Copper ore exploration method and system combining multi-source remote sensing data and AI

Abstract

The invention belongs to the technical field of mineral resource exploration, and provides a multi-source remote sensing data and AI combined copper ore exploration method and system, and the method comprises the following steps: obtaining multi-source remote sensing data of a target region through a plurality of remote sensing devices; preprocessing the multi-source remote sensing data to obtain fusion data; the fusion data are analyzed through an AI algorithm, and mineralization alteration areas and geological structure information are obtained through recognition; performing overlay analysis on the mineralization alteration area and the known geochemical anomaly area, analyzing the spatial relationship between the mineralization area and the structure in combination with the geological structure information, and judging the favorable mineralization part of the copper ore to obtain an analysis result; performing field investigation on the potential copper ore area determined by the analysis result, collecting rock samples for laboratory analysis, and verifying the accuracy of the analysis result; according to the method, the multi-source remote sensing data and the AI technology are comprehensively utilized, the target area is rapidly analyzed, the potential copper ore area is positioned, the workload of blind exploration is reduced, and the exploration efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of mineral resource exploration, and specifically relates to a copper mine exploration method and system that combines multi-source remote sensing data with AI. Background Art

[0002] As an important non-ferrous metal that is indispensable in industrial production, the development of copper exploration technology has always attracted much attention. Traditional copper exploration technology mainly relies on geological mapping, geophysical exploration (such as gravity exploration, magnetic exploration) and geochemical sampling. Geological mapping requires exploration personnel to conduct a lot of field surveys and measurements in the field, and determine the stratigraphic structure and structural characteristics by observing rock outcrops and drawing geological profiles. However, this method is extremely inefficient and it is difficult to complete the exploration of large areas in a short period of time. In addition, the manpower and time costs are extremely high.

[0003] Geophysical exploration methods use differences in the physical properties of rocks and ores (such as density, magnetism, and electrical properties) to detect underground geological structures and ore body distribution. For example, gravity exploration infers underground density differences by measuring gravity anomalies. However, this method is easily affected by terrain fluctuations and underground non-ore anomalies, resulting in multiple interpretations of the data and difficulty in accurately locating ore bodies. Although magnetic exploration can quickly identify magnetic geological bodies, it has poor detection effects on weakly magnetic or non-magnetic copper deposits, which can easily lead to missed or misjudgment. Geochemical sampling collects samples such as soil and stream sediments and analyzes the content of copper and associated elements to identify anomalous areas. However, the distribution and number of sampling points have a significant impact on the results. Intensive sampling is costly, while sparse sampling may miss important information. Geochemical anomalies are affected by multiple factors, and not all anomalies are related to copper deposits. Further verification work is required, resulting in long exploration cycles and low efficiency.

[0004] With the development of remote sensing technology, single optical remote sensing data was once applied to copper mine exploration, identifying mineralized and altered areas by analyzing the spectral characteristics and texture information of remote sensing images. However, optical remote sensing is limited by weather and lighting conditions, and effective data cannot be obtained in cloudy, foggy, or rainy weather. It also has difficulty detecting deep geological structures. The accuracy of mineralization identification based solely on spectral and texture characteristics is limited and is easily interfered with by factors such as vegetation cover and rock weathering. As a result, existing technologies lack the ability to deeply integrate and analyze multi-source data with geological and geochemical information, making it difficult to fully reveal the mechanisms and patterns of copper mineralization and accurately locate favorable mineralization areas. This leads to blind field investigations and increases exploration costs and risks.

[0005] To this end, technicians in this field have proposed a copper mine exploration method and system that combines multi-source remote sensing data with AI. The method aims to comprehensively utilize multi-source remote sensing data and AI technology to quickly analyze the target area and quickly locate potential copper mine areas, reducing the workload of blind exploration and improving exploration efficiency. Summary of the Invention

[0006] In order to solve the above technical problems, the present invention provides a copper mine exploration method and system that combines multi-source remote sensing data with AI to solve the problems raised in the background technology.

[0007] According to a first aspect of the present disclosure, a copper mine exploration method combining multi-source remote sensing data with AI is proposed, comprising the following steps:

[0008] S1. Acquire multi-source remote sensing data of the target area through multiple remote sensing devices; the multi-source remote sensing data includes optical remote sensing data, thermal infrared remote sensing data, and radar remote sensing data;

[0009] S2. Preprocessing the multi-source remote sensing data, wherein the preprocessing includes radiation correction, geometric correction, and image fusion to obtain fused data;

[0010] S3. performing cluster analysis on the fused data using a machine learning algorithm to identify mineralized alteration areas;

[0011] S4. Identify and analyze the fused data using a deep learning algorithm to identify and obtain geological structure information;

[0012] S5. Superimposing and analyzing the mineralized alteration area with the known geochemical anomaly area, verifying each other, and obtaining verification information;

[0013] S6. Based on the verification information and in combination with the geological structure information, analyze the spatial relationship between the mineralized area and the structure, determine the favorable location for copper mineralization, and obtain an analysis result;

[0014] S7. Conduct field investigations in the potential copper deposit areas identified by the analysis results, collect rock samples for laboratory analysis, and verify the accuracy of the analysis results.

[0015] Preferably, the cluster analysis of the fused data using a machine learning algorithm to identify the mineralized alteration area includes:

[0016] The fused data is represented as a dataset D = {x1, x2, ..., x n}, where x i is a d-dimensional feature vector, representing the feature description of the i-th data point in the fused data, n is the total number of data points, and d is the dimension of the feature;

[0017] Randomly select K data points as the initial cluster centers, denoted as C = {c1, c2, ..., c K}, where c j is the d-dimensional vector of the j-th cluster center;

[0018] Use the Euclidean distance to compare each data point x in the dataset i , calculate its relationship with each cluster center c j Distance:

[0019]

[0020] Among them, x i,k and c j,k They are x i and c j The k-th dimension component of i Assign it to the cluster with the closest cluster center;

[0021] For each cluster, calculate the mean of all data points in the cluster as the new cluster center. The update formula is expressed as:

[0022]

[0023] Among them, S j is the set of data points in the jth cluster, c j is the new cluster center, |S j | is the set S j The number of data points in ;

[0024] By continuously updating the distribution of cluster centers and data points until the cluster centers no longer change significantly or the preset number of iterations is reached; at this point, the data points in different clusters have similar characteristics; combining geological knowledge and prior experience, the clustering results are screened and judged to determine the mineralized alteration area.

[0025] Preferably, the identifying and analyzing the fused data by a deep learning algorithm to obtain geological structure information includes:

[0026] Represent the fused data as a three-dimensional tensor Where H represents the height of the image, W represents the width of the image, and C represents the number of channels of the image;

[0027] Using a deep learning algorithm, after multiple convolutional layers and pooling layers, the feature map is flattened into a one-dimensional vector and input to the fully connected layer. The fully connected layer synthesizes the extracted features and uses the following formula to obtain the output result x of the fully connected layer. l :

[0028] x l=W l ×x l-1 +b l

[0029] Among them, W l is the weight matrix of the lth fully connected layer, and b l is the bias vector, and x l-1 is the input vector, and

[0030] The output is converted into a probability distribution using the Softmax function, representing the probability of belonging to different geological structure types:

[0031]

[0032] Among them, x i is the output of the i-th neuron in the output layer, y i is the corresponding probability, k is the total number of geological structure types; by outputting the probability that different areas in the fused data belong to various geological structure types, the geological structure information is identified.

[0033] Preferably, the superposition analysis of the mineralized alteration area and the known geochemical anomaly area to verify each other to obtain verification information includes:

[0034] Based on the mineralized alteration area, it is represented as a binary image, where the pixel value of the mineralized alteration area is 1 and the pixel value of the non-mineralized alteration area is 0, denoted as an image matrix M, with a size of m×n, where m and n are the number of rows and columns of the image respectively;

[0035] Based on the geochemical anomaly area, it is represented as a binary image, where the pixel value of the geochemical anomaly area is 1 and the pixel value of other areas is 0, which is recorded as an image matrix G with a size of m×n.

[0036] The mineralized alteration area M and the geochemical anomaly area G are logically ANDed using the following formula to obtain a new matrix V as verification information:

[0037] V ij =M ij ∧G ij

[0038] Where i=1,2,...,m,j=1,2,...,n; ∧ represents the logical AND operation, that is, when M ij =1 and G ij =1, V ij =1, indicating that the pixel point belongs to both the mineralization alteration area and the geochemical anomaly area; otherwise, V ij =0;

[0039] By counting the pixels with a value of 1 in the matrix V, the number N and the proportion P of the total number of pixels are calculated, that is, Taking P as a verification indicator, the higher the ratio, the higher the overlap between the mineralized alteration area and the geochemical anomaly area, and the better the mutual verification effect.

[0040] Preferably, the analysis of the spatial relationship between the mineralized area and the structure based on the verification information and in combination with the geological structure information to determine the favorable location for copper mineralization and obtain the analysis results includes:

[0041] The mineralized area V obtained based on the verification information is represented by polygonal vector data, and its boundary is represented by a series of coordinate points (x vi ,y vi ), i=1,2,...,n v , n v is the number of boundary points;

[0042] The geological structure F obtained based on the geological structure information is represented by a series of line segments, each of which consists of two endpoint coordinates (x fi ,y fi ) and (x fj ,y fj ) Determine, i, j = 1, 2, ..., n f , n f is the number of fracture structural segments;

[0043] The comprehensive evaluation index A is obtained through spatial relationship analysis using the following formula:

[0044] A=ω1P d +ω2(1-|cosθ-1|)+ω3P overlop

[0045] Among them, ω1, ω2, ω3 are the corresponding weights, P d is the number of points in the mineralized area that are within the preset threshold d0 from the geological structure and the proportion of the total number of points in the mineralized area. cosθ is the cosine value of the angle θ between the long axis direction of the mineralized area V and the strike of the geological structure F. P overlop The area of ​​overlap between the mineralized area V and the geological structure F, and the proportion of the overlap to the area of ​​the mineralized area;

[0046] The favorable locations for copper mineralization are determined based on the value of the comprehensive evaluation index A. The higher the value of A, the more favorable the spatial relationship between the mineralized area and the geological structure in the region is for copper mineralization, thereby obtaining analysis results and determining potential favorable locations for copper mineralization.

[0047] According to a second aspect of the present disclosure, a copper mine exploration system combining multi-source remote sensing data with AI is proposed, comprising:

[0048] Data acquisition module, used to obtain multi-source remote sensing data of the target area through various remote sensing devices;

[0049] A data processing module, configured to pre-process the multi-source remote sensing data to obtain fused data;

[0050] An AI processing module is used to perform AI analysis on the fused data to identify mineralized alteration areas and geological structure information;

[0051] A collaborative analysis module is used to overlay and analyze the mineralized alteration area with the known geochemical anomaly area to verify each other; and combined with the geological structure information, analyze the spatial relationship between the mineralized area and the structure, determine the favorable location for copper mineralization, and obtain analysis results;

[0052] The field verification module is used to conduct field investigations in potential copper mining areas determined based on the analysis results, collect rock samples for laboratory analysis, and verify the accuracy of the analysis results.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] 1. By comprehensively utilizing multi-source remote sensing data and AI technology, the present invention can quickly and extensively analyze target areas, quickly locate potential copper mining areas, reduce the workload of blind exploration, and improve exploration efficiency.

[0055] 2. The present invention reduces the limitations of single data or methods by integrating multi-source data and combining multiple analysis methods, complementing and verifying each other, and improving the accuracy of identifying mineralized areas and judging favorable mineralization locations. It can not only identify mineralized areas, but also obtain various information such as geological structure, providing rich data support for in-depth research on the mineralization mechanism and laws of copper mines, and helping to grasp the regional mineralization characteristics as a whole.

[0056] 3. The present invention uses remote sensing and AI technology to conduct large-scale screening and analysis in the early stage, which can carry out field investigations in a more targeted manner, reduce unnecessary field work investment, and reduce exploration costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 This is a flow chart of the copper mine exploration method combining multi-source remote sensing data with AI of the present invention;

[0058] Figure 2 This is a block diagram of the copper mine exploration system that combines multi-source remote sensing data with AI. DETAILED DESCRIPTION

[0059] The following embodiments of the present invention are described in further detail with reference to the accompanying drawings and examples. The following examples are used to illustrate the present invention but are not intended to limit the scope of the present invention.

[0060] Example 1: As shown in the attached Figure 1 As shown, the present invention provides a copper mine exploration method combining multi-source remote sensing data with AI, comprising the following steps:

[0061] S1. Obtain multi-source remote sensing data for the target area through a variety of remote sensing equipment; multi-source remote sensing data includes optical remote sensing data, thermal infrared remote sensing data and radar remote sensing data; obtain optical remote sensing data, thermal infrared remote sensing data and radar remote sensing data of the target area through a variety of remote sensing equipment, observe the target area from different angles and characteristics, and provide a rich data basis for subsequent analysis; among them, optical remote sensing data can provide high-resolution surface images for identifying terrain, landforms and other features; thermal infrared remote sensing data can reflect the difference in surface thermal radiation, which is helpful for discovering thermal anomalies related to mineralization; radar remote sensing data is not restricted by weather and lighting conditions, and can obtain information such as terrain undulation and surface roughness. The combination of the three can comprehensively cover a variety of surface and shallow geological information of the target area.

[0062] S2. Preprocess multi-source remote sensing data. The preprocessing includes radiation correction, geometric correction and image fusion to obtain fused data. Radiation correction eliminates the influence of the sensor itself and atmospheric factors on the radiation value, so that the data can truly reflect the reflection and emission characteristics of the surface. Geometric correction ensures the spatial position accuracy of the image and facilitates the precise matching between different data. Image fusion combines the advantages of different remote sensing data to form more informative and interpretable fused data.

[0063] S3. Perform cluster analysis on the fused data through machine learning algorithm to identify the mineralized alteration area; express the fused data as a data set D = {x1, x2, ..., x n}, where x i is a d-dimensional feature vector, representing the feature description of the i-th data point in the fused data, n is the total number of data points, and d is the dimension of the feature;

[0064] Randomly select K data points as the initial cluster centers, denoted as C = {c1, c2, ..., c K}, where c j is the d-dimensional vector of the j-th cluster center;

[0065] Use the Euclidean distance to compare each data point x in the dataset i , calculate its relationship with each cluster center c j Distance:

[0066]

[0067] Among them, x i,k and c j,k They are x i and c j The k-th dimension component of i Assign it to the cluster with the closest cluster center;

[0068] For each cluster, calculate the mean of all data points in the cluster as the new cluster center. The update formula is expressed as:

[0069]

[0070] Among them, S j is the set of data points in the jth cluster, c j is the new cluster center, |S j | is the set S j The number of data points in ;

[0071] By continuously updating the distribution of cluster centers and data points until the cluster centers no longer change significantly or the preset number of iterations is reached; at this point, the data points in different clusters have similar characteristics; combining geological knowledge and prior experience, the clustering results are screened and judged to determine the mineralized alteration area.

[0072] By using machine learning algorithms to perform cluster analysis on the fused data, the pixels in the image are divided into different categories according to the characteristics of the data, thereby identifying possible mineralized alteration areas and automatically discovering potential patterns and regularities in the data without the need to know the specific characteristics of the mineralized alteration areas in advance. This improves the objectivity and accuracy of the identification and provides important target area information for subsequent precise exploration.

[0073] S4. Use deep learning algorithms to identify and analyze the fused data to obtain geological structure information; represent the fused data as a three-dimensional tensor Where H represents the height of the image, W represents the width of the image, and C represents the number of channels of the image;

[0074] Using a deep learning algorithm, after multiple convolutional layers and pooling layers, the feature map is flattened into a one-dimensional vector and input to the fully connected layer. The fully connected layer synthesizes the extracted features and uses the following formula to obtain the output result x of the fully connected layer. l :

[0075] x l =W l ×x l-1 +b l

[0076] Among them, W l is the weight matrix of the lth fully connected layer, and b l is the bias vector, and x l -1 is the input vector, and

[0077] The output is converted into a probability distribution using the Softmax function, representing the probability of belonging to different geological structure types:

[0078]

[0079] Among them, x i is the output of the i-th neuron in the output layer, y i is the corresponding probability, k is the total number of geological structure types; by outputting the probability that different areas in the fused data belong to various geological structure types, the geological structure information is identified.

[0080] By analyzing the fused data through deep learning algorithms, it is possible to accurately identify the details and features of geological structures from complex remote sensing images, thereby extracting geological structure information.

[0081] S5. Superimpose and analyze the mineralized alteration area with the known geochemical anomaly area, verify each other, and obtain verification information; based on the mineralized alteration area, represent it as a binary image, where the pixel value of the mineralized alteration area is 1 and the pixel value of the non-mineralized alteration area is 0, denoted as image matrix M, with a size of m×n, where m and n are the number of rows and columns of the image respectively;

[0082] Based on the geochemical anomaly area, it is represented as a binary image, where the pixel value of the geochemical anomaly area is 1 and the other areas are 0, which is recorded as the image matrix G with a size of m×n.

[0083] Perform a logical AND operation on the mineralized alteration area M and the geochemical anomaly area G using the following formula to obtain a new matrix V as verification information:

[0084] V ij =M ij ∧G ij

[0085] Where i=1,2,...,m,j=1,2,...,n; ∧ represents the logical AND operation, that is, when M ij =1 and G ij =1, V ij =1, indicating that the pixel point belongs to both the mineralization alteration area and the geochemical anomaly area; otherwise, V ij =0;

[0086] By counting the pixels with a value of 1 in the matrix V, we can calculate their number N and their proportion P to the total number of pixels, that is, Taking P as a verification indicator, the higher the ratio, the higher the overlap between the mineralized alteration area and the geochemical anomaly area, and the better the mutual verification effect.

[0087] The mineralized alteration areas are superimposed and analyzed with known geochemical anomaly areas to verify each other and improve the reliability of the results. If the mineralized alteration areas and geochemical anomaly areas coincide with each other in space or have a high degree of overlap, then the possibility that these areas are potential copper mines is greatly increased, reducing the misjudgment that may be caused by a single data source.

[0088] S6. Based on the verification information and combined with the geological structure information, the spatial relationship between the mineralized area and the structure is analyzed to determine the favorable location for copper mineralization and obtain the analysis results;

[0089] The mineralized area V obtained based on the verification information is represented by polygonal vector data, and its boundary is represented by a series of coordinate points (x vi ,y vi ), i=1,2,...,n v , n v is the number of boundary points;

[0090] The geological structure F obtained based on geological structure information is represented by a series of line segments, each of which consists of two endpoint coordinates (x fi ,y fi ) and (x fj ,y fj ) Determine, i, j = 1, 2, ..., n f , n f is the number of fracture structural segments;

[0091] The comprehensive evaluation index A is obtained through spatial relationship analysis using the following formula:

[0092] A=ω1P d +ω2(1-|cosθ-1|)+ω3P overlop

[0093] Among them, ω1, ω2, ω3 are the corresponding weights, P d is the number of points in the mineralized area that are within the preset threshold d0 from the geological structure and the proportion of the total number of points in the mineralized area. cosθ is the cosine value of the angle θ between the long axis direction of the mineralized area V and the strike of the geological structure F. P overlop The area of ​​overlap between the mineralized area V and the geological structure F, and the proportion of the overlap to the area of ​​the mineralized area;

[0094] The favorable locations for copper mineralization are determined based on the value of the comprehensive evaluation index A. The higher the value of A, the more favorable the spatial relationship between the mineralized area and the geological structure in the region is for copper mineralization. The analysis results are then obtained to determine the potential favorable locations for copper mineralization.

[0095] Based on verification information and geological structure information, the spatial relationship between the mineralized area and the structure is analyzed to determine the favorable locations for copper mineralization. By comprehensively considering multiple factors, the possible locations of copper mines can be predicted more accurately, providing more precise guidance for field surveys and improving exploration efficiency and success rate.

[0096] S7. Conduct field surveys of potential copper deposits identified by the analysis results, collect rock samples for laboratory analysis, and verify the accuracy of the analysis results. Field surveys and sample analysis of potential copper deposits identified by the analysis results will verify the accuracy of the analysis results. This means that through actual sampling and laboratory analysis, we can directly obtain information on the composition and structure of the ore, verify whether the remote sensing and data analysis results are consistent with the actual situation, and provide direct evidence for the final determination of the existence and scale of copper resources.

[0097] Example 2: As shown in the attached Figure 2 As shown, the present invention also provides a copper mine exploration system that combines multi-source remote sensing data with AI, including:

[0098] Data acquisition module, used to obtain multi-source remote sensing data of the target area through various remote sensing devices;

[0099] Data processing module, used to pre-process multi-source remote sensing data to obtain fused data;

[0100] AI processing module, used to perform AI analysis on fused data to identify mineralized alteration areas and geological structure information;

[0101] The collaborative analysis module is used to overlay and analyze mineralized alteration areas with known geochemical anomaly areas for mutual verification; and combined with geological structural information, it analyzes the spatial relationship between mineralized areas and structures, determines favorable locations for copper mineralization, and obtains analytical results;

[0102] The field verification module is used to conduct field investigations in potential copper deposit areas identified based on the analysis results, collect rock samples for laboratory analysis, and verify the accuracy of the analysis results.

[0103] As can be seen from the above, the integration of multi-source remote sensing data, such as optical, thermal infrared, and radar, and its combination with machine learning and deep learning AI technologies, fully leverages the advantages of multi-source data and the powerful data analysis capabilities of AI technology. The use of machine learning and deep learning algorithms enables the automatic identification and analysis of mineralized alteration areas and geological structure information, reducing the subjectivity and workload of manual interpretation and improving the speed and accuracy of analysis. Furthermore, by analyzing the spatial relationship between mineralized areas and geological structures in detail and combining information such as geochemical anomalies for comprehensive judgment, this multi-factor comprehensive analysis method more comprehensively considers the various controlling factors of copper mineralization and is more scientific and accurate than traditional single-factor or simple overlay analysis methods.

[0104] It is important to note that the construction and arrangement of the present application shown in a plurality of different exemplary embodiments are merely illustrative. Although only a few embodiments are described in detail in this disclosure, it will be readily understood by those who consult this disclosure that many modifications are possible without departing substantially from the novel teachings and advantages of the subject matter described in this application. Other substitutions, modifications, changes, and omissions may be made in the design, operating conditions, and arrangement of the exemplary embodiments without departing from the scope of the present invention. Therefore, the present invention is not limited to specific embodiments, but extends to a variety of modifications that still fall within the scope of the appended claims.

[0105] Additionally, in order to provide a concise description of exemplary embodiments, all features of an actual embodiment (i.e., those features that are not relevant to the best mode presently contemplated for carrying out the invention or those that are not relevant to implementing the invention) may not be described.

[0106] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A copper mine exploration method combining multi-source remote sensing data with AI, characterized by: The following steps are involved: S1. Acquire multi-source remote sensing data of the target area through multiple remote sensing devices; the multi-source remote sensing data includes optical remote sensing data, thermal infrared remote sensing data, and radar remote sensing data; S2. Preprocessing the multi-source remote sensing data, wherein the preprocessing includes radiation correction, geometric correction, and image fusion to obtain fused data; S3. performing cluster analysis on the fused data using a machine learning algorithm to identify mineralized alteration areas; S4. Identify and analyze the fused data using a deep learning algorithm to identify and obtain geological structure information; S5. Superimposing and analyzing the mineralized alteration area with the known geochemical anomaly area, verifying each other, and obtaining verification information; S6. Based on the verification information and in combination with the geological structure information, analyze the spatial relationship between the mineralized area and the structure, determine the favorable location for copper mineralization, and obtain an analysis result; S7. Conduct field investigations in the potential copper deposit areas identified by the analysis results, collect rock samples for laboratory analysis, and verify the accuracy of the analysis results.

2. The copper mine exploration method combining multi-source remote sensing data with AI as claimed in claim 1, characterized in that: The cluster analysis of the fused data using a machine learning algorithm to identify mineralized alteration areas includes: The fused data is represented as a dataset D = {x1, x2, ..., x n }, where x i is a d-dimensional feature vector, representing the feature description of the i-th data point in the fused data, n is the total number of data points, and d is the dimension of the feature; Randomly select K data points as the initial cluster centers, denoted as C = {c1, c2, ..., c K }, where c j is the d-dimensional vector of the j-th cluster center; Use the Euclidean distance to compare each data point x in the dataset i , calculate its relationship with each cluster center c j Distance: Among them, x i,k and c j,k They are x i and c j The k-th dimension component of i Assign it to the cluster with the closest cluster center; For each cluster, calculate the mean of all data points in the cluster as the new cluster center. The update formula is expressed as: Among them, S j is the set of data points in the jth cluster, c j is the new cluster center, |S j | is the set S j The number of data points in ; By continuously updating the distribution of cluster centers and data points until the cluster centers no longer change significantly or the preset number of iterations is reached; at this point, the data points in different clusters have similar characteristics; combining geological knowledge and prior experience, the clustering results are screened and judged to determine the mineralized alteration area.

3. The copper mine exploration method combining multi-source remote sensing data with AI as claimed in claim 1, characterized in that: The identifying and analyzing the fused data by a deep learning algorithm to obtain geological structure information includes: Represent the fused data as a three-dimensional tensor Where H represents the height of the image, W represents the width of the image, and C represents the number of channels of the image; Using a deep learning algorithm, after multiple convolutional layers and pooling layers, the feature map is flattened into a one-dimensional vector and input to the fully connected layer. The fully connected layer synthesizes the extracted features and uses the following formula to obtain the output result x of the fully connected layer. l : x l =W l ×x l-1 +b l Among them, W l is the weight matrix of the lth fully connected layer, and b l is the bias vector, and x l-1 is the input vector, and The output is converted into a probability distribution using the Softmax function, representing the probability of belonging to different geological structure types: Among them, x i is the output of the i-th neuron in the output layer, y i is the corresponding probability, k is the total number of geological structure types; by outputting the probability that different areas in the fused data belong to various geological structure types, the geological structure information is identified.

4. The copper mine exploration method combining multi-source remote sensing data with AI as claimed in claim 1, characterized in that: The mineralized alteration area is superimposed and analyzed with the known geochemical anomaly area, and the verification information obtained by mutual verification includes: Based on the mineralized alteration area, it is represented as a binary image, where the pixel value of the mineralized alteration area is 1 and the pixel value of the non-mineralized alteration area is 0, denoted as an image matrix M, with a size of m×n, where m and n are the number of rows and columns of the image respectively; Based on the geochemical anomaly area, it is represented as a binary image, where the pixel value of the geochemical anomaly area is 1 and the pixel value of other areas is 0, which is recorded as an image matrix G with a size of m×n. The mineralized alteration area M and the geochemical anomaly area G are logically ANDed using the following formula to obtain a new matrix V as verification information: In ij =M ij ∧G ij Where i=1,2,...,m,j=1,2,...,n; ∧ represents the logical AND operation, that is, when M ij =1 and G ij =1, V ij =1, indicating that the pixel point belongs to both the mineralization alteration area and the geochemical anomaly area; otherwise, V ij =0; By counting the pixels with a value of 1 in the matrix V, the number N and the proportion P of the total number of pixels are calculated, that is, Taking P as a verification indicator, the higher the ratio, the higher the overlap between the mineralized alteration area and the geochemical anomaly area, and the better the mutual verification effect.

5. The copper mine exploration method combining multi-source remote sensing data with AI as claimed in claim 1, characterized in that: The verification information is combined with the geological structure information to analyze the spatial relationship between the mineralized area and the structure, determine the favorable location for copper mineralization, and obtain analysis results, including: The mineralized area V obtained based on the verification information is represented by polygonal vector data, and its boundary is represented by a series of coordinate points (x vi ,y vi ), i=1,2,...,n v , n v is the number of boundary points; The geological structure F obtained based on the geological structure information is represented by a series of line segments, each of which consists of two endpoint coordinates (x fi ,y fi ) and (x fj ,y fj ) Determine, i, j = 1, 2, ..., n f , n f is the number of fracture structural segments; The comprehensive evaluation index A is obtained through spatial relationship analysis using the following formula: A=ω1P d +ω2(1-|cosθ-1|)+ω3P overlop Among them, ω1, ω2, ω3 are the corresponding weights, P d is the number of points in the mineralized area that are within the preset threshold d0 from the geological structure and the proportion of the total number of points in the mineralized area. cosθ is the cosine value of the angle θ between the long axis direction of the mineralized area V and the strike of the geological structure F. P overlop The area of ​​overlap between the mineralized area V and the geological structure F, and the proportion of the overlap to the area of ​​the mineralized area; The favorable locations for copper mineralization are determined based on the value of the comprehensive evaluation index A. The higher the value of A, the more favorable the spatial relationship between the mineralized area and the geological structure in the region is for copper mineralization, thereby obtaining analysis results and determining potential favorable locations for copper mineralization.

6. The copper mine exploration system combining multi-source remote sensing data with AI is characterized by: include: Data acquisition module, used to obtain multi-source remote sensing data of the target area through various remote sensing devices; A data processing module, configured to pre-process the multi-source remote sensing data to obtain fused data; An AI processing module is used to perform AI analysis on the fused data to identify mineralized alteration areas and geological structure information; A collaborative analysis module is used to overlay and analyze the mineralized alteration area with the known geochemical anomaly area to verify each other; and combined with the geological structure information, analyze the spatial relationship between the mineralized area and the structure, determine the favorable location for copper mineralization, and obtain analysis results; The field verification module is used to conduct field investigations in potential copper mining areas determined based on the analysis results, collect rock samples for laboratory analysis, and verify the accuracy of the analysis results.

Images