can be randomly selected from the potential slope clusters after small cluster threshold screening. The average K-nearest neighbor distance of each sample point can be calculated using the K-nearest neighbor algorithm. The average K-nearest neighbor distances of all sample points can be taken as the neighborhood radius of the DBSCAN algorithm. Then, based on the neighborhood radius Through formula Calculate the volume of the neighborhood spheres; then count the total number of potential slope clusters. Point cloud count for each potential slope cluster And calculate the total bounding box volume of all potential slope clusters. Through formula Calculate the minimum cluster size for the DBSCAN algorithm. ,in, Finally, based on the determined neighborhood radius... and minimum cluster size The DBSCAN clustering operation is performed to identify and remove discrete point clusters with a density lower than the minimum cluster size as noise points corresponding to rock protrusions and blasting residues, thus obtaining pure, denoised slope clusters.

[0089] Accordingly, the specific steps of the implementation may include: randomly selecting a preset number of sample point cloud points from the potential slope clusters, calculating the average K-nearest neighbor distance for each sample point cloud point, determining the mean of the average K-nearest neighbor distances for all sample point cloud points as the neighborhood radius of the DBSCAN algorithm, and calculating the neighborhood sphere volume based on the neighborhood radius; calculating the minimum cluster size of the DBSCAN algorithm based on the number of potential slope clusters, the number of point clouds in each potential slope cluster, the bounding box volume of the slope cluster, and the neighborhood sphere volume; performing DBSCAN clustering based on the neighborhood radius and the minimum cluster size, removing noise points corresponding to rock protrusions and blasting residues, and obtaining denoised slope clusters. The denoised slope clusters are a collection of clean slope point clouds obtained after DBSCAN clustering processing and removal of low-density noise points such as rock protrusions and blasting residues, exhibiting higher slope consistency.

[0090] This technique determines the neighborhood radius by randomly selecting sample points and calculating the average distance of K-nearest neighbors. It also calculates the minimum cluster size by combining the spatial distribution characteristics of slope clusters. This enables adaptive configuration of the core parameters of the DBSCAN algorithm, avoiding the subjectivity and insufficient adaptability of manually set parameters. Leveraging the density-based clustering characteristics of the DBSCAN algorithm, it can accurately identify and remove discrete noise points such as rock protrusions and blasting residues, while fully preserving high-density effective slope clusters. This significantly improves the purity and consistency of slope point cloud data.

[0091] Step 240-4: Select the largest slope cluster with the most point cloud data from the denoised slope clusters, project all the point cloud points that make up the largest slope cluster onto the XY plane, and construct a two-dimensional image using the elevation value of each point cloud point as the gray value.

[0092] Among them, the largest slope cluster is the point cloud cluster with the most point cloud points after denoising. It is the main point cloud set that can completely represent the core topographic features of the main slope of the open-pit mine. The two-dimensional image is a planar image constructed with the XY plane as the carrier and the gray value corresponding to the point cloud elevation value. It can intuitively present the height distribution and undulation features of the slope, providing an image basis for subsequent edge detection.

[0093] In this embodiment of the disclosure, the largest slope cluster containing the most point cloud points can be selected from the slope clusters after DBSCAN clustering and denoising. The three-dimensional spatial coordinates of all point cloud points in the cluster are extracted, and the X and Y coordinates of each point are projected onto the XY plane as planar position parameters. At the same time, the Z-axis elevation value corresponding to each point is converted into the gray value of the two-dimensional image. Through the one-to-one mapping relationship between gray value and elevation value, a two-dimensional image that can intuitively reflect the slope height distribution characteristics is constructed.

[0094] This technical step selects the largest slope cluster with the most points, focusing on the core topographic features of the main slope of the open-pit mine. This eliminates the interference of minor small clusters on subsequent processing and ensures the relevance of the data. The three-dimensional point cloud is projected onto the XY plane and a two-dimensional image is constructed using the elevation value as the gray value. This successfully transforms the height difference of the three-dimensional terrain into the brightness contrast of the two-dimensional image, which can intuitively present the undulation distribution pattern of the slope.

[0095] Step 240-5: Perform Gaussian filtering on the two-dimensional image, use the Canny operator to perform edge detection on the Gaussian-filtered two-dimensional image, and perform non-maximum suppression processing during the edge detection process to extract the top and bottom lines of the slope.

[0096] Among them, Gaussian filtering is a linear smoothing filtering technique based on the Gaussian function. It suppresses noise interference while preserving the core edge features of the image by performing a weighted average calculation on image pixels and their neighboring pixels. The Canny operator is a multi-stage edge detection algorithm that accurately extracts continuous and effective edge contours in the image through gradient calculation, non-maximum suppression, and double threshold processing. Non-maximum suppression is a key step in edge detection by the Canny operator. It removes local maximum points of non-edges by comparing the gray values ​​of the target pixel with those of its neighboring pixels in the gradient direction, thereby refining the edge contours and avoiding edge widening distortion. The slope crest line is a spatial straight line representing the boundary of the top of the slope. It is the dividing line between the slope and the slope crest platform or flat land. It is obtained by capturing the gray-level change features at the top of the slope through image edge detection. The slope bottom line is a spatial straight line representing the boundary of the bottom of the slope. It is the dividing line between the slope and the slope foot area. It is obtained by capturing the gray-level change features at the bottom of the slope through image edge detection.

[0097] In this embodiment of the present disclosure, a two-dimensional image constructed with elevation as grayscale value can first be smoothed by Gaussian filtering. The Gaussian function is used to weight the pixel and its neighboring pixels to suppress noise interference and preserve the slope boundary features. Then, the Canny operator is used to calculate the gradient magnitude and gradient direction of the filtered image. Non-local maximum points on the edge are removed by non-maximum suppression to refine the edge contour. Combined with double threshold processing, strong edges and connected weak edges are screened out. Finally, the top and bottom slope lines that can represent the top and bottom boundaries of the slope are accurately extracted from the image.

[0098] Gaussian filtering effectively suppresses noise interference in two-dimensional images, avoiding false edges caused by noise and providing a clear image foundation for subsequent edge detection. The Canny operator can accurately capture the gray-scale abrupt change features of the slope boundary through gradient calculation, while non-maximum suppression can further refine the edge contour and eliminate redundant edge points, ensuring that the extracted slope top line and slope bottom line are continuous, accurate and distortion-free.

[0099] Step 240-6: Using the top and bottom slope lines as spatial boundaries, constrain the coordinates of all candidate slope point cloud points in the denoised slope cluster, retain the candidate slope point cloud points whose coordinates are within the corresponding spatial interval of the top and bottom slope lines, and obtain the reconstructed slope free surface.

[0100] In this embodiment of the disclosure, the slope top line and slope bottom line extracted by Gaussian filtering preprocessing and Canny operator edge detection can be used as spatial boundaries. First, the corresponding three-dimensional spatial coordinate intervals are determined. Then, the coordinates of all candidate slope point cloud points in the slope cluster after DBSCAN clustering and denoising are verified one by one. Only valid slope point cloud points whose coordinates are within the spatial interval are retained, and non-slope points and residual interference points that exceed the boundary are removed. Finally, a reconstructed slope free surface that can accurately represent the real terrain features of the slope is formed.

[0101] Step 250: Extract the top three-dimensional coordinates of the target borehole from the geological target point cloud data. Based on the top three-dimensional coordinates and the preset borehole parameters, construct a cylindrical borehole three-dimensional model that is consistent with the actual borehole structure. The borehole parameters include borehole radius, borehole depth and plugging length.

[0102] Among them, the top three-dimensional coordinates of the target borehole are the X, Y, and Z axis coordinates accurately extracted from the geological target point cloud data, representing the specific position of the top of the target borehole in three-dimensional space, and serve as the benchmark for positioning the three-dimensional model of the borehole. The preset borehole parameters are key parameters that describe the actual structure and size of the borehole, including the borehole radius (radius of the borehole cross-section), borehole depth (vertical length from the top of the borehole to the bottom), and plugging length (length of the closed section at the bottom of the borehole without explosives), which are the core data for restoring the true structure of the borehole. The cylindrical borehole three-dimensional model is a digital model constructed from the top three-dimensional coordinates of the borehole, combined with parameters such as the borehole radius and depth. Its size and spatial posture are completely consistent with the actual borehole, and it can intuitively present the spatial distribution and internal structure of the borehole.

[0103] In this embodiment of the present disclosure, the three-dimensional coordinates of the top of the target borehole can be accurately extracted from the preprocessed geological target point cloud data. Using these coordinates as the starting reference, and combined with preset borehole parameters including borehole radius, borehole depth, and plugging length, a cylindrical borehole three-dimensional model that completely matches the actual structure, size, and spatial posture of the target borehole can be constructed through adapted three-dimensional modeling technology, thus fully restoring the spatial morphology and internal structural features of the borehole.

[0104] Step 260: Based on the axial length and plugging length of the three-dimensional model of the cylindrical borehole, locate the explosive section inside the target borehole and calculate the average elevation of the top and bottom of the explosive section.

[0105] Among them, the axial length (i.e., borehole depth) refers to the vertical length from the top to the bottom of the borehole in the three-dimensional model of the cylindrical borehole. It is the core dimension for defining the overall spatial range of the borehole and locating the explosive section. The plugging length is the preset length of the closed section at the bottom of the borehole without explosive. It is a key parameter for distinguishing between the explosive section and the plugging section and directly determines the bottom boundary of the explosive section. The explosive section is the core area in the borehole where explosive is loaded. Its spatial range is defined by the axial length of the borehole and the plugging length. It is the main carrier for the release of blasting energy. The elevation mean is the arithmetic mean of the Z-axis coordinates (elevation values) of the top and bottom of the explosive section in the three-dimensional rectangular coordinate system. It is the core quantitative data for determining the elevation of the geometric center of the explosive.

[0106] In this embodiment of the present disclosure, a cylindrical borehole three-dimensional model that is completely consistent with the actual borehole structure can be constructed. The overall vertical range of the borehole can be determined by the axial length of the model (i.e., the borehole depth). The spatial position of the bottom blocking section of the borehole can be locked by combining the preset blocking length. Then, the explosive section filled with explosives in the borehole can be accurately located. The elevation values ​​(Z-axis coordinates) corresponding to the top and bottom of the explosive section can be extracted. The average elevation value of the explosive section can be obtained by calculating the arithmetic mean of the two elevation values.

[0107] This technical step utilizes a 3D model of a cylindrical borehole to accurately reconstruct the borehole structure. Combined with quantitative constraints on axial length and plugging length, it ensures the accuracy of spatial positioning of the explosive section, avoiding boundary deviations caused by traditional manual estimation. By calculating the average elevation of the top and bottom of the explosive section, a reliable elevation basis can be provided for subsequently fusing the horizontal coordinates of the borehole axis to determine the 3D coordinates of the explosive's geometric center. This directly guarantees the accuracy of subsequent minimum resistance line calculations and provides key data support for the design of blasting parameters in open-pit mines, enhancing the safety and stability of blasting operations.

[0108] Step 270: Determine the three-dimensional coordinates of the explosive geometric center based on the average elevation and the horizontal coordinates of the borehole axis. Calculate the Euclidean distance from the explosive geometric center to each point cloud point on the free surface of the slope based on the three-dimensional coordinates. Select the minimum Euclidean distance as the minimum resistance line of the target borehole.

[0109] In this embodiment of the present disclosure, the average elevation of the top and bottom of the explosive section can be used as the Z-axis coordinate of the geometric center of the explosive. Combined with the horizontal coordinate (X coordinate, Y coordinate) of the borehole axis, the three-dimensional coordinates of the geometric center of the explosive can be determined. Then, according to the Euclidean distance calculation formula, the Euclidean distance from the geometric center of the explosive to each point cloud point on the free surface of the slope is calculated one by one. The Euclidean distance with the smallest value is selected from all the calculation results. This minimum value is the minimum resistance line of the target borehole.

[0110] To facilitate understanding of this solution, the following is combined with... Figure 3 The flowchart illustrating the principle of minimum resistance line extraction in deep-hole blasting provides a complete explanation of the technical solution of this application: The technical solution for minimum resistance line extraction in deep-hole blasting of this application begins with the point cloud data preprocessing stage. This involves using point cloud oversampling to perform spatial dimensionality reduction on the original three-dimensional point cloud of the open-pit mine to compress the data size. Simultaneously, non-geological target removal operations are performed to remove interfering points such as personnel and equipment, resulting in clean geological target point cloud data. Next, the process proceeds to the key parameter determination and regional growth rule configuration stage, where the calculation of zenith distance and the determination of the effective range of zenith distance on the slope are completed respectively. The roughness of cloud points was calculated and the difference threshold was determined to construct the constraint rules for regional growth. Subsequently, the extraction and optimization of the slope free surface were carried out. First, the DBSCAN clustering algorithm was used to perform density clustering on potential slope clusters to remove noise points, resulting in denoised slope clusters. Then, the slope top line and slope bottom line were extracted as spatial boundaries to reconstruct an accurate slope free surface. Finally, based on the slope free surface, combined with data such as the 3D model of the blast hole and the location of the explosive section, the Euclidean distance from the geometric center of the explosive to each point on the slope free surface was calculated and the minimum value was selected. Finally, the minimum resistance line of the target blast hole was extracted.

[0111] In summary, the technical solution in this application preprocesses the original 3D point cloud of the open-pit mine using octree downsampling and progressive morphological filtering. This not only compresses the point cloud size and improves subsequent processing efficiency while maintaining the integrity of terrain features, but also effectively eliminates interference from non-geological targets and ensures the purity of the geological target point cloud data. The K-nearest neighbor algorithm and elbow rule are used to determine the effective range of slope zenith distance and point cloud roughness. Combined with the normal vector angle condition, a triple region growth constraint rule is configured, which can accurately lock slope features and improve the reliability of seed point selection. Then, starting from the seed point cloud points that meet the constraint rules, region growth is performed to generate initial slope clusters, which are then subjected to small cluster removal and adaptive... By using DBSCAN clustering to remove noise points such as rock protrusions and blasting residues, and then extracting the slope crest and bottom lines through Gaussian filtering and Canny edge detection, combined with boundary constraints, accurate reconstruction of the slope free surface can be achieved. Finally, by locating the geometric center of the explosive through 3D modeling of the blast holes and calculating the minimum Euclidean distance between it and the slope free surface, the minimum resistance line can be obtained. The fully automated operation can avoid the safety risks of manual operation on steep slopes and overcome the shortcomings of traditional methods, such as being time-consuming and subjective. It can significantly improve the accuracy and efficiency of minimum resistance line extraction in complex mining scenarios, and provide a reliable basis for open-pit mine blasting block size control and borehole mesh parameter design.

[0112] Furthermore, as Figure 1 and Figure 2 The specific implementation of the method shown in this embodiment provides a minimum resistance line extraction device for medium-deep hole blasting, such as... Figure 4 As shown, the device includes: a processing module 41, a configuration module 42, a reconstruction module 43, and a calculation module 44;

[0113] Processing module 41 can be used to preprocess the original three-dimensional point cloud of the open-pit mine to obtain geological target point cloud data. The preprocessing includes at least spatial dimensionality reduction processing and non-geological target removal processing.

[0114] Configuration module 42 can be used to determine the effective range of slope zenith distance and the roughness of each point cloud point based on geological target point cloud data, and configure regional growth constraint rules based on the effective range of slope zenith distance, the judgment threshold of roughness and the angle condition of normal vectors.

[0115] Reconstruction module 43 can be used to filter seed point cloud points that meet the regional growth constraint rules in geological target point cloud data, perform regional growth processing starting from the seed point cloud points to obtain initial slope clusters, perform density clustering and edge detection processing on the initial slope clusters in sequence, and obtain the reconstructed slope free surface by combining boundary constraints.

[0116] The calculation module 44 can be used to determine the geometric center of the explosive based on the top coordinates and parameters of the target borehole in the geological target point cloud data, and to calculate the minimum Euclidean distance between the geometric center of the explosive and the free surface of the slope, which serves as the minimum resistance line of the target borehole.

[0117] In some embodiments of this application, the processing module 41 can be specifically used to perform spatial dimensionality reduction processing on the original three-dimensional point cloud of the open-pit mine using an octree downsampling algorithm to obtain compressed point cloud data; for the compressed point cloud data, progressive morphological filtering is used to perform iterative operations of morphological dilation and erosion to remove point cloud data of non-geological targets and obtain point cloud data of geological targets, which at least include workers, pickup trucks and bulldozers.

[0118] In some embodiments of this application, the configuration module 42 can be specifically used to solve the point cloud normal vector of each point cloud point based on geological target point cloud data using the K-nearest neighbor algorithm, calculate the angle between the point cloud normal vector and the positive direction of the Z-axis of the spatial rectangular coordinate system to obtain the zenith distance distribution, and use the elbow rule to perform cluster analysis on the zenith distance distribution to determine the effective range of slope zenith distance; based on geological target point cloud data, the K-nearest neighbor method is used to select K neighboring points of each point cloud point, and the standard deviation of the K neighboring points with respect to the elevation value is calculated as the roughness of the corresponding point cloud point.

[0119] In some embodiments of this application, the regional growth constraint rules include zenith distance constraints, roughness constraints, and normal vector angle constraints; the configuration module 42 can also be used to configure zenith distance constraints based on the effective range of slope zenith distance, wherein the zenith distance constraint is that the zenith distance of the point cloud to be grown is within the effective range of slope zenith distance, wherein the point cloud to be grown is the point cloud to be included in the growth range among the K nearest neighbors of the seed point cloud, which has not yet been verified to meet the regional growth constraint rules; an initial seed point set composed of all point cloud points in the geological target point cloud data is determined, and the roughness of each point cloud point in the initial seed point set is used as the basis for the determination of the initial seed point set. The roughness standard deviation of the initial seed point set is calculated as the roughness difference threshold. Roughness constraints are configured based on the roughness difference threshold, which stipulates that the roughness difference between the point cloud to be grown and the seed point cloud does not exceed the roughness difference threshold. Based on the geological target point cloud data, the normal vector of each point cloud is solved by the K-nearest neighbor algorithm. By analyzing the overall distribution characteristics of the normal vectors of all point cloud points, the normal vector angle threshold is determined. Normal vector angle constraints are configured based on the normal vector angle threshold, which stipulates that the angle between the normal vector of the point cloud to be grown and the normal vector of the seed point cloud does not exceed the normal vector angle threshold.

[0120] In some embodiments of this application, the reconstruction module 43 can be specifically used to take the selected seed point cloud points as the growth starting point, traverse the K nearest neighbors of each seed point cloud point, include the K nearest neighbors that satisfy the regional growth constraint rules into the growth range and update them as new seed point cloud points, continuously iterate and expand the growth boundary to generate discrete initial slope clusters; set a small cluster threshold based on the total number of point clouds in the geological target point cloud data, screen the initial slope clusters one by one, remove pseudo slope clusters and small-scale noise clusters with a point cloud number less than the small cluster threshold, and retain potential slope clusters with slope characteristics; use the DBSCAN clustering algorithm to perform density clustering processing on the potential slope clusters, and remove noise points corresponding to rock protrusions and blasting residues. The denoised slope clusters are obtained. The largest slope cluster with the most point cloud data is selected from the denoised slope clusters. All point cloud data points constituting the largest slope cluster are projected onto the XY plane, and a two-dimensional image is constructed using the elevation value of each point cloud data point as the gray value. Gaussian filtering is applied to the two-dimensional image, and edge detection is performed on the Gaussian-filtered two-dimensional image using the Canny operator. Non-maximum suppression is performed during the edge detection process to extract the top line and bottom line of the slope. The coordinates of all candidate slope point cloud data points in the denoised slope clusters are constrained using the top line and bottom line as spatial boundaries. Candidate slope point cloud data points whose coordinates are within the corresponding spatial interval of the top line and bottom line are retained to obtain the reconstructed free surface of the slope.

[0121] In some embodiments of this application, when using the DBSCAN clustering algorithm to perform density clustering on potential slope clusters, removing noise points corresponding to rock protrusions and blasting residues to obtain denoised slope clusters, the reconstruction module 43 can specifically be used to randomly select a preset number of sample point cloud points in the potential slope clusters, calculate the average K-nearest neighbor distance of each sample point cloud point, determine the mean of the average K-nearest neighbor distances of all sample point cloud points as the neighborhood radius of the DBSCAN algorithm, and calculate the neighborhood sphere volume based on the neighborhood radius; calculate the minimum cluster size of the DBSCAN algorithm based on the number of potential slope clusters, the number of point clouds in each potential slope cluster, the bounding box volume of the slope cluster, and the neighborhood sphere volume; and perform DBSCAN clustering operation based on the neighborhood radius and the minimum cluster size to remove noise points corresponding to rock protrusions and blasting residues to obtain denoised slope clusters.

[0122] In some embodiments of this application, the calculation module 44 can be specifically used to extract the top three-dimensional coordinates of the target borehole from the geological target point cloud data, construct a cylindrical borehole three-dimensional model consistent with the actual borehole structure based on the top three-dimensional coordinates and preset borehole parameters, the borehole parameters including borehole radius, borehole depth and plugging length; locate the explosive segment inside the target borehole based on the axial length and plugging length of the cylindrical borehole three-dimensional model, calculate the average elevation of the top and bottom of the explosive segment; determine the three-dimensional coordinates of the geometric center of the explosive based on the average elevation and the horizontal coordinate of the borehole axis; calculate the Euclidean distance from the geometric center of the explosive to each point cloud point on the free surface of the slope one by one according to the three-dimensional coordinates, and select the minimum Euclidean distance as the minimum resistance line of the target borehole.

[0123] It should be noted that other corresponding descriptions of the functional units involved in the minimum resistance line extraction device for medium-deep hole blasting provided in this embodiment can be found in [reference needed]. Figure 1 and Figure 2 The corresponding descriptions in [the document] will not be repeated here.

[0124] Based on the above, Figure 1 and Figure 2 Accordingly, this embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the above-described method. Figure 1 and Figure 2 The method for extracting the minimum resistance line in medium-deep hole blasting is shown.

[0125] Based on this understanding, the technical solution of this application can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as CD-ROM, USB flash drive, mobile hard drive, etc.) and includes several instructions to cause an electronic device (such as personal computer, server, or network device, etc.) to execute the methods of various implementation scenarios of this application.

[0126] Based on the above, Figure 1 and Figure 2 The method shown, and Figure 4 To achieve the above objectives, the present application also provides an electronic device, specifically a personal computer, tablet computer, server, or other network device, as shown in the virtual device embodiment. This device includes a storage medium and a processor; the storage medium stores a computer program; the processor executes the computer program to achieve the above-described objectives. Figure 1 and Figure 2 The method for extracting the minimum resistance line in medium-deep hole blasting is shown.

[0127] Optionally, the aforementioned physical devices may also include a user interface, a network interface, a camera, radio frequency (RF) circuitry, sensors, audio circuitry, a Wi-Fi module, etc. The user interface may include a display screen, input units such as a keyboard, etc., and optional user interfaces may also include USB interfaces, card reader interfaces, etc. The network interface may optionally include standard wired interfaces, wireless interfaces (such as Wi-Fi interfaces), etc.

[0128] Those skilled in the art will understand that the physical device structure provided in this embodiment does not constitute a limitation on the physical device, and may include more or fewer components, or combine certain components, or have different component arrangements.

[0129] The storage medium may also include an operating system and a network communication module. The operating system is a program that manages the hardware and software resources of the aforementioned physical device, supporting the operation of information processing programs and other software and / or programs. The network communication module is used to enable communication between the various components within the storage medium, as well as communication with other hardware and software in the information processing physical device.

[0130] Through the above description of the embodiments, those skilled in the art can clearly understand that this application can be implemented by means of software plus necessary general-purpose hardware platform, or it can be implemented by hardware.

[0131] This invention preprocesses the original 3D point cloud of an open-pit mine using octree downsampling and progressive morphological filtering. This not only compresses the point cloud size and improves subsequent processing efficiency while maintaining the integrity of terrain features, but also effectively eliminates interference from non-geological targets and ensures the purity of the point cloud data for geological targets. The effective range of slope zenith distance and point cloud roughness are determined using the K-nearest neighbor algorithm and the elbow rule. Combined with the normal vector angle condition, a triple region growth constraint rule is configured, which can accurately lock slope features and improve the reliability of seed point selection. Then, starting from the seed point cloud points that meet the constraint rules, region growth is performed to generate initial slope clusters. After small cluster removal and adaptive parameter adjustment... DBSCAN clustering removes noise points such as rock protrusions and blasting residues. Gaussian filtering and Canny edge detection are then used to extract the top and bottom slope lines, combined with boundary constraints, enabling accurate reconstruction of the free surface of the slope. Finally, the geometric center of the explosive is located by 3D modeling of the blast holes, and the minimum Euclidean distance between the explosive and the free surface of the slope is calculated to obtain the minimum resistance line. The fully automated operation avoids the safety risks of manual operation on steep slopes and overcomes the shortcomings of traditional methods, such as being time-consuming and subjective. It can significantly improve the accuracy and efficiency of minimum resistance line extraction in complex mining scenarios, providing a reliable basis for open-pit mine blasting block size control and borehole mesh parameter design.

[0132] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of a preferred embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing this application. Those skilled in the art will understand that the modules in the apparatus of the embodiment can be distributed within the apparatus of the embodiment as described, or can be modified to be located in one or more apparatuses different from this embodiment. The modules of the above-described embodiment can be combined into one module, or further divided into multiple sub-modules.

[0133] The serial numbers in this application are for descriptive purposes only and do not represent the superiority or inferiority of any particular implementation scenario. The above disclosures are merely a few specific implementation scenarios of this application; however, this application is not limited thereto, and any variations conceived by those skilled in the art should fall within the protection scope of this application.

Claims

1. A method for extracting the minimum resistance line in medium-deep hole blasting, characterized in that, include: The original three-dimensional point cloud of the open-pit mine is preprocessed to obtain geological target point cloud data. The preprocessing includes at least spatial dimensionality reduction processing and non-geological target removal processing. Based on the geological target point cloud data, the effective range of the slope zenith distance and the roughness of each point cloud point are determined, and the region growth constraint rules are configured based on the effective range of the slope zenith distance, the judgment threshold of the roughness and the normal vector angle condition. Seed point cloud points that satisfy the regional growth constraint rules are selected from the geological target point cloud data. Regional growth processing is performed starting from the seed point cloud points to obtain an initial slope cluster. Density clustering and edge detection processing are performed on the initial slope cluster in sequence. Combined with boundary constraints, the reconstructed slope free surface is obtained. Based on the top coordinates and borehole parameters of the target borehole in the geological target point cloud data, the geometric center of the explosive is determined, and the minimum Euclidean distance between the geometric center of the explosive and the free surface of the slope is calculated as the minimum resistance line of the target borehole.

2. The method according to claim 1, characterized in that, The preprocessing of the original 3D point cloud of the open-pit mine to obtain geological target point cloud data includes: An octree downsampling algorithm was used to perform spatial dimensionality reduction on the original 3D point cloud of the open-pit mine to obtain compressed point cloud data. For the compressed point cloud data, progressive morphological filtering is used to perform iterative operations of morphological dilation and erosion to remove point cloud data of non-geological targets and obtain point cloud data of geological targets. The non-geological targets include at least workers, pickup trucks and forklifts.

3. The method according to claim 1, characterized in that, The determination of the effective range of the slope zenith distance and the roughness of each point cloud point based on the geological target point cloud data includes: Based on the geological target point cloud data, the point cloud normal vector of each point cloud point is solved by the K-nearest neighbor algorithm, and the angle between the point cloud normal vector and the positive direction of the Z-axis of the spatial rectangular coordinate system is calculated to obtain the zenith distance distribution. The elbow rule is used to perform cluster analysis on the zenith distance distribution to determine the effective range of slope zenith distance. Based on the geological target point cloud data, the K nearest neighbor method is used to select K neighboring points for each point cloud point, and the standard deviation of the K neighboring points with respect to the elevation value is calculated as the roughness of the corresponding point cloud point.

4. The method according to claim 1, characterized in that, The region growth constraint rules include zenith distance constraint, roughness constraint, and normal vector angle constraint. Based on the effective range of the slope zenith distance, the roughness determination threshold, and the normal vector angle condition, a region growth constraint rule is configured, including: The zenith distance constraint is configured based on the effective range of the slope zenith distance. The zenith distance constraint is that the zenith distance of the point cloud to be grown is within the effective range of the slope zenith distance. The point cloud to be grown is the point cloud to be included in the growth range among the K nearest neighbors of the seed point, which has not yet been verified whether it meets the regional growth constraint rules. An initial seed point set is determined, consisting of all point cloud points in the geological target point cloud data. The roughness standard deviation of the initial seed point set is calculated based on the roughness of each point cloud point in the initial seed point set, and used as a roughness difference threshold. The roughness constraint is configured based on the roughness difference threshold, wherein the roughness difference between the point cloud point to be grown and the seed point cloud point does not exceed the roughness difference threshold. Based on the geological target point cloud data, the normal vector of each point cloud point is solved by the K-nearest neighbor algorithm. By analyzing the overall distribution characteristics of the normal vectors of all point cloud points, the normal vector angle threshold is determined. Based on the normal vector angle threshold, the normal vector angle constraint is configured. The normal vector angle constraint is that the angle between the normal vector of the point cloud point to be grown and the normal vector of the seed point cloud point does not exceed the normal vector angle threshold.

5. The method according to claim 1, characterized in that, Starting from the seed point cloud points, region growing is performed to obtain an initial slope cluster. Density clustering and edge detection are then performed on the initial slope cluster sequentially. Combined with boundary constraints, the reconstructed free slope surface is obtained, including: Using the selected seed point cloud points as the starting point for growth, the K nearest neighbors of each seed point cloud point are traversed. The K nearest neighbors that satisfy the region growth constraint rules are included in the growth range and updated as new seed point cloud points. The growth boundary is continuously expanded iteratively to generate discrete initial slope clusters. Based on the total number of point cloud data of the geological target, a small cluster threshold is set, and the initial slope clusters are screened one by one. Pseudo-slope clusters and small-scale noise clusters with a point cloud number less than the small cluster threshold are removed, and potential slope clusters with slope characteristics are retained. The DBSCAN clustering algorithm was used to perform density clustering on the potential slope clusters, and noise points corresponding to rock protrusions and blasting residues were removed to obtain the denoised slope clusters. Select the largest slope cluster with the most point cloud data from the denoised slope clusters, project all the point cloud points constituting the largest slope cluster onto the XY plane, and construct a two-dimensional image using the elevation value of each point cloud point as the gray value. The two-dimensional image is subjected to Gaussian filtering, and the Canny operator is used to perform edge detection on the Gaussian-filtered two-dimensional image. During the edge detection process, non-maximum suppression is performed to extract the top and bottom lines of the slope. Using the slope crest line and the slope bottom line as spatial boundaries, the coordinates of all candidate slope point cloud points in the denoised slope cluster are constrained, and candidate slope point cloud points whose coordinates are in the corresponding spatial interval between the slope crest line and the slope bottom line are retained to obtain the reconstructed slope free surface.

6. The method according to claim 5, characterized in that, The DBSCAN clustering algorithm is used to perform density clustering on the potential slope clusters, removing noise points corresponding to rock protrusions and blasting residues, resulting in denoised slope clusters, including: A preset number of sample point cloud points are randomly selected from the potential slope cluster. The average distance of the K nearest neighbors of each sample point cloud point is calculated. The mean of the average distances of the K nearest neighbors of all sample point cloud points is determined as the neighborhood radius of the DBSCAN algorithm. The neighborhood sphere volume is calculated based on the neighborhood radius. Based on the number of potential slope clusters, the number of point clouds in each potential slope cluster, the bounding box volume of the slope cluster, and the volume of the neighborhood sphere, the minimum cluster size of the DBSCAN algorithm is calculated. Based on the neighborhood radius and the minimum cluster size, DBSCAN clustering is performed to remove noise points corresponding to rock protrusions and blasting residues, resulting in denoised slope clusters.

7. The method according to claim 1, characterized in that, Based on the top coordinates and borehole parameters of the target point cloud data, the geometric center of the explosive is determined. The minimum Euclidean distance between the geometric center of the explosive and the free surface of the slope is calculated as the minimum resistance line of the target borehole, including: The top three-dimensional coordinates of the target borehole are extracted from the geological target point cloud data. Based on the top three-dimensional coordinates and preset borehole parameters, a cylindrical borehole three-dimensional model consistent with the actual borehole structure is constructed. The borehole parameters include borehole radius, borehole depth and plugging length. Based on the axial length and plugging length of the cylindrical borehole 3D model, the explosive section inside the target borehole is located, and the average elevation of the top and bottom of the explosive section is calculated. The three-dimensional coordinates of the explosive's geometric center are determined based on the average elevation and the horizontal coordinates of the borehole axis. Calculate the Euclidean distance from the geometric center of the explosive to each point cloud point on the free surface of the slope based on the three-dimensional coordinates, and select the minimum Euclidean distance as the minimum resistance line of the target blast hole.

8. A device for extracting the minimum resistance line in medium-deep hole blasting, characterized in that, include: The processing module is used to preprocess the original three-dimensional point cloud of the open-pit mine to obtain geological target point cloud data. The preprocessing includes at least spatial dimensionality reduction processing and non-geological target removal processing. The configuration module is used to determine the effective range of the slope zenith distance and the roughness of each point cloud point based on the geological target point cloud data, and to configure the regional growth constraint rules based on the effective range of the slope zenith distance, the judgment threshold of the roughness and the normal vector angle condition. The reconstruction module is used to filter seed point cloud points that meet the regional growth constraint rules in the geological target point cloud data, perform regional growth processing with the seed point cloud points as the starting point to obtain an initial slope cluster, and perform density clustering and edge detection processing on the initial slope cluster in sequence, and combine the boundary constraints to obtain the reconstructed slope free surface. The calculation module is used to determine the geometric center of the explosive based on the top coordinates and parameters of the target borehole in the geological target point cloud data, and to calculate the minimum Euclidean distance between the geometric center of the explosive and the free surface of the slope, which is used as the minimum resistance line of the target borehole.

9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

10. An electronic device comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

Images