An open-pit mine step line high-precision extraction method, device and electronic equipment

Abstract

The present application provides a kind of open-pit mine step line high-precision extraction method, device and electronic equipment, wherein the method comprises: using Gaussian filter and bilateral filter to carry out denoising processing to the target model of open-pit mine;The target model includes digital surface model, digital elevation model;The second derivative of the target model after denoising is calculated, and the initial discrete points of step line are extracted based on the second derivative;The initial discrete points of step line are corrected, and the step line discrete points obtained by correction are split to obtain the step line discrete point set of different height steps;Each step line discrete point in the step line discrete point set is sorted and smoothed, and the target step line is connected to obtain. By the present application, the open-pit mine step line is extracted with high precision and automation, and the problem of weak anti-interference ability and low extraction precision in the existing related technology is solved.

Description

Technical Field

[0001] This invention relates to the field of open-pit mine planning technology, and in particular to a method, apparatus and electronic equipment for high-precision extraction of open-pit mine bench lines. Background Technology

[0002] Mine safety production is a systematic project encompassing multiple stages, including production planning, production statistics, safety monitoring, and resource reserve accounting. These stages traditionally rely on a manual, manual work model, which suffers from drawbacks such as low efficiency, large errors, narrow coverage, difficulty in accessing high-risk areas, and tendency to deviate from actual terrain. The bench line, as the layered boundary of open-pit mine slopes, is a core topographical element connecting all these stages. Its extraction directly determines the accuracy of operational boundaries in production planning, the precision of mining and stripping volumes in production statistics, the reliability of slope early warning systems in safety monitoring, and the authenticity of resource reserve accounting boundaries, thus becoming the fundamental support for refined mine operations.

[0003] Existing open-pit mine bench line extraction technologies mainly fall into four categories: First, the on-site manual measurement method, which uses equipment such as total stations and GPS to collect bench feature points on-site. Its advantages are high accuracy, but its disadvantages include the need for personnel to enter the work area, extremely low efficiency, high safety risks, and inability to adapt to large-scale dynamic monitoring. Second, the point cloud extraction method, which extracts bench lines based on lidar point cloud data using segmentation algorithms. Its advantages are rich detail, but its disadvantages include large data volume, long processing time, and susceptibility to interference from vegetation and equipment. Third, the digital surface model or digital elevation model extraction method, which relies on UAV aerial survey digital surface model or digital elevation model data to extract bench lines through terrain analysis. Its advantages are convenient data acquisition and high processing efficiency, but its disadvantages include the potential for bench line anomalies caused by ground debris, vehicle barriers, ditches, overhead pipelines, etc., in complex scenarios. Fourth, the photogrammetric method, which extracts bench lines after generating a 3D model through multi-view image matching. Its advantages are low cost, but its disadvantages include significant influence from lighting and weather, and insufficient accuracy in complex terrain areas. In addition, deep learning segmentation methods have emerged in recent years, which use semantic segmentation models to extract step lines. The advantage is that they are highly automated, but the disadvantage is that they are highly dependent on training data and have poor generalization in mining scenarios.

[0004] The various shortcomings of the existing technologies mentioned above create a contradiction between the precise requirements for bench lines in all aspects of mine safety production and the deficiencies of the existing technologies. Therefore, developing a bench line extraction method with strong anti-interference capabilities, high extraction accuracy, and broad applicability to various mining scenarios has become the key to solving the pain points of refined operation in multiple aspects of mining.

[0005] There is currently no effective solution to the problems of weak anti-interference ability and low extraction accuracy in existing related technologies. Summary of the Invention

[0006] This invention provides a method, apparatus, and electronic equipment for high-precision extraction of bench lines in open-pit mines, which solves the defects of weak anti-interference ability and low extraction accuracy in existing related technologies.

[0007] In a first aspect, the present invention provides a method for high-precision extraction of bench lines in open-pit mines, comprising:

[0008] Gaussian filtering and bilateral filtering are used to denoise the target model of the open-pit mine; the target model includes a digital surface model and a digital elevation model.

[0009] Calculate the second derivative of the denoised target model, and extract the initial discrete points of the step line based on the second derivative;

[0010] The initial discrete points of the step line are corrected, and the corrected discrete points of the step line are split to obtain the set of discrete points of the step line for steps of different heights.

[0011] The discrete points of each step line in the set of discrete points are sorted and smoothed, and then connected to obtain the target step line.

[0012] According to the present invention, a high-precision extraction method for open-pit mine bench lines is provided, which uses Gaussian filtering and bilateral filtering to denoise the target model of the open-pit mine, including:

[0013] Gaussian filtering is used to perform the first round of denoising on the target model to filter out high-frequency random noise;

[0014] The target model is then subjected to secondary denoising using bilateral filtering to capture the strong edge contours of the open-pit mine steps caused by abrupt elevation changes.

[0015] According to the present invention, a high-precision extraction method for open-pit mine bench lines is provided, which calculates the second derivative of the denoised target model and extracts the initial discrete points of the bench lines based on the second derivative, including:

[0016] Calculate the second derivative of the denoised target model;

[0017] The second derivative is thresholded, and the initial discrete points of the step line are extracted based on the thresholding results.

[0018] According to the present invention, a high-precision extraction method for open-pit mine bench lines is provided, which involves thresholding the second derivative and extracting the initial discrete points of the bench lines based on the thresholding results, including:

[0019] The discrete points of the step line where the absolute value of the second derivative is greater than the threshold and the second derivative is greater than zero are the initial discrete points of the slope crest line.

[0020] The discrete points of the step line where the absolute value of the second derivative is greater than the threshold and the second derivative is less than zero are the initial discrete points of the slope bottom line.

[0021] According to the present invention, a high-precision extraction method for open-pit mine bench lines is provided, which corrects the initial discrete points of the bench lines, including:

[0022] For each initial discrete point of the step line, a horizontal search range and an elevation search range are set based on the initial discrete point of the step line, and the elevation search range is divided into multiple sub-elevation search ranges.

[0023] Count the number of valid target model pixels within each sub-elevation search range;

[0024] Based on the statistical results, the effective elevation range is determined. Among the effective target model pixels in the outermost layer of the effective elevation range, the pixel with the closest Euclidean distance to the initial discrete point of the step line is taken as the corrected discrete point of the step line.

[0025] According to the present invention, a high-precision extraction method for open-pit mine bench lines is provided, which determines the effective elevation range by combining statistical results, including:

[0026] The sub-elevation search range with the largest number of valid pixels that is greater than the preset standard is taken as the valid elevation range.

[0027] If no valid elevation range exists, the initial discrete points of the step line are discarded.

[0028] According to the present invention, a high-precision extraction method for open-pit mine bench lines is provided, which involves splitting the corrected bench line discrete points to obtain a set of bench line discrete points for benches of different heights, including:

[0029] Adaptive clustering based on elevation difference splits the discrete points of the step line into independent sets of discrete points for each step at different elevations.

[0030] According to the present invention, a high-precision extraction method for open-pit mine bench lines is provided, which sorts and smooths each discrete point of the bench line discrete point set, and connects them to obtain the target bench line, including:

[0031] Region growing based on nearest neighbor distance involves iteratively finding the nearest neighbor of each discrete point on the step line within the set of discrete points on the step line, forming an ordered set of discrete points on the step line from head to tail.

[0032] The sorted discrete points of the step line are smoothed using B-spline functions;

[0033] The target step line is obtained by sequentially connecting the discrete points of the step line.

[0034] Secondly, the present invention also provides a high-precision extraction device for open-pit mine bench lines, comprising:

[0035] A denoising module is used to denoise the target model of the open-pit mine using Gaussian filtering and bilateral filtering; the target model includes a digital surface model and a digital elevation model.

[0036] The extraction module is used to calculate the second derivative of the denoised target model and extract the initial discrete points of the step line based on the second derivative.

[0037] The correction module is used to correct the initial discrete points of the step line and split the corrected discrete points of the step line to obtain a set of discrete points of the step line for steps of different heights.

[0038] The integration module is used to sort and smooth each discrete point of the step line in the discrete point set, and connect them to obtain the target step line.

[0039] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the high-precision extraction method for open-pit mine bench lines as described in the first aspect above.

[0040] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the high-precision extraction method for open-pit mine bench lines as described in the first aspect above.

[0041] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the high-precision extraction method for open-pit mine bench lines as described in the first aspect above.

[0042] Compared with the prior art, the present invention has the following beneficial effects:

[0043] The high-precision extraction method for open-pit mine step lines provided by this invention uses techniques such as Gaussian filtering + bilateral filtering for noise reduction, second derivative threshold segmentation, multi-dimensional step line discrete point correction, elevation adaptive clustering, and smoothing to extract open-pit mine step lines with high precision and automation, solving the problems of weak anti-interference ability and low extraction accuracy in existing related technologies. Attached Figure Description

[0044] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0045] Figure 1 This is a flowchart of the high-precision extraction method for open-pit mine bench lines provided by the present invention;

[0046] Figure 2 This is a structural block diagram of the high-precision extraction device for open-pit mine bench lines provided by the present invention;

[0047] Figure 3 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0048] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0049] This invention provides a high-precision method for extracting bench lines in open-pit mines. Figure 1 This is a flowchart of the high-precision extraction method for open-pit mine bench lines provided by the present invention, as follows: Figure 1 As shown, the method includes the following steps:

[0050] Step S101: Gaussian filtering and bilateral filtering are used to denoise the target model of the open-pit mine; the target model includes a digital surface model and a digital elevation model.

[0051] Step S102: Calculate the second derivative of the denoised target model and extract the initial discrete points of the step line based on the second derivative;

[0052] Step S103: Correct the initial discrete points of the step line, and split the corrected discrete points of the step line to obtain a set of discrete points of the step line for steps of different heights.

[0053] Step S104: Sort and smooth each discrete point in the discrete point set of the step line, and connect them to obtain the target step line.

[0054] The following description uses a digital surface model as an example; the operation method for a digital elevation model is the same. During the acquisition process, the original mine digital surface model data is easily mixed with sensor noise from UAV aerial surveys, artifact noise from minor ground undulations, and local interference noise from operating equipment and accumulated materials. This noise can obscure the edge features of steps and interfere with the accuracy of subsequent step line extraction. Therefore, in this method, Gaussian filtering and bilateral filtering are first used to denoise the open-pit mine digital surface model, laying a solid data foundation for subsequent step line extraction. Then, the second derivative of the denoised digital surface model is calculated, and the initial discrete points of the step lines are extracted based on the second derivative. Since the initial discrete points of the step lines extracted in the above steps may be located at the bottom of ditches, the top of vehicle barriers, or on ground features, the position of the initial discrete points of the step lines is corrected in this method to improve their accuracy. Moreover, since the extracted step line discrete points are a mixture of discrete points from all height steps, they cannot be used directly. It is necessary to split the step line discrete points into independent sets of discrete points for each step line to ensure that the step lines of different height steps do not interfere with each other. Because of the significant height difference between adjacent steps, the discrete points of the step line on the same step are relatively close in height. Therefore, this method splits the discrete points of the step line into independent sets of discrete points for each step at different heights. Since each set of discrete points is internally unordered and cannot be directly connected to form a step line, and the discrete points may exhibit local jaggedness, this method sorts and smooths each discrete point in its set before connecting them to obtain the target step line. Through this process, high-precision and automated extraction of open-pit mine step lines can be achieved, solving the problems of weak anti-interference ability and low extraction accuracy in existing related technologies.

[0055] In some embodiments, step S101 involves denoising the target model of the open-pit mine using Gaussian filtering and bilateral filtering, including: performing a first round of denoising on the target model using Gaussian filtering to filter out high-frequency random noise; and performing a second round of denoising on the target model using bilateral filtering to capture the strong edge contours of the open-pit mine steps formed by abrupt changes in elevation.

[0056] In this embodiment, Gaussian filtering is used to perform the first round of denoising on the digital surface model, quickly filtering out high-frequency random noise. Then, bilateral filtering is used for secondary processing, which further weakens the residual noise while accurately capturing and preserving the strong edge contours formed by the abrupt elevation changes of the open-pit mine steps, laying a solid data foundation for the subsequent extraction of step lines.

[0057] In some embodiments, step S102, calculating the second derivative of the denoised target model and extracting the initial discrete points of the step line based on the second derivative, includes: calculating the second derivative of the denoised target model; performing threshold segmentation on the second derivative and extracting the initial discrete points of the step line based on the threshold segmentation result.

[0058] Specifically, the second derivative is thresholded, and the initial discrete points of the step line are extracted based on the thresholding results. The initial discrete points of the step line are: the absolute value of the second derivative is greater than the threshold and the second derivative is greater than zero. The initial discrete points of the step line are: the absolute value of the second derivative is greater than the threshold and the second derivative is less than zero.

[0059] In this embodiment, by thresholding the second derivative of the digital surface model, the most original step change features are directly captured. It also has a strong ability to extract step lines even for steps with complex terrain and narrow width, solving the problems of missed step line extraction and broken step lines caused by traditional methods that indirectly extract step lines through slope segmentation and morphological operations.

[0060] In some embodiments, step S103, correcting the initial discrete points of the step line, includes: for each initial discrete point of the step line, setting a horizontal search range and an elevation search range based on the initial discrete point of the step line, and dividing the elevation search range into multiple sub-elevation search ranges; counting the number of effective target model pixels in each sub-elevation search range; determining the effective elevation range based on the statistical results, and taking the pixel with the closest Euclidean distance to the initial discrete point of the step line among the effective target model pixels in the outermost layer of the effective elevation range as the corrected discrete point of the step line.

[0061] Specifically, the effective elevation range is determined by combining statistical results, including: taking the sub-elevation search range with the most effective pixels that is greater than the preset standard as the effective elevation range; if no effective elevation range exists, the initial discrete points of the step line are discarded.

[0062] For example, using the initial discrete point of the step line as a reference, a certain radius around the point is defined as the horizontal search range, and a certain elevation range above and below the point is defined as the elevation search range. This elevation search range is then divided into multiple sub-elevation search ranges. Next, the number of valid digital surface model pixels within each sub-elevation search range is counted. Then, the number of valid digital surface model pixels within each sub-elevation search range is compared. The sub-elevation search range with the largest number of valid pixels that exceeds a preset standard is taken as the valid elevation range. The pixel in the outermost layer of valid digital surface model pixels within this valid elevation range that has the closest Euclidean distance to the initial discrete point of the step line is taken as the corrected discrete point of the step line. If no valid elevation range exists, the initial discrete point of the step line is discarded.

[0063] In this embodiment, by correcting the position of discrete points of the step line in both horizontal and vertical directions, the problems of incorrectly extracting the step line to the top of the vehicle stop, the bottom of the ditch, overhead pipelines, the edge of ground features, and the vertical and horizontal oscillation of the step line are solved. The extracted step line fits the slope better and is more flush with the step platform, and the accuracy of the step line is greatly improved.

[0064] Based on this, the obtained discrete points of the step line are split to obtain discrete point sets of step lines for steps of different heights, including: adaptive clustering based on elevation difference to split the discrete points of the step line into independent discrete point sets of step lines for each height.

[0065] In this embodiment, adaptive elevation clustering is used to split the discrete points of the step line into independent discrete point sets for each step, which solves the problem of incorrect connection of the step line across different steps and improves the accuracy of the step line.

[0066] In some embodiments, step S104, which involves sorting and smoothing each discrete point in the discrete point set of the step line and connecting them to obtain the target step line, includes: region growing based on nearest neighbor distance, iteratively finding the nearest neighbor of each discrete point in the discrete point set of the step line to form an ordered discrete point set of the step line from head to tail; smoothing the sorted discrete points of the step line using a B-spline function; and sequentially connecting the discrete points of the step line to obtain the target step line.

[0067] This method achieves a technological breakthrough in addressing the pain points of open-pit mine bench lines, such as weak anti-interference capability, low extraction accuracy, and poor generalization to various mining scenarios. By providing accurate and complete bench line data, it significantly improves the management accuracy of mining enterprises in core aspects such as production planning, production statistics, safety monitoring, and resource reserve accounting, empowering the open-pit mining industry to achieve refined and intelligent operations. Furthermore, this method is adaptable to different types and regions of open-pit mines, possessing broad industry application value and providing a low-cost, highly adaptable solution for industry technological upgrades.

[0068] The present invention also provides a high-precision extraction device for open-pit mine bench lines. The high-precision extraction device for open-pit mine bench lines provided by the present invention is described below. The high-precision extraction device for open-pit mine bench lines described below can be referred to in correspondence with the high-precision extraction method for open-pit mine bench lines described above. Figure 2 This is a structural block diagram of the high-precision extraction device for open-pit mine bench lines provided by the present invention, as shown in the figure. Figure 2 As shown, the device includes:

[0069] The denoising module 201 is used to denoise the target model of the open-pit mine using Gaussian filtering and bilateral filtering; the target model includes a digital surface model and a digital elevation model.

[0070] Extraction module 202 is used to calculate the second derivative of the denoised target model and extract the initial discrete points of the step line based on the second derivative;

[0071] The correction module 203 is used to correct the initial discrete points of the step line and split the corrected discrete points of the step line to obtain a set of discrete points of the step line for steps of different heights.

[0072] The integration module 204 is used to sort and smooth each discrete point of the step line in the discrete point set, and connect them to obtain the target step line.

[0073] The raw digital surface model data of the mine is easily mixed with sensor noise from UAV aerial surveys, artifact noise from minor ground undulations, and local interference noise from operating equipment and accumulated materials during the acquisition process. This noise can mask the edge features of the steps and interfere with the accuracy of subsequent step line extraction. Therefore, the denoising module 201 first uses Gaussian filtering and bilateral filtering to denoise the digital surface model of the open-pit mine, laying a solid data foundation for subsequent step line extraction. Then, the extraction module 202 calculates the second derivative of the denoised digital surface model and extracts the initial discrete points of the step lines based on the second derivative. Since the initial discrete points of the step lines extracted in the above steps may be located at the bottom of ditches, the top of vehicle barriers, or on ground features, the correction module 203 corrects the position of the initial discrete points of the step lines to improve their accuracy. Moreover, since the extracted step line discrete points are a mixture of discrete points of all height steps, they cannot be used directly. It is necessary to split the step line discrete points into independent sets of discrete points for each step line to ensure that the step lines of different height steps do not interfere with each other. Because of the significant height difference between adjacent steps, the discrete points of the step line on the same step are relatively close in height. Therefore, the discrete points of the step line are split into independent sets of discrete points for each step at different heights. Since each set of discrete points is internally unordered and cannot be directly connected to form a step line, and the discrete points may exhibit local jagged edges, the integration module 204 sorts and smooths each discrete point in its set, connecting them to obtain the target step line. Through this process, the step lines of open-pit mines can be extracted with high precision and automatically, solving the problems of weak anti-interference ability and low extraction accuracy in existing related technologies.

[0074] Figure 3 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 3As shown, the electronic device may include: a processor 301, a communication interface 302, a memory 303, and a communication bus 304. The processor 301, communication interface 302, and memory 303 communicate with each other via the communication bus 304. The processor 301 can call logical instructions stored in the memory 303 to execute a high-precision extraction method for open-pit mine bench lines. This method includes:

[0075] Gaussian filtering and bilateral filtering were used to denoise the target model of the open-pit mine; the target model included a digital surface model and a digital elevation model.

[0076] Calculate the second derivative of the denoised target model, and extract the initial discrete points of the step line based on the second derivative;

[0077] The initial discrete points of the step line are corrected, and the corrected discrete points of the step line are split to obtain the set of discrete points of the step line for steps of different heights.

[0078] The discrete points of each step line in the discrete point set are sorted and smoothed, and then connected to obtain the target step line.

[0079] Furthermore, the logical instructions in the aforementioned memory 303 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, essentially, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0080] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute the high-precision extraction method for open-pit mine bench lines provided by the above methods, the method comprising:

[0081] Gaussian filtering and bilateral filtering were used to denoise the target model of the open-pit mine; the target model included a digital surface model and a digital elevation model.

[0082] Calculate the second derivative of the denoised target model, and extract the initial discrete points of the step line based on the second derivative;

[0083] The initial discrete points of the step line are corrected, and the corrected discrete points of the step line are split to obtain the set of discrete points of the step line for steps of different heights.

[0084] The discrete points of each step line in the discrete point set are sorted and smoothed, and then connected to obtain the target step line.

[0085] In another aspect, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the high-precision extraction method for open-pit mine bench lines provided by the methods described above, the method comprising:

[0086] Gaussian filtering and bilateral filtering were used to denoise the target model of the open-pit mine; the target model included a digital surface model and a digital elevation model.

[0087] Calculate the second derivative of the denoised target model, and extract the initial discrete points of the step line based on the second derivative;

[0088] The initial discrete points of the step line are corrected, and the corrected discrete points of the step line are split to obtain the set of discrete points of the step line for steps of different heights.

[0089] The discrete points of each step line in the discrete point set are sorted and smoothed, and then connected to obtain the target step line.

[0090] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0091] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0092] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for high-precision extraction of bench lines in open-pit mines, characterized in that, include: Gaussian filtering and bilateral filtering are used to denoise the target model of the open-pit mine; The target model includes a digital surface model and a digital elevation model; Calculate the second derivative of the denoised target model, and extract the initial discrete points of the step line based on the second derivative; The initial discrete points of the step line are corrected, and the corrected discrete points of the step line are split to obtain the set of discrete points of the step line for steps of different heights. The discrete points of each step line in the set of discrete points are sorted and smoothed, and then connected to obtain the target step line. The initial discrete points of the step line are corrected, including: For each initial discrete point of the step line, a horizontal search range and an elevation search range are set based on the initial discrete point of the step line, and the elevation search range is divided into multiple sub-elevation search ranges. Count the number of valid target model pixels within each sub-elevation search range; Based on the statistical results, the effective elevation range is determined. The pixel point with the closest Euclidean distance to the initial discrete point of the step line is taken from the outermost effective target model pixels of the effective elevation range as the corrected discrete point of the step line. The effective elevation range is determined based on statistical results, including: The sub-elevation search range with the largest number of valid pixels that is greater than the preset standard is taken as the valid elevation range. If no valid elevation range exists, the initial discrete points of the step line are discarded.

2. The high-precision extraction method for open-pit mine bench lines according to claim 1, characterized in that, The target model of the open-pit mine is denoised using Gaussian filtering and bilateral filtering, including: Gaussian filtering is used to perform the first round of denoising on the target model to filter out high-frequency random noise; The target model is denoised a second time by bilateral filtering to capture the strong edge contours of the open-pit mine steps caused by abrupt changes in elevation.

3. The high-precision extraction method for open-pit mine bench lines according to claim 1, characterized in that, Calculate the second derivative of the denoised target model, and extract the initial discrete points of the step line based on the second derivative, including: Calculate the second derivative of the denoised target model; The second derivative is thresholded, and the initial discrete points of the step line are extracted based on the thresholding results.

4. The high-precision extraction method for open-pit mine bench lines according to claim 3, characterized in that, The second derivative is thresholded, and the initial discrete points of the step line are extracted based on the thresholding results, including: The discrete points of the step line where the absolute value of the second derivative is greater than the threshold and the second derivative is greater than zero are the initial discrete points of the slope crest line. The discrete points of the step line where the absolute value of the second derivative is greater than the threshold and the second derivative is less than zero are the initial discrete points of the slope bottom line.

5. The high-precision extraction method for open-pit mine bench lines according to claim 1, characterized in that, The corrected discrete points of the step line are split to obtain discrete point sets of step lines for steps of different heights, including: Adaptive clustering based on elevation difference splits the discrete points of the step line into independent sets of discrete points for each step at different elevations.

6. The method for high-precision extraction of open-pit mine bench lines according to claim 1, characterized in that, The discrete points of each step line in the set of discrete points are sorted and smoothed, and then connected to obtain the target step line, including: Region growing based on nearest neighbor distance involves iteratively finding the nearest neighbor of each discrete point on the step line within the set of discrete points on the step line, forming an ordered set of discrete points on the step line from head to tail. The sorted discrete points of the step line are smoothed using B-spline functions; The target step line is obtained by sequentially connecting the discrete points of the step line.

7. A high-precision extraction device for open-pit mine bench lines, characterized in that, The method for high-precision extraction of open-pit mine bench lines according to any one of claims 1-6 includes: A denoising module is used to denoise the target model of the open-pit mine using Gaussian filtering and bilateral filtering; the target model includes a digital surface model and a digital elevation model. The extraction module is used to calculate the second derivative of the denoised target model and extract the initial discrete points of the step line based on the second derivative. The correction module is used to correct the initial discrete points of the step line and split the corrected discrete points of the step line to obtain a set of discrete points of the step line for steps of different heights. The integration module is used to sort and smooth each discrete point of the step line in the discrete point set, and connect them to obtain the target step line.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the high-precision extraction method for open-pit mine bench lines as described in any one of claims 1 to 6.

Images