Geological exploration data visualization method and device, equipment and medium

Abstract

The invention provides a geological exploration data visualization method and device, equipment and a medium, and belongs to the technical field of data visualization, and the method comprises the steps: carrying out the anomaly detection of each first exploration data point in target geological exploration volume data, and obtaining the anomaly score of each first exploration data point; screening the first exploration data points to obtain a plurality of screened second exploration data points; determining a hyper-parameter of a target clustering algorithm based on the abnormal score of each second exploration data point; performing clustering processing on the features corresponding to the second exploration data points based on the target clustering algorithm after hyper-parameter determination to obtain an initial clustering result; performing second denoising processing on the initial clustering result to obtain a target clustering result; and carrying out projection visualization processing on the geological exploration data based on the clusters in the target clustering result. According to the method, the noise processing accuracy of the geological exploration volume data can be improved, so that the visualization effect of the geological exploration data is improved.

Description

Technical Field

[0001] This application belongs to the field of data visualization technology, and more specifically, relates to a geological exploration data visualization method, device, equipment, and medium. Background Technology

[0002] Geological exploration data visualization is a key technology for stratigraphic structure analysis and geological anomaly identification, and its visualization effect directly affects the accuracy of exploration decisions. In actual exploration processes, the collected geological exploration body data is easily affected by factors such as instrument malfunction, environmental electromagnetic interference, and surface clutter, resulting in a large amount of noisy data.

[0003] Existing technologies often use single filtering or fixed threshold removal to process noisy data. However, this often results in excessive removal of normal signals or a large amount of residual noise, leading to problems such as blurred geological boundaries and low anomaly identification in the final visualized images. This makes it difficult to meet the application requirements of high-precision geological exploration. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus, equipment, and medium for visualizing geological exploration data, so as to improve the accuracy of noise processing of geological exploration data and thus improve the visualization effect of geological exploration data.

[0005] A first aspect of this application provides a method for visualizing geological exploration data, including: Acquire data on the target geological exploration bodies in the target exploration area; Anomaly detection is performed on each first exploration data point in the target geological exploration body data based on the isolated forest algorithm, and anomaly scores are obtained for each first exploration data point. Based on the anomaly scores of each first exploration data point, the first exploration data points are filtered to obtain multiple filtered second exploration data points, so as to achieve the first denoising process of the target geological exploration body data. Feature extraction was performed on multiple second exploration data points to obtain the geological features corresponding to each second exploration data point; Based on the feature distribution of geological characteristics of each second exploration data point, a target clustering algorithm that is compatible with the second exploration data points is determined. The hyperparameters of the target clustering algorithm are determined based on the anomaly scores of each second exploration data point; Based on the target clustering algorithm after the hyperparameters are determined, the features corresponding to each second exploration data point are clustered to obtain the initial clustering results; A second denoising process is performed on clusters in the initial clustering results whose data volume of the third exploration data points is less than a preset threshold, to obtain the target clustering results; the target clustering results contain multiple clusters, and different clusters represent different geological features in the target exploration area; Based on the clusters in the target clustering results, the geological exploration data is projected and visualized to achieve the visualization of geological exploration data in the target exploration area.

[0006] A second aspect of this application provides a geological exploration data visualization device, comprising: The data acquisition module is used to acquire data on the target geological exploration bodies in the target exploration area; Anomaly detection module is used to perform anomaly detection on each first exploration data point in the target geological exploration body data based on the isolated forest algorithm, and obtain anomaly score for each first exploration data point; The first denoising module is used to filter each first exploration data point based on the anomaly score of each first exploration data point to obtain multiple second exploration data points after filtering, so as to achieve the first denoising process of the target geological exploration body data. The feature extraction module is used to extract features from multiple second exploration data points to obtain the geological features corresponding to each second exploration data point. The clustering algorithm determination module is used to determine the target clustering algorithm that is compatible with the second exploration data points based on the feature distribution of the geological characteristics of each second exploration data point; The parameter determination module is used to determine the hyperparameters of the target clustering algorithm based on the anomaly scores of each second exploration data point; The first clustering module is used to perform clustering processing on the features corresponding to each second exploration data point based on the target clustering algorithm after the hyperparameters are determined, and to obtain the initial clustering results; The second clustering module is used to perform a second denoising process on clusters in the initial clustering results whose data volume of the third exploration data points is less than a preset threshold, to obtain the target clustering result; the target clustering result contains multiple clusters, and different clusters represent different geological features in the target exploration area; The visualization module is used to perform projection visualization processing of geological exploration data based on clusters in the target clustering results, so as to realize the visualization of geological exploration data in the target exploration area.

[0007] A third aspect of this application provides an electronic device, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of the above-described geological exploration data visualization method.

[0008] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the above-described geological exploration data visualization method.

[0009] The beneficial effects of the geological exploration data visualization method, apparatus, equipment, and medium provided in this application embodiment are as follows: This application's embodiment achieves optimization through a two-step precise denoising process. First, the application uses the isolated forest algorithm to detect anomalies in the original geological exploration body data, obtaining anomaly scores for each data point and filtering them to complete the first denoising step. Second, geological features are extracted, and a suitable target clustering algorithm is determined. Hyperparameters are set using the existing anomaly scores. Since anomaly scores directly reflect the validity of the data, setting hyperparameters based on anomaly scores makes the target clustering algorithm more suitable for the geological data features. Then, clusters with data volume less than a preset threshold in the clustering results undergo a second denoising process to further purify the data. Compared to the existing technology that uses single filtering or fixed threshold removal to process noisy data, this application's embodiment can improve the accuracy of noise processing, resulting in improved final visualization effects. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments of this application, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 A flowchart illustrating a geological exploration data visualization method provided in an embodiment of this application; Figure 2 This is a structural block diagram of a geological exploration data visualization device provided in an embodiment of this application; Figure 3 This is a schematic block diagram of an electronic device provided in an embodiment of this application. Detailed Implementation

[0012] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.

[0013] To make the objectives, technical solutions, and advantages of this application clearer, the following description will be provided in conjunction with the accompanying drawings and specific embodiments.

[0014] Please refer to Figure 1 , Figure 1This is a flowchart illustrating a geological exploration data visualization method provided in an embodiment of this application. The method can be executed by an electronic device and may include: S101-S109.

[0015] S101: Obtain target geological exploration body data for the target exploration area.

[0016] In this embodiment, the target exploration area refers to a specific spatial range within which geological resource exploration and geological structure analysis are to be carried out. Target geological exploration body data refers to a three-dimensional spatial exploration data volume covering the target exploration area; it is the original core data of geological exploration and may contain physical property parameters at different depths and locations underground. Common types of target geological exploration body data include three-dimensional seismic exploration body data (including seismic wave amplitude at each sampling point) and three-dimensional electrical resistivity tomography (EMT) exploration body data (including resistivity or polarizability at each sampling point). The data structure can be a three-dimensional grid, with each grid corresponding to exploration parameters at a spatial location. S102: Based on the isolated forest algorithm, anomaly detection is performed on each first exploration data point in the target geological exploration body data to obtain the anomaly score of each first exploration data point.

[0017] In this embodiment, the first exploration data point refers to the smallest data unit in the target geological exploration body data, corresponding to a specific grid point in three-dimensional space, containing the original exploration parameters at that location (such as the amplitude value of a single seismic trace sampling point, the resistivity value of a single electrical resistivity measurement point). The Isolation Forest algorithm is an unsupervised anomaly detection algorithm. Its core principle is to repeatedly segment the data space by randomly selecting feature dimensions and randomly setting segmentation thresholds. Anomalies, being more isolated in space, are segmented more quickly. The anomaly score of each first exploration data point refers to the quantitative index output by the Isolation Forest algorithm used to measure the degree of anomaly of the data point, typically ranging from 0 to 1. The closer the score is to 1, the more likely the data point is to be an anomaly, i.e., noise (such as instrument noise, environmental interference, or invalid points caused by data acquisition errors); the closer the score is to 0, the more likely the data point is to be a valid high-quality data point. The hyperparameters of the Isolation Forest algorithm can be based on the algorithm's default parameter settings or set empirically. In this embodiment, the processing procedure of the Isolation Forest algorithm will not be elaborated further.

[0018] S103: Based on the anomaly scores of each first exploration data point, the first exploration data points are filtered to obtain multiple filtered second exploration data points, so as to achieve the first denoising process of the target geological exploration body data.

[0019] In this embodiment, each first exploration data point can be filtered based on a preset third scoring threshold. Specifically, the third scoring threshold is set by the user based on actual needs, for example, it can be 0.65. If the scoring threshold of a certain first exploration data point is 0.7, which is greater than 0.65, then the first exploration data point is filtered out. The above threshold filtering is performed on each first exploration data point, and finally the remaining first exploration data points are marked as second exploration data points to achieve the first noise reduction process.

[0020] S104: Extract features from multiple second exploration data points to obtain the geological features corresponding to each second exploration data point.

[0021] In this embodiment, the process of extracting indicators with clear geological significance and reflecting subsurface geological characteristics from the original parameters of the second exploration data point is described. The core objective is to address the issues of high dimensionality and ambiguous physical meaning in the original data, thus adapting to subsequent clustering requirements. Geological characteristics refer to the extracted quantitative indicators that have geological interpretation value. Different exploration data types correspond to different characteristics; for example, the characteristics of seismic data include the mean amplitude, dominant frequency, or continuity of the phase axis; the characteristics of electrical resistivity data include the logarithm of resistivity or the polarizability gradient.

[0022] S105: A target clustering algorithm that adapts to the second exploration data points is determined based on the feature distribution of geological characteristics of each second exploration data point.

[0023] In this embodiment, the target clustering algorithm is either the K-means algorithm or a density-based clustering algorithm. These two algorithms are suited to different scenarios; the K-means algorithm is more suitable for processing uniformly distributed data, while the density-based clustering algorithm is more suitable for processing unevenly distributed data. Therefore, in this embodiment, the geological characteristics of each second exploration data point can be evaluated by calculating the coefficient of variation. The smaller the coefficient of variation, the more uniform the feature distribution; conversely, the larger the coefficient of variation, the higher the dispersion.

[0024] In this embodiment, a coefficient of variation threshold can be determined based on multiple experiments. If the calculated coefficient of variation is greater than the threshold, the target clustering algorithm is determined to be a density-based clustering algorithm; otherwise, the target clustering algorithm is determined to be a K-means algorithm.

[0025] S106: Determine the hyperparameters of the target clustering algorithm based on the anomaly scores of each second exploration data point.

[0026] In this embodiment, hyperparameters refer to the parameters that need to be preset before the target clustering algorithm runs. Different clustering algorithms correspond to different types of hyperparameters. For example, the hyperparameters of the K-means algorithm can be the initial number of clusters K and the initial cluster centers, while the hyperparameters of density-based clustering algorithms can be the neighborhood radius and the minimum number of points. Generally, the hyperparameters of clustering algorithms are set using the algorithm's default data or based on experience. However, in practice, the hyperparameters of clustering algorithms are usually manually adjusted based on the clustering results. This results in high time and labor costs, and the clustering effect is generally poor.

[0027] In this embodiment, considering that the anomaly score is used to evaluate the credibility of the corresponding second exploration data point, the hyperparameters of the target clustering algorithm can be determined based on the anomaly score of each second exploration data point, so that the target clustering algorithm after the hyperparameters are determined is more suitable for this scenario.

[0028] S107: Based on the target clustering algorithm after the hyperparameters are determined, the features corresponding to each second exploration data point are clustered to obtain the initial clustering results.

[0029] In this embodiment, the initial clustering result refers to the cluster partitioning result directly output by the target clustering algorithm after the hyperparameters are determined. Since the essence of the first denoising process is to filter based on a fixed threshold, there may still be some noisy data in the filtered data. In the clustering process of this embodiment, this part of the data can be clustered, which can be represented as clusters with fewer data points in the initial clustering result.

[0030] S108: Perform a second denoising process on clusters in the initial clustering results where the amount of data for the third exploration data point is less than a preset threshold, to obtain the target clustering results.

[0031] In this embodiment, the target clustering result contains multiple clusters, and different clusters represent different geological conditions in the target exploration area. The second denoising process refers to a fine denoising operation based on the anomaly scores of the small clusters in the initial clustering result. Unlike the first denoising process which removes explicit noise, the second denoising process filters out latent noise clusters, making the clustering result more consistent with the actual geological conditions.

[0032] In one embodiment, clusters containing third exploration data points in the initial clustering results with a data volume less than a preset threshold undergo a second denoising process to obtain the target clustering result, including: For each cluster containing less than a preset threshold of data points from the third exploration data point, the following anomaly detection operation is performed to obtain the target clustering result: Extract the outlier scores corresponding to all third exploration data points within the cluster. If the average outlier score of all third exploration data points within the cluster is greater than the preset second score threshold, then remove the cluster; if the average outlier score of all third exploration data points within the cluster is less than or equal to the second score threshold, then retain the cluster.

[0033] In this embodiment, the preset threshold is a quantitative standard used to determine whether a cluster in the initial clustering result is a small cluster. It can be set by technicians based on the data density of the exploration area and the minimum size of the geological body. For example, in 3D seismic data, it is set to 50, meaning that if the number of third exploration data points within a cluster is less than 50, it is determined to be a small cluster and needs to enter a secondary denoising process. The third exploration data point refers to the second exploration data point in the initial clustering result that is assigned to the same cluster, and the third exploration data points within the same cluster have similar geological characteristics.

[0034] In this embodiment, the mean anomaly score refers to the arithmetic mean of the anomaly scores of all third exploration data points within a cluster. Compared to the anomaly score of a single data point, the mean better reflects the overall attributes of the cluster, avoiding misjudgments caused by fluctuations in the anomaly scores of individual data points within the cluster. The second score threshold is the core threshold for determining the cluster's attributes; it needs to be set based on the anomaly score distribution after the first denoising and is lower than the third score threshold.

[0035] In this embodiment, a noise cluster refers to a cluster whose average abnormal score is greater than the second score threshold. Essentially, it is a latent noise (such as weak environmental interference or data interpolation error) that was not completely removed during the first denoising process. It has no geological interpretation value and needs to be removed.

[0036] S109: Based on the clusters in the target clustering results, perform projection visualization processing of geological exploration data to achieve visualization of geological exploration data in the target exploration area.

[0037] In one embodiment, geological exploration data is projected and visualized based on clusters in the target clustering results to visualize the geological exploration data of the target exploration area, including: For each cluster in the target clustering results, a dimensionality reduction projection algorithm is used to project the feature vectors of each third exploration data point in the cluster onto a three-dimensional visualization plane to obtain the corresponding projection distribution pattern of the cluster, so as to realize the visualization of geological exploration data of the target exploration area.

[0038] In this embodiment, the dimension reduction projection algorithm can be principal component analysis or a 3D projection algorithm, etc. The 3D visualization plane is a 3D coordinate system used to display the spatial distribution of underground geological bodies. The three coordinate axes typically correspond to the geographical coordinates of the exploration area (X-axis: east-west, Y-axis: north-south) and depth coordinates (Z-axis: vertical underground direction), which can intuitively present the location, shape, and scale of the geological bodies corresponding to the clusters in space. The projected distribution graphic refers to the spatial geometric shape formed on the 3D visualization plane after each cluster is dimensionally reduced and projected. The projected graphics of different clusters can be distinguished by different colors or transparency, making it easy for geologists to identify them.

[0039] In this embodiment, since the same cluster corresponds to the same type of geological body, cluster processing can directly reflect the spatial distribution of geological bodies and avoid the messy visualization of a single data point. Therefore, the clusters in the target clustering results are used as the basic units, and a dimension reduction projection algorithm is applied to the feature vector of each cluster to map it to the geographic-depth coordinate system of the three-dimensional visualization plane, thereby realizing the transformation from high-dimensional features to three-dimensional spatial coordinates.

[0040] In this embodiment, the depth interval of the three-dimensional horizontal slices can also be set (such as 50m, 100m, which can be customized by the user). Horizontal slices are extracted from each data in the target clustering result to obtain electrical structure slices at different depths. By aligning spatial coordinates, the electrical structure slices and shallow terrain are nested and superimposed to display the surface elevation undulations (such as using contour lines or three-dimensional terrain rendering) and the underground electrical distribution at the corresponding depth, intuitively presenting the spatial relationship between the terrain and the underground structure. The specific visualization process will not be described in detail in this embodiment.

[0041] As can be seen from the above, the embodiments of this application achieve optimization through two steps of precise denoising. First, this application uses the isolated forest algorithm to detect anomalies in the original geological exploration body data, obtains anomaly scores for each data point, and filters them, completing the first denoising step. Second, geological features are extracted and a suitable target clustering algorithm is determined. Hyperparameters are set using the existing anomaly scores. Since the anomaly scores can directly reflect the validity of the data, setting hyperparameters based on anomaly scores can make the target clustering algorithm more suitable for the geological data features. Then, clusters with data volume less than a preset threshold in the clustering results undergo a second denoising process to further purify the data. Compared with the existing technology that uses a single filter or fixed threshold to process noisy data, the embodiments of this application can improve the accuracy of noise processing, thereby improving the final visualization effect.

[0042] In one embodiment of this application, the target clustering algorithm is the K-means algorithm, and the hyperparameters of the K-means algorithm include: the initial number of clusters K; The hyperparameters of the target clustering algorithm are determined based on the abnormal scores of each second exploration data point, including: determining the frequency histogram of the abnormal scores of each second exploration data point; identifying the number of peaks in the frequency histogram where the abnormal scores are within a preset low score range, and determining the number of peaks as the initial cluster number K of the K-means algorithm.

[0043] In this embodiment, the frequency histogram of anomaly scores is a statistical graph plotted with anomaly scores on the horizontal axis and the number of second exploration data points in the corresponding score interval on the vertical axis. It is used to visually display the distribution pattern of anomaly scores. The peak value represents the score interval with the largest number of valid data points, corresponding to the geological body with the highest underground proportion. The preset low score interval refers to a pre-set interval with lower anomaly scores (usually 0~0.4). The second exploration data points in this interval are valid geological data without noise interference (which can truly reflect the characteristics of underground geological bodies), excluding the interference of high-scoring suspected noise points on the setting of the K value.

[0044] In this embodiment, the number of peaks in the preset low score interval is determined as the initial cluster number K because each peak corresponds to a geological body with a significant proportion. The number of peaks directly reflects the number of major underground geological bodies. Determining the value of K in this way allows the K-means algorithm to fit the actual geological reality from the beginning, thus improving the clustering effect.

[0045] In one embodiment of this application, the hyperparameters of the K-means algorithm further include: initial cluster centers; The method for determining the hyperparameters of the target clustering algorithm based on the anomaly scores of each second exploration data point also includes: assigning weight values ​​to each second exploration data point according to its anomaly scores; for each second exploration data point, the anomaly score of that second exploration data point is negatively correlated with the weight value corresponding to that second exploration data point; performing weighted calculation on the features of each second exploration data point based on the weight values ​​corresponding to each second exploration data point to obtain the target features; and selecting the K exploration data points farthest from the target features in the feature space as the initial clustering centers of the K-means algorithm.

[0046] In this embodiment, the initial cluster center refers to the starting coordinates (points in the feature space) of the K-means algorithm's iterative optimization, and is one of the core hyperparameters of K-means. The selection of the initial cluster center directly determines whether the algorithm gets trapped in a local optimum. If the selected points deviate from the feature center of the real geological body, the final clustering result will be out of sync with the actual geological body distribution.

[0047] In this embodiment, the negative correlation between abnormal scores and weight values ​​is intended to weaken the influence of high-scoring points that are suspected to be weak noise points, and to strengthen the feature dominance of low-scoring valid data. The target feature refers to the global reference value in the feature space obtained after weighted calculation of the features of each second exploration data point. The logic of selecting the K points farthest from the target feature is to ensure that the initial cluster centers are as dispersed as possible, covering different core regions (corresponding to different geological bodies) in the feature space, and avoiding the algorithm getting trapped in local optima due to initial center clustering. In this embodiment, K is the initial number of clusters.

[0048] In this embodiment, the process of determining weight values ​​based on abnormal scores can be based on a preset mapping relationship, a simple linear formula, or a nonlinear formula. Taking a linear formula as an example, the slope and intercept of the linear formula can be set by the user based on multiple experiments and actual needs. The specific mapping relationship and formula are not limited.

[0049] As can be seen from the above, the embodiments of this application reuse the process data of the first denoising process, use anomaly scores to filter out low-scoring but effective high-quality data, and determine the initial cluster number K by the peak value of its frequency histogram. Each peak corresponds to a core geological body, making the K value closely match the actual underground geological distribution, avoiding the problem of traditional K value setting being divorced from geological meaning and causing clustering distortion. Secondly, based on the negative correlation weighting of anomaly scores, the influence of weak noise is weakened and the dominance of effective data is strengthened. Then, K scattered points are selected as the initial cluster centers to avoid the algorithm getting trapped in local optima, further improving the clustering accuracy, improving the noise processing accuracy of geological exploration body data, and thus improving the visualization effect of geological exploration data.

[0050] In one embodiment of this application, the target clustering algorithm is a density-based clustering algorithm, and the hyperparameters of the density-based clustering algorithm include: neighborhood radius; The hyperparameters of the target clustering algorithm are determined based on the abnormal scores of each second exploration data point, including: screening candidate exploration data points whose abnormal scores are lower than a preset first score threshold among multiple second exploration data points; calculating the mean distance between the features of the candidate exploration data points, and determining the mean distance as the neighborhood radius.

[0051] In this embodiment, the density-based clustering algorithm can be the DBSCAN algorithm, whose hyperparameters include the neighborhood radius, which refers to the radius of the circular neighborhood of the feature space centered on a single data point. This determines the range of the neighborhood. If the neighborhood radius is too small, the same geological body will be split into multiple clusters, and if the neighborhood radius is too large, different geological bodies will be merged.

[0052] In this embodiment, the first score threshold is used to filter out anomaly scores of core, effective high-quality data points, ensuring that the data source for calculating the neighborhood radius is free from noise interference. The first score threshold is lower than the third score threshold. Candidate exploration data points refer to the second exploration data points whose anomaly scores are less than or equal to the first score threshold. These represent real geological data free from noise interference, and their density distribution reflects the spatial characteristics of the actual geological body. The mean distance between features is the arithmetic mean of the Euclidean distances between the geological feature vectors of the candidate exploration data points. Euclidean distance is a core indicator for measuring the similarity of data points in the feature space. The mean distance reflects the natural dispersion of the data, and setting the neighborhood radius accordingly allows the neighborhood range to match the density distribution of effective high-quality data.

[0053] In this embodiment, screening candidate exploration data points is to eliminate noise interference in density calculation, ensure that the data source used to guarantee the neighborhood radius is real and effective geological data, and avoid calculation distortion caused by noise. Secondly, calculating the distance mean of the candidate point features is to convert the density distribution pattern of the screened data into a quantitative value. The larger the distance mean, the more dispersed the effective high-quality data is in the feature space, and a larger neighborhood radius is required. Conversely, a smaller neighborhood radius needs to be set.

[0054] In one embodiment of this application, the hyperparameter of the density-based clustering algorithm further includes: minimum number of points; The hyperparameters of the target clustering algorithm are determined based on the anomaly scores of each second exploration data point. The method also includes: for each candidate exploration data point, determining the number of neighboring points of the candidate data in the region centered on the candidate data and with the neighborhood radius as the radius; calculating the mean number of neighboring points of each candidate data, and determining the minimum number of points based on the mean number of neighboring points.

[0055] In this embodiment, the minimum number of points refers to the threshold for determining core points. If the number of points within the neighborhood of a data point (with the neighborhood radius as the radius) is greater than or equal to the minimum number of points, then that point is a core point (capable of forming the core of a cluster); otherwise, it is a boundary point / noise point. This parameter directly determines the density of an area that can be determined as an effective mass cluster, and is key to distinguishing effective mass clusters from scattered noise points. The number of neighborhood points refers to the number of other candidate exploration data points contained within the feature space neighborhood centered on a single candidate exploration data point and with the determined neighborhood radius as the radius. The number of neighborhood points reflects the density around a single data point and is a basic indicator for measuring the local density of effective mass data. The average number of neighborhood points can reflect the overall average density of effective mass data. Using this as a basis to determine the minimum number of points allows the determination criteria for core points to match the actual density distribution of underground geological bodies. In this embodiment, the minimum number of points is generally determined to be 1.2-1.5 times the average number of neighborhood points to avoid directly using the average, which could lead to overly lenient (including scattered noise) or overly strict (splitting effective mass clusters) determinations for core points.

[0056] As can be seen from the above, the embodiments of this application screen noise-free candidate exploration data points through a first scoring threshold, eliminating the interference of noise on density calculation, and then determine the neighborhood radius based on the average characteristic distance of candidate points, so that the neighborhood range accurately matches the natural dispersion of effective geological data, avoiding the blindness of traditional empirical value setting. Secondly, to address the problem that the minimum number of points does not fit the geological density, the minimum number of points is determined based on the average number of neighborhood points of candidate points, so that the core point determination criteria match the actual situation of underground geological bodies, improving the accuracy of noise processing of geological exploration body data, and thus improving the visualization effect of geological exploration data.

[0057] Corresponding to the geological exploration data visualization method in the above embodiment, Figure 2 This is a structural block diagram of a geological exploration data visualization device provided in one embodiment of this application. For ease of explanation, only the parts relevant to the embodiment of this application are shown. References Figure 2 The geological exploration data visualization device 20 includes: a data acquisition module 21, an anomaly detection module 22, a first denoising module 23, a feature extraction module 24, a clustering algorithm determination module 25, a parameter determination module 26, a first clustering module 27, a second clustering module 28, and a visualization module 29.

[0058] Among them, the data acquisition module 21 is used to acquire the target geological exploration body data of the target exploration area; Anomaly detection module 22 is used to perform anomaly detection on each first exploration data point in the target geological exploration body data based on the isolated forest algorithm, and obtain anomaly score for each first exploration data point; The first denoising module 23 is used to filter each first exploration data point based on the anomaly score of each first exploration data point to obtain multiple second exploration data points after filtering, so as to achieve the first denoising process of the target geological exploration body data. Feature extraction module 24 is used to extract features from multiple second exploration data points to obtain the geological features corresponding to each second exploration data point; Clustering algorithm determination module 25 is used to determine the target clustering algorithm that is compatible with the second exploration data points based on the feature distribution of the geological features of each second exploration data point; The parameter determination module 26 is used to determine the hyperparameters of the target clustering algorithm based on the anomaly scores of each second exploration data point; The first clustering module 27 is used to perform clustering processing on the features corresponding to each second exploration data point based on the target clustering algorithm after the hyperparameters are determined, so as to obtain the initial clustering result; The second clustering module 28 is used to perform a second denoising process on clusters in the initial clustering results whose data volume of the third exploration data points is less than a preset threshold, to obtain the target clustering result; the target clustering result contains multiple clusters, and different clusters represent different geological features in the target exploration area; The visualization module 29 is used to perform projection visualization processing of geological exploration data based on clusters in the target clustering results, so as to realize the visualization of geological exploration data of the target exploration area.

[0059] In one embodiment of this application, the target clustering algorithm is the K-means algorithm, and the hyperparameters of the K-means algorithm include: the initial number of clusters K; The parameter determination module 26 is specifically used to determine the frequency histogram of the anomaly scores for each second exploration data point; The number of peaks in the frequency histogram that are abnormal scores within a preset low score range is identified, and the number of peaks is determined as the initial cluster number K of the K-means algorithm.

[0060] In one embodiment of this application, the hyperparameters of the K-means algorithm further include: initial cluster centers; The parameter determination module 26 is further used to assign weight values ​​to each second exploration data point based on the abnormal scores of each second exploration data point; for each second exploration data point, the abnormal score of the second exploration data point is negatively correlated with the weight value corresponding to the second exploration data point; The features of each second exploration data point are weighted and calculated based on the weight value corresponding to each second exploration data point to obtain the target features. The K exploration data points that are farthest from the target features in the feature space are selected as the initial clustering centers of the K-means algorithm.

[0061] In one embodiment of this application, the target clustering algorithm is a density-based clustering algorithm, and the hyperparameters of the density-based clustering algorithm include: neighborhood radius; The parameter determination module 26 is also specifically used to filter candidate exploration data points among multiple second exploration data points whose abnormal scores are lower than a preset first score threshold. Calculate the mean distance between the features of candidate exploration data points, and determine the mean distance as the neighborhood radius.

[0062] In one embodiment of this application, the hyperparameter of the density-based clustering algorithm further includes: minimum number of points; The parameter determination module 26 is further used to determine the number of neighboring points of each candidate exploration data point within an area centered on the candidate data and with a neighborhood radius as the radius. Calculate the average number of neighborhood points for each candidate data point, and determine the minimum number of points based on the average number of neighborhood points.

[0063] In one embodiment of this application, the second clustering module 28 is specifically used to perform the following anomaly judgment operation for each cluster containing third exploration data points whose data volume is less than a preset threshold, to obtain the target clustering result: Extract the outlier scores corresponding to all third exploration data points within the cluster. If the average outlier score of all third exploration data points within the cluster is greater than the preset second score threshold, then remove the cluster; if the average outlier score of all third exploration data points within the cluster is less than or equal to the second score threshold, then retain the cluster.

[0064] In one embodiment of this application, the visualization module 29 is specifically used to project the feature vectors of each third exploration data point in the target cluster to a three-dimensional visualization plane using a dimensionality reduction projection algorithm for each cluster, thereby obtaining the projection distribution graphic corresponding to the cluster and realizing the visualization of geological exploration data of the target exploration area.

[0065] See Figure 3 , Figure 3 This is a schematic block diagram of an electronic device provided according to an embodiment of this application. Figure 3 The electronic device 300 in this embodiment may include one or more processors 301, one or more input devices 302, one or more output devices 303, and one or more memories 304. The processors 301, input devices 302, output devices 303, and memories 304 communicate with each other via a communication bus 305. The memories 304 store computer programs, including program instructions. The processors 301 execute the program instructions stored in the memories 304. Specifically, the processors 301 are configured to invoke the program instructions to perform the functions of each module / unit in the above-described device embodiments, for example... Figure 2 The functions of the data acquisition module 21, anomaly detection module 22, first denoising module 23, feature extraction module 24, clustering algorithm determination module 25, parameter determination module 26, first clustering module 27, second clustering module 28, and visualization module 29 are shown.

[0066] It should be understood that, in the embodiments of this application, the processor 301 may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0067] Input device 302 may include a touchpad, a fingerprint sensor (for collecting the user's fingerprint information and fingerprint orientation information), a microphone, etc., and output device 303 may include a display (LCD, etc.), a speaker, etc.

[0068] The memory 304 may include read-only memory and random access memory, and provides instructions and data to the processor 301. A portion of the memory 304 may also include non-volatile random access memory. For example, the memory 304 may also store device type information.

[0069] In specific implementations, the processor 301, input device 302, and output device 303 described in the embodiments of this application can execute the implementation method described in the geological exploration data visualization method provided in the embodiments of this application, or they can execute the implementation method of the electronic device described in the embodiments of this application, which will not be repeated here.

[0070] In another embodiment of this application, a computer-readable storage medium is provided. This computer-readable storage medium stores a computer program, which includes program instructions. When executed by a processor, the program instructions implement all or part of the processes in the methods described above. Alternatively, the computer program can instruct related hardware to complete the process. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include any entity or device capable of carrying computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0071] The computer-readable storage medium can be an internal storage unit of the electronic device in any of the foregoing embodiments, such as a hard disk or memory of the electronic device. The computer-readable storage medium can also be an external storage device of the electronic device, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc., equipped on the electronic device. Furthermore, the computer-readable storage medium can include both internal and external storage units of the electronic device. The computer-readable storage medium is used to store computer programs and other programs and data required by the electronic device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0072] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this application.

[0073] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the electronic devices and units described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0074] In the several embodiments provided in this application, it should be understood that the disclosed electronic devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces or units, or it may be an electrical, mechanical, or other form of connection.

[0075] 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 units can be selected to achieve the purpose of the embodiments of this application, depending on actual needs.

[0076] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0077] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions.

[0078] The above are merely specific embodiments of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in this application, and these modifications or substitutions should all be covered within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for visualizing geological exploration data, characterized in that, include: Acquire data on the target geological exploration bodies in the target exploration area; Anomaly detection is performed on each first exploration data point in the target geological exploration body data based on the isolated forest algorithm, and anomaly scores are obtained for each first exploration data point. Based on the anomaly scores of each first exploration data point, the first exploration data points are filtered to obtain multiple filtered second exploration data points, so as to achieve the first denoising process of the target geological exploration body data. Feature extraction is performed on the multiple second exploration data points to obtain the geological features corresponding to each second exploration data point; Based on the feature distribution of the geological characteristics of each second exploration data point, a target clustering algorithm adapted to the second exploration data points is determined; The hyperparameters of the target clustering algorithm are determined based on the anomaly scores of each second exploration data point; Based on the target clustering algorithm after the hyperparameters are determined, the features corresponding to each second exploration data point are clustered to obtain the initial clustering results; A second denoising process is performed on clusters in the initial clustering results where the amount of data for the third exploration data point is less than the preset number threshold, to obtain the target clustering results; The target clustering results contain multiple clusters, and different clusters represent different geological features in the target exploration area; Based on the clusters in the target clustering results, the geological exploration data is projected and visualized to achieve the visualization of geological exploration data in the target exploration area.

2. The geological exploration data visualization method as described in claim 1, characterized in that, The target clustering algorithm is the K-means algorithm, and the hyperparameters of the K-means algorithm include: the initial number of clusters K; The process of determining the hyperparameters of the target clustering algorithm based on the anomaly scores of each second exploration data point includes: Determine the frequency histogram of the anomaly scores for each of the second exploration data points; Identify the number of peaks in the frequency histogram whose abnormal scores fall within a preset low score range, and determine the number of peaks as the initial cluster number K of the K-means algorithm.

3. The geological exploration data visualization method as described in claim 2, characterized in that, The hyperparameters of the K-means algorithm also include: initial cluster centers; The method of determining the hyperparameters of the target clustering algorithm based on the anomaly scores of each second exploration data point further includes: Each second exploration data point is assigned a weight value based on its anomaly score; for each second exploration data point, the anomaly score of that second exploration data point is negatively correlated with the weight value corresponding to that second exploration data point. The features of each second exploration data point are weighted and calculated based on the weight value corresponding to each second exploration data point to obtain the target features. The K exploration data points that are farthest from the target features in the feature space are selected as the initial clustering centers of the K-means algorithm.

4. The geological exploration data visualization method as described in claim 1, characterized in that, The target clustering algorithm is a density-based clustering algorithm, and the hyperparameters of the density-based clustering algorithm include: neighborhood radius; The process of determining the hyperparameters of the target clustering algorithm based on the anomaly scores of each second exploration data point includes: Filter candidate exploration data points whose abnormal scores are lower than a preset first score threshold from among the plurality of second exploration data points; Calculate the mean distance between the features of the candidate exploration data points, and determine the mean distance as the neighborhood radius.

5. The geological exploration data visualization method as described in claim 4, characterized in that, The hyperparameters of the density-based clustering algorithm also include: minimum number of points; The method of determining the hyperparameters of the target clustering algorithm based on the anomaly scores of each second exploration data point further includes: For each candidate exploration data point, determine the number of neighboring points within the region centered on the candidate data point and with a neighborhood radius as the radius. Calculate the average number of neighboring points for each candidate data point, and determine the minimum number of points based on the average number of neighboring points.

6. The geological exploration data visualization method as described in claim 1, characterized in that, The second denoising process is performed on clusters in the initial clustering result where the amount of data for the third exploration data point is less than the preset threshold, to obtain the target clustering result, including: For each cluster containing a third exploration data point whose data volume is less than the preset threshold, the following anomaly detection operation is performed to obtain the target clustering result: Extract the abnormal scores corresponding to all third exploration data points within the cluster. If the average of the abnormal scores corresponding to all third exploration data points within the cluster is greater than the preset second score threshold, then the cluster is removed; if the average of the abnormal scores corresponding to all third exploration data points within the cluster is less than or equal to the second score threshold, then the cluster is retained.

7. The geological exploration data visualization method as described in claim 1, characterized in that, The projection visualization processing of geological exploration data based on the clusters in the target clustering results to achieve visualization of geological exploration data in the target exploration area includes: For each cluster in the target clustering results, a dimensionality reduction projection algorithm is used to project the feature vectors of each third exploration data point in the cluster onto a three-dimensional visualization plane to obtain the corresponding projection distribution pattern of the cluster, so as to realize the visualization of geological exploration data of the target exploration area.

8. A geological exploration data visualization device, characterized in that, include: The data acquisition module is used to acquire data on the target geological exploration bodies in the target exploration area; An anomaly detection module is used to perform anomaly detection on each first exploration data point in the target geological exploration body data based on the isolated forest algorithm, and to obtain anomaly scores for each first exploration data point. The first denoising module is used to filter each first exploration data point based on the anomaly score of each first exploration data point to obtain multiple second exploration data points after filtering, so as to achieve the first denoising process of the target geological exploration body data. The feature extraction module is used to extract features from the plurality of second exploration data points to obtain the geological features corresponding to each second exploration data point. The clustering algorithm determination module is used to determine the target clustering algorithm that is compatible with the second exploration data points based on the feature distribution of the geological characteristics of each second exploration data point; The parameter determination module is used to determine the hyperparameters of the target clustering algorithm based on the anomaly scores of each second exploration data point; The first clustering module is used to perform clustering processing on the features corresponding to each second exploration data point based on the target clustering algorithm after the hyperparameters are determined, and to obtain the initial clustering results; The second clustering module is used to perform a second denoising process on clusters in the initial clustering result where the amount of data of the third exploration data points is less than the preset number threshold, so as to obtain the target clustering result. The target clustering results contain multiple clusters, and different clusters represent different geological features in the target exploration area; The visualization module is used to perform projection visualization processing of geological exploration data based on the clusters in the target clustering results, so as to realize the visualization of geological exploration data in the target exploration area.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 7.

Images