A geological safety monitoring and early warning method and system

Abstract

The present invention relates to the field of image data processing technology, and in particular to a method and system for geological safety monitoring and early warning. The method comprises: obtaining an elevation map of a mining area and the current navigation path of a drone used for performing safety exploration in the mining area, and locating the path on the elevation map; recording pixel points located on the path as navigation pixel points, and obtaining multiple prediction lines of the navigation pixel points; obtaining predicted elevation data of the navigation pixel points; obtaining corrected elevation data of the navigation pixel points; obtaining the collapse coefficient of the navigation pixel points; and determining the dangerous areas of the mining area based on the collapse coefficient to achieve early warning based on geological safety monitoring. The present invention avoids the error transmission caused by single-direction prediction by making predictions based on the historical elevation data of the navigation pixel points and their multiple prediction lines, thereby improving the accuracy and reliability of elevation data predictions, and further improving the accuracy of safety monitoring and early warning of mines.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and in particular to a geological safety monitoring and early warning method and system. Background Art

[0002] As a key component of resource development and environmental protection, geological safety monitoring has developed into a multidisciplinary technical framework. Traditional methods rely on manual inspections, ground-based monitoring equipment, and remote sensing satellite data. While these methods can cover large areas, they suffer from limitations such as long monitoring cycles, high costs, and difficulty capturing subtle deformations. In recent years, unmanned aerial vehicle (UAV) technology, with its flexibility and high-resolution data acquisition capabilities, has gradually become a core tool for mining area monitoring, particularly in areas with complex terrain or difficult-to-reach locations.

[0003] In the mining industry, goaf management is a long-standing challenge. Backfilling techniques (such as paste filling and high-water-density material filling) are affected by variable geological conditions, material aging (such as sulfate attack and carbonization), and groundwater seepage, resulting in potential collapse risks in some areas. Statistics show that goaf collapses cause economic losses exceeding 10 billion yuan annually, resulting in damaged farmland, cracked buildings, and ground subsidence, seriously threatening the safety of surrounding residents and the ecological environment.

[0004] In current mine geological safety monitoring and early warning, the accuracy of elevation data prediction has always faced severe challenges. Existing methods often use a single direction to predict the elevation data of pixel points. This limitation makes the prediction results extremely susceptible to interference from pixel errors in a single direction. Due to the complex and changeable mining environment, data deviations are inevitable during the data collection process. The single-direction prediction model cannot effectively avoid such errors and it is difficult to fully capture the characteristics of terrain changes, resulting in low prediction accuracy. Monitoring plans based on low-precision prediction data cannot accurately determine the potential risk of surface collapse, making it difficult to achieve early identification of collapse hazards. At the same time, inaccurate prediction results can mislead the implementation of mine planning, land reclamation and other work, which may not only waste resources, but also bring unpredictable risks to the ecological environment, infrastructure and personnel safety in the mining area and surrounding areas, seriously restricting the effectiveness and reliability of mine geological safety monitoring and early warning work. Summary of the Invention

[0005] In order to solve the problem that the existing method uses a single direction to predict the elevation data of pixel points during the elevation data prediction process, which makes the prediction result extremely susceptible to interference from the errors of pixel points in a single direction, resulting in a decrease in prediction accuracy and further reducing the accuracy of mine safety monitoring and early warning, the present invention provides a geological safety monitoring and early warning method and system.

[0006] In a first aspect, the present invention provides a geological safety monitoring and early warning method, which adopts the following technical solutions:

[0007] A geological safety monitoring and early warning method includes: obtaining an elevation map of a mining area and a current navigation path of a drone used for performing safety exploration in the mining area, locating the path on the elevation map; recording pixel points on the path as navigation pixel points, recording any navigation pixel point as a target pixel point, obtaining multiple prediction lines of the target pixel point based on a vector formed by the target pixel point and the next navigation pixel point of the target pixel point, determining a prediction value of the target pixel point based on each prediction line based on the most recent historical elevation data of all pixels on each prediction line of the target pixel point, and taking the average of the prediction values as the target pixel point. The predicted elevation data of the pixel point; the normalized result of the difference between the predicted elevation data and the current navigation elevation data of the target pixel point is used as the correction requirement of the target pixel point; the current navigation elevation data of the target pixel point is corrected according to the size of the correction requirement to determine the corrected elevation data of the target pixel point; the collapse coefficient of the target pixel point is determined according to the difference between the corrected elevation data of all navigation pixel points and the most recent historical elevation data, as well as the difference between the corrected elevation data of the target pixel point and multiple historical elevation data; the dangerous area of the mining area is determined according to the collapse coefficient to realize geological safety monitoring and early warning.

[0008] The beneficial effects are as follows: by constructing multiple prediction lines based on the vector formed by the target pixel point and its next navigation pixel point, and comprehensively utilizing the historical elevation data on multiple prediction lines to predict the elevation of the target pixel point, the error influence caused by single-direction prediction is effectively avoided, and the accuracy of elevation data prediction is significantly improved; the elevation data of this navigation is targetedly corrected based on the correction need, reducing anomalies caused by environmental changes or data collection errors, and ensuring the timeliness and authenticity of the elevation data; by performing difference analysis between the corrected elevation data and multiple historical data, the collapse coefficient is accurately calculated, which can more comprehensively and sensitively capture the surface deformation trend and realize early risk identification and dynamic early warning of geological safety in mining areas; based on high-precision prediction and early warning, it assists the mining area in rationally demarcating dangerous areas, accurately guides safety exploration, mining area planning and land reclamation, effectively avoids environmental damage and resource waste, and ensures the safety of the ecological environment, infrastructure and personnel in and around the mining area; through the multi-line prediction and correction mechanism, it has strong robustness, can adapt to different geological conditions and dynamic changes, and improve the overall stability and reliability of the mining area geological safety monitoring and early warning system.

[0009] Furthermore, the elevation map is obtained by obtaining historical elevation data of all locations in the mining area through the National Mineral Resources Database and the National Administration of Surveying, Mapping and Geoinformation, and processing the historical elevation data using geographic information system software. The processing includes denoising, filling in gaps and cropping the area where the mining area is located, converting the processed historical elevation data into a contour map, and then using the smart map function to generate an elevation map of the mining area.

[0010] Furthermore, the geographic information system software adopts ArcGIS software.

[0011] Furthermore, the multiple prediction lines are obtained by taking the opposite direction of the vector formed by the target pixel point and the next navigation pixel point of the target pixel point as the prediction angle direction, and dividing the circular area with the target pixel point as the center into multiple parts based on the prediction angle direction, and recording the connecting lines of all two adjacent parts as prediction lines.

[0012] Furthermore, the predicted value satisfies:

[0013] Where, Navigation pixels Based on the The predicted value of the prediction line, and Navigation pixels No. The first prediction line Pixels and The most recent historical elevation data of pixels, Navigation pixels In the The most recent historical elevation data of the next pixel on the prediction line, Navigation pixels No. The number of all pixels on the predicted line.

[0014] The beneficial effects are: by calculating the average value of the elevation difference between adjacent pixel points on the prediction line, the overall elevation change trend of the prediction line is reflected, avoiding the sensitivity of the simple averaging method to local outliers, and helping to more accurately simulate the continuous changes in terrain; by utilizing the difference of historical elevation data on the prediction line and the historical elevation of the next navigation pixel point, the spatial correlation along the prediction line is comprehensively considered to enhance the stability and scientificity of the prediction, thereby improving the reliability of the prediction results.

[0015] Furthermore, the corrected elevation data satisfies:

[0016] Where, Navigation pixels Corrected elevation data, Navigation pixels The correction demand, is the preset correction threshold, Navigation pixels The predicted elevation data, Navigation pixels The altitude data of this voyage, Navigation pixels The similarity between the predicted elevation data of the local sequence and the most recent historical elevation data, Navigation pixels The similarity between the current navigation elevation data and the most recent historical elevation data of the local sequence.

[0017] The beneficial effects are: when the correction demand is lower than the threshold, the measured elevation data is directly used as the correction elevation data to ensure data robustness; otherwise, the predicted data in the local sequence and the current data are weightedly fused according to the similarity weights with the historical data, so that the correction results are more reasonable and dynamically adaptable to actual observations; through the segmented correction strategy, differentiated processing can be made for different degrees of correction needs, effectively responding to complex and changeable geological environments, and reducing the negative impact of outliers on the overall prediction of the system; more accurate correction elevation data can significantly improve the calculation accuracy of subsequent collapse coefficients, thereby enhancing the ability to identify early geological risks in mining areas and the early warning effect.

[0018] Furthermore, the similarity is a normalized result of the Pearson correlation coefficient.

[0019] Furthermore, the collapse coefficient satisfies:

[0020] Where, Navigation pixels The collapse coefficient, Navigation pixels The difference between the revised elevation data and the most recent historical elevation data, is the average of the differences between the corrected elevation data of all navigation pixels and the most recent historical elevation data. Navigation pixels The average of the differences between the corrected elevation data and multiple historical elevation data, for function.

[0021] The beneficial effects are: the collapse coefficient combines the single point elevation change with the average of the elevation change of the entire navigation path, as well as the historical multiple elevation changes of the point, comprehensively reflects the relationship between local and overall terrain changes, and improves the scientificity and accuracy of collapse risk assessment; The function performs nonlinear mapping on the normalized differences, which can smoothly convert the elevation change differences into risk indicators in the range of 0 to 1, enhancing the sensitivity and discrimination of potential collapse risks, and facilitating the accurate delineation of dangerous areas; by comparing the differences between current and historical elevation data, it can effectively capture the dynamic changes of surface subsidence and collapse, and support real-time monitoring and early warning of the geological safety situation in mining areas.

[0022] Furthermore, the determination of dangerous areas in the mining area includes: clustering the navigation pixel points based on the collapse coefficients of all navigation pixel points to obtain two first clusters, obtaining the mean collapse coefficient in each first cluster, and recording the pixel points in the first cluster with the largest mean collapse coefficient as collapsed pixel points; clustering the collapsed pixel points based on the spatial distance between all collapsed pixel points to obtain multiple second clusters, obtaining the mean collapse coefficient in each second cluster, and in response to the mean collapse coefficient in any second cluster being greater than a preset danger threshold, recording the mining area location where the second cluster is located as a dangerous area; in response to the mean collapse coefficient in any second cluster being less than a preset mild danger threshold, recording the mining area location where the second cluster is located as a mild danger area; otherwise, recording the mining area location where the second cluster is located as a moderate danger area.

[0023] The beneficial effects are: through the two-time clustering method, high collapse risk points are first identified, and then high-risk areas are subdivided, which can more accurately divide areas of different risk levels in the mining area and realize refined identification of dangerous areas; the mean value of the collapse coefficient is combined with the preset threshold to quantify the risk level of different areas, and the three-level warning of danger, moderate danger and mild danger is used to facilitate the implementation of targeted safety management measures; the clear division of each risk level area in the mining area helps mine managers give priority to high-risk areas, rationally allocate resources, and improve monitoring efficiency and management effects.

[0024] In a second aspect, the present invention provides a geological safety monitoring and early warning system, which adopts the following technical solutions:

[0025] A geological safety monitoring and early warning system includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned geological safety monitoring and early warning method is implemented.

[0026] By adopting the above technical solution, a computer program based on the above geological safety monitoring and early warning method is generated and stored in a memory to be loaded and executed by a processor, thereby making a terminal device based on the memory and the processor for easy use.

[0027] The present invention has the following technical effects:

[0028] (1) The method of predicting elevation data in a single direction is changed. Multiple prediction lines are obtained based on the vector formed by the target pixel point and the next navigation pixel point. The predicted elevation data of the target pixel point is determined by the average of the predicted values of multiple prediction lines. This reduces the interference of the error of the pixel point in a single direction, captures the terrain change characteristics more comprehensively, and effectively improves the accuracy of elevation data prediction.

[0029] (2) Accurate predicted elevation data can enable monitoring programs to accurately determine the potential risk of surface collapse, achieve early identification of potential collapse hazards, and avoid formulating monitoring programs based on low-precision elevation data, which may lead to misjudgment.

[0030] (3) Accurate prediction results avoid misleading mining area planning, land reclamation and other work due to low-precision prediction data, reduce resource waste, and reduce the risks to the mining area and surrounding ecological environment, infrastructure and personnel safety, and improve the effectiveness and reliability of mining area geological safety monitoring and early warning work, which is conducive to the sustainable development of the mining area.

[0031] (4) Considering the complex and changeable environment of mining areas and the fact that data collection is prone to deviation, the present invention can effectively avoid the impact of data deviation by processing and correcting elevation data multiple times, and can still provide accurate monitoring and early warning in complex environments, and has strong environmental adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 This is a flow chart of a method based on geological safety monitoring and early warning in an embodiment of the present invention.

[0033] Figure 2 This is a schematic diagram of the predicted angle direction of step S2 in a geological safety monitoring and early warning method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work shall fall within the scope of protection of the present invention.

[0035] The embodiment of the present invention discloses a geological safety monitoring and early warning method, referring to Figure 1 , including steps S1 to S6:

[0036] S1: Obtain an elevation map of the mining area and a navigation path of a drone used for performing safety exploration in the mining area, and locate the path on the elevation map.

[0037] Specifically, the elevation map is obtained as follows:

[0038] Historical elevation data for all locations within the mining area was obtained through the National Mineral Resources Database and the National Administration of Surveying, Mapping and Geoinformation. This data was processed using geographic information system software, including denoising, void filling, and cropping of the mining area. The processed historical elevation data was converted into a contour map, and an elevation map of the mining area was generated using the smart map function.

[0039] Implementers can choose the correction method for scenes in the contour map that require planar correction based on the specific implementation situation. For example, they can use high-precision control points (GCPs) to correct the contour map.

[0040] Specifically, the geographic information system software adopts ArcGIS software.

[0041] In another embodiment, the geographic information system software adopts QGIS software.

[0042] S2: Recording pixel points on the path as navigation pixel points, and obtaining multiple prediction lines of the navigation pixel points.

[0043] It should be noted that when drones use monitoring equipment above the mining area to obtain real-time elevation data for each area they pass through, they will be affected by environmental factors, resulting in inaccurate elevation data. Therefore, it is necessary to predict the elevation data on the drone's navigation path based on the drone's navigation path. However, when analyzing the elevation data of the mining area corresponding to its navigation path based on the drone's historical elevation data, if only the historical elevation data in the direction of the drone's advance is analyzed, the connection between the elevation data in the forward direction and the elevation data of the surrounding area will be ignored. Moreover, if there is an error in the historical elevation data in the forward direction, it will have a significant impact on the predicted value. Therefore, the pixels within a certain range around the navigation pixel point are delineated, and then multiple prediction lines are obtained to facilitate the subsequent calculation of the predicted value.

[0044] The pixel points on the path are recorded as navigation pixel points, any navigation pixel point is recorded as a target pixel point, and multiple prediction lines of the target pixel point are obtained based on a vector formed by the target pixel point and the next navigation pixel point of the target pixel point.

[0045] Specifically, the multiple prediction lines are obtained as follows:

[0046] like Figure 2 As shown in the figure, the opposite direction of the vector formed by the target pixel and the next navigation pixel of the target pixel is used as the predicted angle direction (if the target pixel is the last navigation pixel, the previous navigation pixel of the target pixel is selected). Based on the predicted angle direction, the circular area with the target pixel as the center is divided into multiple parts, and the connecting lines of all two adjacent parts are recorded as predicted lines.

[0047] Implementers can set the angle of equal division according to the specific implementation situation. For example, based on the predicted angle direction, rotate counterclockwise around the center of the circle, and divide once every 90°, eventually obtaining 4 predicted lines.

[0048] If the navigation pixel point is close to the edge of the elevation map and cannot form a complete circle, set the angle of equalization according to the actual situation and obtain the predicted line (at least The circle, that is, the right-angled sector, is divided every 30°, and the two sides of the right-angled sector are also used as prediction lines, and finally 4 prediction lines are obtained).

[0049] S3: Obtain the predicted elevation data of the navigation pixel points.

[0050] It should be noted that a joint evaluation is performed based on the multi-directional elevation data in a circular range around the target pixel point, that is, the navigation pixel point on the UAV's navigation path is located, and based on the change trend of the corresponding elevations of all pixels in different directions around this navigation pixel point, the elevation data of the navigation pixel point at the center of the circle is predicted.

[0051] According to the latest historical elevation data of all pixels on each prediction line of the target pixel point, the prediction value of the target pixel point based on each prediction line is determined, and the average of the prediction values is used as the predicted elevation data of the target pixel point.

[0052] Specifically, the predicted value satisfies:

[0053] ;

[0054] Where, Navigation pixels Based on the The predicted value of the prediction line, and Navigation pixels No. The first prediction line Pixels and The most recent historical elevation data of pixels, Navigation pixels In the The most recent historical elevation data of the next pixel on the prediction line, Navigation pixels No. The number of all pixels on the predicted line.

[0055] in, Indicates the The average change trend of the most recent historical elevation data of the pixel points on the prediction line. The larger the value, the more drastic the terrain change in the direction of the prediction line, and the larger the prediction value; vice versa. Indicates the basic reference value for prediction. The larger the value, the larger the predicted value; and vice versa.

[0056] S4: Obtain the corrected elevation data of the navigation pixel points.

[0057] It's important to note that in the complex environment of mining areas, elevation data collected by drones can be affected by factors such as sensor errors, terrain obstruction, and electromagnetic interference, resulting in data bias. Relying solely on predicted elevation data or data collected during a single flight cannot accurately reflect the actual terrain. Therefore, it is necessary to dynamically optimize elevation data by combining predicted and measured data (the elevation data from the current flight) and incorporating a correction requirement as a basis for judgment. This provides more reliable data support for geological safety in mining areas.

[0058] A normalized result of the difference between the predicted elevation data and the current navigation elevation data of the target pixel point (exemplarily, the normalization adopts standard normalization) is used as the correction requirement of the target pixel point; based on the size of the correction requirement, the current navigation elevation data of the target pixel point is corrected to determine the corrected elevation data of the target pixel point.

[0059] Specifically, the corrected elevation data satisfies:

[0060] ;

[0061] Where, Navigation pixels Corrected elevation data, Navigation pixels The correction demand, is the preset correction threshold, Navigation pixels The predicted elevation data, Navigation pixels The altitude data of this voyage, Navigation pixels The similarity between the predicted elevation data of the local sequence and the most recent historical elevation data, Navigation pixels The similarity between the current navigation elevation data and the most recent historical elevation data of the local sequence.

[0062] The implementer can set the correction threshold and the length of the local sequence according to the specific implementation situation. For example, the correction threshold is 0.6; the length of the local sequence is based on the navigation pixel points. As the center, 25 navigation pixels are selected on each side to form the navigation pixel points Belong to the local sequence, if the navigation pixel If the number of navigation pixels on either side is less than 25, fill it up from the other side to make the number of pixels in the sequence reach 51.

[0063] Among them, when When , it means that the difference between this navigation data and the predicted data is small, and the navigation pixel point The altitude data of this navigation is accurate, so it is directly As the corrected elevation data, no additional adjustments are made; when When , it indicates that there are significant anomalies in the navigation data, which may be caused by measurement errors. At this time, the data is corrected using the weighted average formula. Weights are assigned according to the similarity between the predicted data and the historical data, and the similarity between the measured data and the historical data, to balance the impact of the two on the final result and improve the accuracy of the corrected elevation data.

[0064] Specifically, the similarity is a normalized result of the Pearson correlation coefficient.

[0065] The normalization is due to the fact that the interval of the Pearson correlation is Therefore, add 1 and divide 2 to the Pearson correlation coefficient and map the Pearson correlation coefficient to between.

[0066] S5: Obtain the collapse coefficient of the navigation pixel point.

[0067] It should be noted that some areas of the mining area may collapse due to loose filling materials, etc. Therefore, it is necessary to calculate the collapse coefficient of the navigation pixel points in the mining area, so as to obtain the area that is about to collapse and carry out timely warning, rectification and personnel transfer. The elevation data changes in the area that is about to collapse on a time scale will be greater than those in other normal areas around it. The collapse of the mining area is a long-term geological change process. The short-term elevation data changes predicted by a single time can only reflect the changes in a certain location within a shorter time evolution interval, and cannot reflect the complete situation of the elevation data changes in this area. Therefore, this step calculates the collapse coefficient of the navigation pixel point based on the changes in multiple historical elevation data of the navigation pixel point.

[0068] The collapse coefficient of the target pixel point is determined based on the difference between the corrected elevation data of all navigation pixels and the most recent historical elevation data, as well as the difference between the corrected elevation data of the target pixel point and multiple historical elevation data.

[0069] Specifically, the collapse coefficient satisfies:

[0070] ;

[0071] Where, Navigation pixels The collapse coefficient, Navigation pixels The difference between the revised elevation data and the most recent historical elevation data, is the average of the differences between the corrected elevation data of all navigation pixels and the most recent historical elevation data. Navigation pixels The average of the differences between the corrected elevation data and multiple historical elevation data, for function.

[0072] Implementers can set the number of historical elevation data based on specific implementation circumstances, for example, historical elevation data within the last 12 months.

[0073] in, Indicates navigation pixels The larger the recent elevation change amplitude is compared with the overall recent elevation change amplitude, the more likely the navigation pixel point is in the process of collapse change. The bigger the better; vice versa. Indicates navigation pixels The magnitude of the recent elevation change compared to the long-term trend of the elevation change of the pixel point. The larger the value, the more the recent change deviates from the long-term trend, suggesting a potential collapse risk. The larger the value, the higher the value; vice versa. The function maps the result of the calculation to The interval is used to smooth the values and avoid interference from extreme values. The center point of the function moves from 0.5 to 0, making It can distinguish between elevation increases (negative values) and decreases (positive values). The function of 2 is to scale the output range to , enhancing the expressiveness of abnormal changes.

[0074] S6: Determine the dangerous areas of the mining area based on the collapse coefficient and implement early warning based on geological safety monitoring.

[0075] It's important to note that subsidence and collapse in mining areas often manifest themselves as blocky areas in planar space. This means that areas prone to collapse may experience a greater height drop than the surrounding normal areas. Therefore, navigation pixels with drastic elevation changes appear clustered together in localized areas on the elevation map, rather than scattered. These navigation pixels also tend to have relatively similar collapse coefficients. Therefore, this step locates areas of mining areas prone to collapse based on the spatial distribution of the collapse coefficients of these navigation pixels.

[0076] Specifically, the determination of dangerous areas in the mining area includes:

[0077] Based on the collapse coefficients of all navigation pixels, the navigation pixels are clustered to obtain two first clusters, the mean of the collapse coefficients in each first cluster is obtained, and the pixel in the first cluster with the largest mean of the collapse coefficient is recorded as a collapsed pixel;

[0078] Based on the spatial distance between all collapsed pixel points, the collapsed pixel points are clustered to obtain multiple second clusters, and the mean collapse coefficient in each second cluster is obtained. In response to the mean collapse coefficient in any second cluster being greater than a preset danger threshold, the mining area where the second cluster is located is recorded as a dangerous area; in response to the mean collapse coefficient in any second cluster being less than a preset mild danger threshold, the mining area where the second cluster is located is recorded as a mild danger area; otherwise, the mining area where the second cluster is located is recorded as a moderate danger area.

[0079] Implementers can set the number of second clusters, the danger threshold, and the mild danger threshold according to specific implementation conditions. For example, the number of second clusters is 5, the danger threshold is 0.8, and the mild danger threshold is 0.4.

[0080] An embodiment of the present invention further discloses a geological safety monitoring and early warning system, comprising a processor and a memory, wherein the memory stores computer program instructions. When the computer program instructions are executed by the processor, a geological safety monitoring and early warning method according to the present invention is implemented.

[0081] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0082] The above are all preferred embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A geological safety monitoring and early warning method, characterized in that: include: Obtaining an elevation map of the mining area and a current navigation path of a drone used for performing safety exploration in the mining area, and locating the path on the elevation map; The pixel points on the path are recorded as navigation pixel points, and any navigation pixel point is recorded as a target pixel point. Based on the vector formed by the target pixel point and the next navigation pixel point of the target pixel point, multiple prediction lines of the target pixel point are obtained, including: The opposite direction of the vector formed by the target pixel and the next navigation pixel of the target pixel is used as the predicted angle direction. Based on the predicted angle direction, the circular area with the target pixel as the center is divided into multiple parts, and the connecting lines of all two adjacent parts are recorded as predicted lines; According to the most recent historical elevation data of all pixels on each prediction line of the target pixel, the predicted value of the target pixel based on each prediction line is determined to satisfy the relationship: ； Where, Navigation pixels Based on the The predicted value of the prediction line, and Navigation pixels No. The first prediction line Pixels and The most recent historical elevation data of pixels, Navigation pixels In the The most recent historical elevation data of the next pixel on the prediction line, Navigation pixels No. The number of all pixels on the prediction line; The average of the predicted values is used as the predicted elevation data of the target pixel point; the normalized difference between the predicted elevation data and the current navigation elevation data of the target pixel point is used as the correction requirement of the target pixel point; and the current navigation elevation data of the target pixel point is corrected according to the correction requirement to determine the corrected elevation data of the target pixel point. Determine the collapse coefficient of the target pixel point based on the difference between the corrected elevation data of all navigation pixels and the most recent historical elevation data, as well as the difference between the corrected elevation data of the target pixel point and multiple historical elevation data; The dangerous areas of the mining area are determined according to the collapse coefficient to realize geological safety monitoring and early warning.

2. A geological safety monitoring and early warning method according to claim 1, characterized in that: The elevation map is obtained as follows: Historical elevation data for all locations within the mining area was collected through the National Mineral Resources Database and the National Administration of Surveying, Mapping and Geoinformation. This data was then processed using geographic information system software, including denoising, void filling, and cropping of the mining area. The processed historical elevation data was converted into a contour map, and an elevation map of the mining area was generated using the smart map function.

3. A geological safety monitoring and early warning method according to claim 2, characterized in that: The geographic information system software adopts ArcGIS software.

4. The geological safety monitoring and early warning method according to claim 1 is characterized in that: The corrected elevation data satisfies: ； Where, Navigation pixels Corrected elevation data, Navigation pixels The correction demand, is the preset correction threshold, Navigation pixels The predicted elevation data, Navigation pixels The altitude data of this voyage, Navigation pixels The similarity between the predicted elevation data of the local sequence and the most recent historical elevation data, Navigation pixels The similarity between the current navigation elevation data and the most recent historical elevation data of the local sequence.

5. A geological safety monitoring and early warning method according to claim 4, characterized in that: The similarity is the result of normalization of the Pearson correlation coefficient.

6. The geological safety monitoring and early warning method according to claim 1 is characterized in that: The collapse coefficient satisfies: ； Where, Navigation pixels The collapse coefficient, Navigation pixels The difference between the revised elevation data and the most recent historical elevation data, is the average of the differences between the corrected elevation data of all navigation pixels and the most recent historical elevation data. Navigation pixels The average of the differences between the corrected elevation data and multiple historical elevation data, for function.

7. The geological safety monitoring and early warning method according to claim 1 is characterized in that: Determining the dangerous areas in the mining area includes: Based on the collapse coefficient of all navigation pixel points, the navigation pixel points are clustered to obtain two first clusters, the mean collapse coefficient in each first cluster is obtained, and the pixel points in the first cluster with the largest mean collapse coefficient are recorded as collapsed pixel points; based on the spatial distance between all collapsed pixel points, the collapsed pixel points are clustered to obtain multiple second clusters, the mean collapse coefficient in each second cluster is obtained, and in response to the mean collapse coefficient in any second cluster being greater than a preset danger threshold, the mine location where the second cluster is located is recorded as a dangerous area; in response to the mean collapse coefficient in any second cluster being less than a preset mild danger threshold, the mine location where the second cluster is located is recorded as a mild danger area; otherwise, the mine location where the second cluster is located is recorded as a moderate danger area.

8. A geological safety monitoring and early warning system, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a geological safety monitoring and early warning method according to any one of claims 1 to 7 is implemented.

Images