Mine geological dynamic identification monitoring method based on multi-source remote sensing data

Abstract

The application discloses a mine geological dynamic identification monitoring method based on multi-source remote sensing data, and relates to the technical field of mine monitoring, comprising the following steps: acquiring multi-source remote sensing image data of a same period of a mine area and a predetermined range area around the mine area; acquiring at least one candidate geological attribute of the mine area and the predetermined range area around the mine area; generating all non-repeated to-be-verified geological combinations; verifying matching rationality indexes of each to-be-verified geological combination; screening out a to-be-verified geological combination with the highest matching rationality index as the geological attribute of the mine area and the predetermined range area around the mine area; and repeating the execution according to a predetermined monitoring period.The application has the advantages that: by establishing a two-way action quantization model of the mine area and the surrounding geological environment, a geological attribute combination matching verification mechanism is realized, the misjudgment problem in traditional one-way analysis is overcome, the misidentification rate can be reduced under complex rock stratum conditions, and a leap from static identification to dynamic collaborative verification is realized.

Description

Technical Field

[0001] This invention relates to the field of mine monitoring technology, specifically to a method for dynamic identification and monitoring of mine geology based on multi-source remote sensing data. Background Technology

[0002] Traditional mine geological monitoring primarily relies on ground exploration and the interpretation of single remote sensing data, focusing mainly on the image feature analysis of the mining area itself. Existing technologies typically treat the mining area as an independent unit, directly inferring geological attributes by extracting its spectral texture features, neglecting the interaction between the surrounding geological environment and the mine. Furthermore, existing methods often employ the assumption of one-way influence when establishing correlation models, meaning they only consider the effects of the surrounding area on the mining area, without addressing the reverse influence mechanisms of the mine's geological attributes on the surrounding environment.

[0003] This technical approach has significant limitations: First, the isolated analysis model leads to a higher false detection rate in identifying similar geological features, especially in areas with complex lithology; second, the one-way correlation model cannot truly reflect the dynamic interaction between the mine and surrounding geological units; and third, it lacks a quantitative verification mechanism for the two-way interaction of geological attributes, making it difficult to capture the co-evolution characteristics of correlated attributes in multi-period monitoring. These shortcomings restrict the accuracy of geological state assessment and the ability to provide dynamic early warning. Summary of the Invention

[0004] To address the aforementioned technical issues, this paper provides a method for dynamic identification and monitoring of mine geology based on multi-source remote sensing data. This technical solution solves the problem that existing methods often adopt the assumption of unidirectional influence when establishing correlation models, that is, they only consider the effect of the surrounding area on the mine area, without involving the reverse influence mechanism of mine geological attributes on the surrounding environment.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] Methods for dynamic identification and monitoring of mine geology based on multi-source remote sensing data include:

[0007] Multi-source remote sensing image data of the mining area and its surrounding predetermined area at the same time are acquired, and the multi-source remote sensing image data are preprocessed.

[0008] Based on the interpretation of pre-processed remote sensing image data of the mining area and its surrounding predetermined area, at least one candidate geological attribute of the mining area and its surrounding predetermined area is obtained;

[0009] From at least one candidate geological attribute in the mining area and at least one candidate geological attribute in the surrounding predetermined area, one candidate geological attribute is selected to form a geological combination to be verified.

[0010] Iterate through and generate all unique geological combinations to be verified;

[0011] Based on the influence of the geological attributes of the surrounding predetermined area on each candidate geological attribute of the mining area, and the analysis of the reverse influence of each candidate geological attribute of the mining area on the geological attributes of the surrounding predetermined area, the matching rationality index of each geological combination to be verified is verified.

[0012] The geological combination with the highest matching rationality index was selected as the geological attribute of the mining area and its surrounding predetermined area.

[0013] The monitoring is repeated according to a predetermined monitoring cycle to obtain the temporal changes in the geological attributes of the mining area and output dynamic monitoring information.

[0014] Preferably, the step of interpreting preprocessed remote sensing image data of the mining area and its surrounding predetermined area to obtain at least one candidate geological attribute specifically includes:

[0015] Multi-dimensional feature extraction is performed on the preprocessed remote sensing image data of the mining area or the surrounding predetermined area to obtain a feature set;

[0016] The feature set is input into the pre-trained geological interpretation model, and the output is a probabilistic geological attribute candidate set. The geological interpretation model adopts a multi-task learning framework and is trained based on cross-modal remote sensing fusion data. The probabilistic geological attribute candidate set includes: attribute type and existence probability.

[0017] Attribute types with a probability exceeding a set threshold are selected as candidate geological attributes for the mining area or a predetermined surrounding area.

[0018] Preferably, the influence relationship between the geological attributes of the surrounding predetermined range area and the candidate geological attributes of the mining area specifically includes:

[0019] Based on historical experience data, we analyzed the frequency of the surrounding predetermined area under each candidate geological attribute and the frequency of the mining area under each candidate geological attribute.

[0020] Based on the frequency of the surrounding predetermined range area under each candidate geological attribute and the frequency of the mining area under each candidate geological attribute, an influence array of the surrounding predetermined range area under each candidate geological attribute is constructed.

[0021] Preferably, the analysis of the reverse influence relationship between each candidate geological attribute of the mining area and the geological attributes of the surrounding predetermined area specifically includes:

[0022] Based on historical experience data, we analyze the frequency with which the surrounding predetermined area falls under each candidate geological attribute for each mining area.

[0023] Based on the frequency of the surrounding predetermined range area being in each candidate geological attribute under each candidate geological attribute of the mining area, an array of reverse influences of the surrounding predetermined range area under each candidate geological attribute of the mining area is constructed.

[0024] Preferably, the indicators for verifying the matching rationality of each geological combination to be verified specifically include:

[0025] Based on the mining area influence array of each surrounding predetermined range area under each candidate geological attribute, extract the product of the frequencies of the mining area under each candidate geological attribute of the surrounding predetermined range area in the geological combination to be verified, and use it as the positive matching index of the geological combination to be verified.

[0026] Based on the reverse influence array of the surrounding predetermined range area of ​​the mining area under each candidate geological attribute, the product of the frequencies of the candidate geological attributes of each surrounding predetermined range area under the candidate geological attributes of the geological combination to be verified is extracted and used as the reverse matching index of the geological combination to be verified.

[0027] By combining the positive and negative matching indices of the geological combination to be verified, a matching rationality index for the geological combination to be verified is obtained.

[0028] Preferably, the method for obtaining the matching rationality index of the geological combination to be verified by comprehensively combining the positive matching index and the negative matching index of the geological combination to be verified is as follows:

[0029] Based on the preset weight ratio, the positive matching index and the negative matching index of the geological combination to be verified are weighted and summed to obtain the matching rationality index of the geological combination to be verified.

[0030] Preferably, the method for obtaining the matching rationality index of the geological combination to be verified by comprehensively combining the positive matching index and the negative matching index of the geological combination to be verified is as follows:

[0031] Based on the TOPSIS algorithm, a formula for evaluating the rationality of matching is constructed.

[0032] Substituting the positive and negative matching indices of each geological combination to be verified into the matching rationality assessment formula, we obtain the matching rationality index of the geological combination to be verified.

[0033] Preferably, the method for dividing the surrounding predetermined range area is as follows:

[0034] Define the geological correlation distance of the mine, and take the area outside the mine outline and extending the geological correlation distance of the mine as the surrounding correlation area;

[0035] Based on the center of the mining area, several segmented sectors are formed according to a preset angle. These segmented sectors divide the surrounding related areas into several predetermined surrounding range areas.

[0036] Preferably, the multi-source remote sensing image data includes at least two of the following: visible-near-infrared remote sensing images, thermal infrared remote sensing images, microwave radar images, and hyperspectral remote sensing images.

[0037] Preferably, the preprocessing of the multi-source remote sensing image data specifically includes: radiometric calibration, atmospheric correction, and geometric fine correction.

[0038] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0039] This invention innovatively achieves a combined matching verification mechanism for geological attributes by establishing a quantitative model of the two-way interaction between the mining area and the surrounding geological environment. Specifically, it utilizes historical frequency data to construct arrays representing the influence of the surrounding area on the mining area and the reaction arrays representing the mining area's influence on the surrounding area, forming a mutually reinforcing correlation verification system. Based on this, it iterates through all possible attribute combinations, calculates the two-way matching degree index through frequency product, and generates a comprehensive rationality assessment value using weighted fusion or the TOPSIS algorithm. This collaborative verification method significantly overcomes the attribute misjudgment problem in traditional one-way analysis and can reduce the false identification rate under complex rock strata conditions. Its sector segmentation mechanism adapts to the geological influence range definition needs of various mining modes, breaking through the cognitive limitations of "isolated interpretation and one-way inference" in geological monitoring technology, and achieving a leap from static identification to dynamic collaborative verification. Attached Figure Description

[0040] Figure 1 This is a flowchart of the mine geological dynamic identification and monitoring method based on multi-source remote sensing data proposed in this invention;

[0041] Figure 2 This is a flowchart illustrating the specific method proposed in this invention for obtaining candidate geological attributes of a mining area and its surrounding predetermined range. Detailed Implementation

[0042] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0043] Reference Figure 1 As shown, the method for dynamic identification and monitoring of mine geology based on multi-source remote sensing data includes:

[0044] Multi-source remote sensing image data of the mining area and its surrounding predetermined area at the same time are acquired separately, and the multi-source remote sensing image data are preprocessed. The multi-source remote sensing image data includes at least two of the following: visible light-near infrared remote sensing images, thermal infrared remote sensing images, microwave radar images, and hyperspectral remote sensing images. The preprocessing steps specifically include: radiometric calibration, atmospheric correction, and geometric fine correction.

[0045] By integrating the complementarity of remote sensing data from multiple bands, such as hyperspectral identification of mineral composition and radar penetration of clouds to monitor deformation, and eliminating radiation distortion and geometric deviation, a physical consistency foundation is laid for cross-regional collaborative analysis, avoiding spatial matching errors caused by differences in data sources or atmospheric interference.

[0046] Based on the interpretation of pre-processed remote sensing image data of the mining area and its surrounding predetermined area, at least one candidate geological attribute of the mining area and its surrounding predetermined area is obtained;

[0047] By using multi-source data collaborative interpretation to generate probabilistic candidate sets, the initial screening coverage of complex geological attributes, such as concealed faults and karst development areas, can be significantly improved, and the false negative rate can be effectively reduced compared with single data interpretation.

[0048] From at least one candidate geological attribute in the mining area and at least one candidate geological attribute in the surrounding predetermined area, one candidate geological attribute is selected to form a geological combination to be verified.

[0049] The traditional isolated attribute determination is transformed into an associated combination verification framework to solve the problem of missing modeling of "multi-attribute coupling effect" in unidirectional causal relationship model;

[0050] Iterate through and generate all unique geological combinations to be verified;

[0051] The exhaustive method ensures that all possible geological scenarios are included in the verification scope, avoiding the omission of local optima that may be caused by heuristic algorithms, and improving the completeness of model decision-making, especially in areas of abrupt lithological changes.

[0052] Based on the influence of the geological attributes of the surrounding predetermined area on the candidate geological attributes of the mining area, and the analysis of the reverse influence of the candidate geological attributes of the mining area on the geological attributes of the surrounding predetermined area, the matching rationality index of each geological combination to be verified is verified.

[0053] Introducing a two-way interaction mechanism to cross-validate the rationality of the combination can effectively reduce the misjudgment rate of associated attributes.

[0054] The geological combination with the highest matching rationality index was selected as the geological attribute of the mining area and its surrounding predetermined area.

[0055] The monitoring is repeated according to the predetermined monitoring cycle to obtain the temporal changes in the geological attributes of the mining area and output dynamic monitoring information.

[0056] By combining dynamic monitoring information, potential geological risks in mining areas can be identified in advance, providing a critical window of opportunity for engineering intervention.

[0057] Reference Figure 2 As shown, based on the interpretation of preprocessed remote sensing image data of the mining area and its surrounding predetermined area, at least one candidate geological attribute of the mining area and its surrounding predetermined area is obtained, specifically including:

[0058] For preprocessed remote sensing image data of the mining area or a predetermined surrounding area, multi-dimensional feature extraction is performed to obtain a feature set. Specifically, the multi-dimensional feature extraction includes:

[0059] Extracting spectral dimension features: Calculating the standardized mineral alteration index and moisture content index for each pixel, where the mineral alteration index is calculated based on the iron ion absorption peak depth ratio, and the moisture content index is calculated based on the short-wave infrared triangular index;

[0060] Extracting spatial texture features: The gray-level co-occurrence matrix parameters are calculated through a sliding window, and the directional gradient histogram is used to detect linearly constructed boundaries.

[0061] Extracting temporal deformation features: When historical image data exists, a surface deformation rate field is generated using temporal interferometric radar technology;

[0062] The feature set is input into the pre-trained geological interpretation model, and the output is a probabilistic geological attribute candidate set. The geological interpretation model adopts a multi-task learning framework and is trained based on cross-modal remote sensing fusion data. The probabilistic geological attribute candidate set includes: attribute type and existence probability.

[0063] Attribute types with a probability exceeding a set threshold are selected as candidate geological attributes for the mining area or a predetermined surrounding area.

[0064] By integrating spectral dimension features, this method accurately quantifies the degree of mineralization and alteration, as well as spatial texture features, based on the depth ratio of iron ion absorption peaks. It identifies tectonic boundaries, rock mass structure features, and temporal deformation characteristics using gray-level co-occurrence matrices and directional gradient histograms. Interferometric radar technology is used to capture surface deformation trends, constructing a multimodal fusion-based quantitative analysis system that overcomes the information limitations of single remote sensing methods. The extracted feature set is input into a multi-task geological interpretation model trained on cross-modal data, simultaneously interpreting the probabilistic distribution results of multiple geological attributes such as lithology, structure, and hydrology. Its multi-task collaborative optimization mechanism significantly suppresses coupling interference errors between attributes. By filtering low-confidence interpretation terms based on probability thresholds, a candidate geological attribute set with high coverage and strong robustness is ultimately formed. This provides a data foundation with clear physical meaning and spatial boundaries for subsequent geological combination verification, fundamentally solving the problems of attribute misjudgment and missed detection caused by single features or lack of correlation in traditional methods.

[0065] Specifically, the influence of the geological attributes of the surrounding predetermined area on the various candidate geological attributes of the mining area includes:

[0066] Based on historical experience data, we analyzed the frequency of the surrounding predetermined area under each candidate geological attribute and the frequency of the mining area under each candidate geological attribute.

[0067] Based on the frequency of the surrounding predetermined area under each candidate geological attribute, and the frequency of the mining area under each candidate geological attribute, an influence array A of the mining area under each candidate geological attribute of the surrounding predetermined area is constructed. i A i =[a i1 (1) … a ij (k) … a im (K i )], where A i Let a be the mining area influence array corresponding to the i-th surrounding predetermined range area. ij (k) represents the frequency at which the i-th surrounding predetermined range area falls under the k-th candidate geological attribute and the j-th candidate geological attribute, K. i Let m be the total number of candidate geological attribute types in the i-th surrounding predetermined range area, and m be the total number of candidate geological attribute types in the mining area.

[0068] The analysis of the inverse influence of each candidate geological attribute in the mining area on the geological attributes of the surrounding predetermined area specifically includes:

[0069] Based on historical experience data, we analyze the frequency with which the surrounding predetermined area falls under each candidate geological attribute for each mining area.

[0070] Based on the frequency of the surrounding predetermined area falling under each candidate geological attribute of the mining area, an array B representing the reverse influence of the surrounding predetermined area of ​​the mining area under each candidate geological attribute is constructed. ij B ij =[b ij (1) … b ij (k) … b ij (K i )],B ij This is an array representing the reverse influence of a mining area under the j-th candidate geological attribute on the i-th surrounding predetermined area. ij (k) represents the frequency of the i-th surrounding predetermined range area of ​​the mining area under the j-th candidate geological attribute under the k-th candidate geological attribute;

[0071] Specifically, the indicators for verifying the matching rationality of each geological combination to be verified include:

[0072] Let the geological assemblage to be verified be {j k1 … k} i … k n}, j represents the mining area being in the j-th candidate geological attribute, k i This represents the i-th surrounding predetermined area being located at the k-th position. i There are several candidate geological attributes, where n is the total number of surrounding predetermined areas;

[0073] Based on the mining area influence array of each surrounding predetermined range area under each candidate geological attribute, extract the product of the frequencies of the mining area under each candidate geological attribute of the surrounding predetermined range area in the geological combination to be verified, and use it as the positive matching index of the geological combination to be verified.

[0074] The formula for calculating the positive matching index is:

[0075]

[0076] P + As a positive matching indicator, a ij (k i ) represents the i-th surrounding predetermined range area located at the k-th position. i The frequency of a mining area being in the j-th candidate geological attribute under a variety of candidate geological attributes;

[0077] Based on the reverse influence array of the surrounding predetermined range area of ​​the mining area under each candidate geological attribute, the product of the frequencies of the candidate geological attributes of each surrounding predetermined range area under the candidate geological attributes of the geological combination to be verified is extracted and used as the reverse matching index of the geological combination to be verified.

[0078]

[0079] P - b is the reverse matching metric. ij (k i ) represents the area within the i-th predetermined surrounding area of ​​the mining area under the j-th candidate geological attribute, and is located at the k-th... i Frequency under each candidate geological attribute;

[0080] By combining the positive and negative matching indices of the geological combination to be verified, a matching rationality index for the geological combination to be verified is obtained.

[0081] In some preferred embodiments, the matching rationality index of the geological combination to be verified can be obtained by weighted summation of the positive matching index and the negative matching index of the geological combination to be verified.

[0082] Specifically, the calculation formula is as follows:

[0083] P = αP + +βP - ;

[0084] Where P is the matching rationality index, α and β are both weights, and α+β=1;

[0085] In other preferred embodiments, the matching rationality index of the geological combination to be verified can be obtained by combining the positive matching index and the negative matching index of the geological combination to be verified based on the TOPSIS algorithm, specifically including:

[0086] Based on the TOPSIS algorithm, a formula for evaluating the rationality of matching is constructed.

[0087] The formula for assessing the reasonableness of a match is as follows:

[0088]

[0089] Where P is the matching rationality index, P +max P is the maximum value among all positive matching indices of the geological assemblies to be verified. +min P is the minimum value among all positive matching indices of the geological assemblies to be verified. -max P is the maximum value among all the reverse matching indices of the geological assemblies to be verified. -min The minimum value among all the reverse matching indices of the geological combinations to be verified;

[0090] Substituting the positive and negative matching indices of each geological combination to be verified into the matching rationality assessment formula, we obtain the matching rationality index of the geological combination to be verified.

[0091] By establishing a two-way probabilistic influence transmission mechanism, an innovative objective quantitative assessment of the matching degree of geological combinations is achieved. Specifically, historical frequency statistics are used to construct a two-way influence array between regions. The positive influence array accurately depicts the conditional probability distribution of the formation of mine regional attributes under specific geological conditions in the surrounding areas, while the negative influence array characterizes the feedback law of the geological state of the mine area on the surrounding environment. In the combination verification stage, the joint frequency product of all combinations to be verified in the positive and negative two-way arrays is calculated to generate positive and negative matching indices with clear statistical significance. By integrating the positive and negative matching indices, the dimensional differences and weight preferences of the two indices are effectively coordinated, fully respecting the dialectical unity of action and reaction forces in the geological system while avoiding the subjective weight setting bias of the weighted summation method. This significantly improves the physical consistency and spatiotemporal synergy of multi-source remote sensing interpretation conclusions in complex geological environments, establishing a statistically significant decision-making basis for geological disaster prevention and control throughout the entire life cycle of mines.

[0092] The method for dividing the surrounding predetermined area is as follows:

[0093] Define the geological correlation distance of the mine. The area extending beyond the mine outline by the geological correlation distance is taken as the surrounding correlation area. The value of the geological correlation distance of the mine is 300-1500 meters, which is determined based on the rock quality.

[0094] Based on the center of the mining area, several segmented sectors are formed according to a preset angle. These segmented sectors divide the surrounding related areas into several predetermined surrounding range areas.

[0095] This regional delineation method establishes a scientific spatial domain definition system through the outward expansion mechanism of the mine outline and a central radial sector model. The dynamic setting of the mine's geological correlation distance precisely controls the outward expansion range, ensuring effective coverage of the influence domain of physical processes such as mine pressure transmission and groundwater seepage. The equiangular sector segmentation based on the geometric center point aligns with the directionality of rock stress transmission and the continuity of tectonic units, avoiding geological information distortion caused by arbitrary cutting of tectonic boundaries. This model significantly improves the integrity of the spatial distribution of geological attributes in the surrounding area, establishing a topologically accurate physical correlation framework for the two-way interaction mechanism between the mine and its surrounding environment, ensuring the spatial reliability of subsequent matching and verification.

[0096] In summary, the advantages of this invention are as follows: By establishing a quantitative model of the two-way interaction between the mining area and the surrounding geological environment, it innovatively realizes a combination matching verification mechanism for geological attributes. Specifically, it utilizes historical frequency data to construct an array of the influence of the surrounding area on the mining area and an array of the reaction of the mining area on the surrounding area, forming a mutually corroborating correlation verification system. Based on this, it iterates through all possible attribute combinations, calculates the two-way matching degree index through frequency product, and generates a comprehensive rationality assessment value using weighted fusion or the TOPSIS algorithm. This collaborative verification method significantly overcomes the attribute misjudgment problem in traditional one-way analysis and can reduce the false identification rate under complex rock strata conditions. Its sector segmentation mechanism adapts to the geological influence range definition needs of various mining modes, breaking through the cognitive limitations of "isolated interpretation and one-way inference" in geological monitoring technology, and achieving a leap from static identification to dynamic collaborative verification.

[0097] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the claimed invention. The scope of protection claimed by the appended claims and their equivalents is defined.

Claims

1. A method for dynamic identification and monitoring of mine geology based on multi-source remote sensing data, characterized in that, include: Multi-source remote sensing image data of the mining area and its surrounding predetermined area at the same time are acquired, and the multi-source remote sensing image data are preprocessed. Based on the interpretation of pre-processed remote sensing image data of the mining area and its surrounding predetermined area, at least one candidate geological attribute of the mining area and its surrounding predetermined area is obtained; From at least one candidate geological attribute in the mining area and at least one candidate geological attribute in the surrounding predetermined area, one candidate geological attribute is selected to form a geological combination to be verified. Iterate through and generate all unique geological combinations to be verified; Based on the influence of the geological attributes of the surrounding predetermined area on each candidate geological attribute of the mining area, and the analysis of the reverse influence of each candidate geological attribute of the mining area on the geological attributes of the surrounding predetermined area, the matching rationality index of each geological combination to be verified is verified. The geological combination with the highest matching rationality index was selected as the geological attribute of the mining area and its surrounding predetermined area. The monitoring is repeated according to a predetermined monitoring cycle to obtain the temporal changes in the geological attributes of the mining area and output dynamic monitoring information. Specifically, the influence relationship between the geological attributes of the surrounding predetermined area and the candidate geological attributes of the mining area includes: Based on historical experience data, we analyzed the frequency of the surrounding predetermined area under each candidate geological attribute and the frequency of the mining area under each candidate geological attribute. Based on the frequency of the surrounding predetermined range area under each candidate geological attribute and the frequency of the mining area under each candidate geological attribute, an influence array of the surrounding predetermined range area under each candidate geological attribute is constructed. The analysis of the inverse influence of each candidate geological attribute of the mining area on the geological attributes of the surrounding predetermined area specifically includes: Based on historical experience data, we analyze the frequency with which the surrounding predetermined area falls under each candidate geological attribute for each mining area. Based on the frequency of the surrounding predetermined range area being in each candidate geological attribute under each candidate geological attribute of the mining area, an array of reverse influences of the surrounding predetermined range area under each candidate geological attribute of the mining area is constructed. The specific indicators for verifying the matching rationality of each geological combination to be verified include: Based on the mining area influence array of each surrounding predetermined range area under each candidate geological attribute, extract the product of the frequencies of the mining area under each candidate geological attribute of the surrounding predetermined range area in the geological combination to be verified, and use it as the positive matching index of the geological combination to be verified. Based on the reverse influence array of the surrounding predetermined range area of ​​the mining area under each candidate geological attribute, the product of the frequencies of the candidate geological attributes of each surrounding predetermined range area under the candidate geological attributes of the geological combination to be verified is extracted and used as the reverse matching index of the geological combination to be verified. By combining the positive and negative matching indices of the geological combination to be verified, a matching rationality index for the geological combination to be verified is obtained.

2. The method for dynamic identification and monitoring of mine geology based on multi-source remote sensing data according to claim 1, characterized in that, The interpretation of remote sensing image data of the preprocessed mining area and its surrounding predetermined area to obtain at least one candidate geological attribute of the mining area and its surrounding predetermined area specifically includes: Multi-dimensional feature extraction is performed on the preprocessed remote sensing image data of the mining area or the surrounding predetermined area to obtain a feature set; The feature set is input into the pre-trained geological interpretation model, and the output is a probabilistic geological attribute candidate set. The geological interpretation model adopts a multi-task learning framework and is trained based on cross-modal remote sensing fusion data. The probabilistic geological attribute candidate set includes: attribute type and existence probability. Attribute types with a probability exceeding a set threshold are selected as candidate geological attributes for the mining area or a predetermined surrounding area.

3. The method for dynamic identification and monitoring of mine geology based on multi-source remote sensing data according to claim 2, characterized in that, The method for obtaining the matching rationality index of the geological combination to be verified by combining the positive matching index and the negative matching index of the geological combination to be verified is as follows: Based on the preset weight ratio, the positive matching index and the negative matching index of the geological combination to be verified are weighted and summed to obtain the matching rationality index of the geological combination to be verified.

4. The method for dynamic identification and monitoring of mine geology based on multi-source remote sensing data according to claim 3, characterized in that, The method for obtaining the matching rationality index of the geological combination to be verified by combining the positive matching index and the negative matching index of the geological combination to be verified is as follows: Based on the TOPSIS algorithm, a formula for evaluating the rationality of matching is constructed. Substituting the positive and negative matching indices of each geological combination to be verified into the matching rationality assessment formula, we obtain the matching rationality index of the geological combination to be verified.

5. The method for dynamic identification and monitoring of mine geology based on multi-source remote sensing data according to claim 1, characterized in that, The method for dividing the surrounding predetermined area is as follows: Define the geological correlation distance of the mine, and take the area outside the mine outline and extending the geological correlation distance of the mine as the surrounding correlation area; Based on the center of the mining area, several segmented sectors are formed according to a preset angle. These segmented sectors divide the surrounding related areas into several predetermined surrounding range areas.

6. The method for dynamic identification and monitoring of mine geology based on multi-source remote sensing data according to claim 1, characterized in that, The multi-source remote sensing image data includes at least two of the following: visible-near-infrared remote sensing images, thermal infrared remote sensing images, microwave radar images, and hyperspectral remote sensing images.

7. The method for dynamic identification and monitoring of mine geology based on multi-source remote sensing data according to claim 1, characterized in that, The preprocessing of the multi-source remote sensing image data specifically includes: radiometric calibration, atmospheric correction, and geometric fine correction.

Images