Intelligent identification method of mineral resources based on multi-source remote sensing data

Abstract

The present invention relates to the field of mineral exploration technology and discloses a method for intelligent identification of mineral resources based on multi-source remote sensing data. The method constructs an intelligent identification model based on deviation analysis and correction mechanism by integrating multi-source remote sensing data, geological data and mineral reaction characteristic signals. The method can dynamically adapt to different environmental conditions and significantly improve the accuracy and reliability of mineral resource exploration. Through dynamic correction and multi-coefficient integration, the method effectively reduces the misjudgment rate, improves the recognition accuracy and precision rate of mineral target areas, and can quickly screen out potential mining areas in a large area, reducing manual intervention and significantly improving exploration efficiency. By generating multiple mineral marking guidance coefficients KWX, the potential of mineral resources is comprehensively evaluated from multiple angles, significantly improving the accuracy and stability of mineral identification.

Description

Technical Field

[0001] The present invention relates to the technical field of mineral exploration, and in particular to a method for intelligently identifying mineral resources based on multi-source remote sensing data. Background Art

[0002] Mineral resource exploration is a key application area for remote sensing technology, particularly the identification and detection of mineral resource distribution through remote sensing data. Specifically, intelligent identification of iron ore is a key task in this field. Leveraging multi-source remote sensing data, combined with image processing and data fusion techniques, it is possible to accurately identify potential iron ore resource areas, significantly improving the efficiency and accuracy of resource exploration.

[0003] Although intelligent mineral resource identification methods based on multi-source remote sensing data have made some progress in iron ore exploration, they still face some key challenges. First, the resolution of remote sensing data remains limited in large-scale iron ore exploration, resulting in the inability to accurately identify subtle features in the mining area, affecting identification accuracy. Second, optical remote sensing data is significantly affected by atmospheric interference and shadow effects, especially in areas with complex terrain. These effects can cause surface features in the mining area to be obscured, thereby reducing the ability to effectively identify mineral resources. Furthermore, in areas with deeply buried ore bodies or complex terrain, traditional remote sensing methods may have difficulty identifying hidden mineral distributions, resulting in omissions or errors in identification results. Furthermore, although multi-source data fusion technology can help improve identification accuracy, due to differences in resolution and quality among different data sources, effectively fusing these data remains a technical challenge, affecting the scope of application and accuracy of mineral resource exploration methods.

[0004] Therefore, we proposed an intelligent identification method of mineral resources based on multi-source remote sensing data to solve the above problems. Summary of the Invention

[0005] The purpose of the present invention is to provide an intelligent identification method for mineral resources based on multi-source remote sensing data to solve the problem of mineral exploration raised in the above background technology.

[0006] To achieve the above-mentioned object, the present invention provides the following technical solution: a method for intelligent identification of mineral resources based on multi-source remote sensing data, characterized in that the specific steps are as follows:

[0007] S1. Data collection: collecting multi-source data in the mineral identification process and preprocessing them to generate a first data set, a second data set, and a third data set;

[0008] S2. Remote sensing analysis: performing calculation and analysis on the first data set to generate a remote sensing deviation analysis coefficient YGF, a remote sensing correction coefficient YGJ, and a first analysis result;

[0009] S3, geological analysis, performing calculation and analysis on the second data set to generate a geological deviation analysis coefficient DZF and a geological correction coefficient DZJ and a second analysis result;

[0010] S4, mineral characteristic signal analysis, performing calculation and analysis on the third data set, thereby generating a mineral response deviation analysis coefficient KWF and a mineral response correction coefficient KWJ and a third analysis result;

[0011] S5. Integrate and calculate the remote sensing deviation analysis coefficient YGF, the remote sensing correction coefficient YGJ, the geological deviation analysis coefficient DZF, the geological correction coefficient DZJ, the mineral reaction deviation analysis coefficient KWF, and the mineral reaction correction coefficient KWJ to generate multiple mineral marking guidance coefficients KWX. Analyze the multiple mineral marking guidance coefficients KWX separately to generate a fourth analysis result. Based on the fourth analysis result, execute the mineral marking instruction on the target.

[0012] Preferably, step S1 completes data collection and preprocessing through the following steps:

[0013] Step S1.1, identifying and extracting remote sensing data in the mineral identification process, geomorphological environment data collected by remote sensing, and mineral reaction data;

[0014] The remote sensing data include spectral reflectance index, terrain deformation rate, and lidar canopy height;

[0015] Geomorphic environmental data include terrain complexity index, vegetation coverage and rainfall erosivity index water;

[0016] Mineral reaction data include shortwave infrared absorption depth, thermal infrared emissivity anomaly index, and radar polarimetric scattering entropy;

[0017] The multi-source dataset consists of spectral reflectance index, terrain deformation rate, lidar canopy height, terrain complexity index, vegetation coverage, rainfall erosivity index, shortwave infrared absorption depth, thermal infrared emissivity anomaly index, and radar polarization scattering entropy;

[0018] Step S1.2, preprocessing and dimensionlessizing the multi-source data set, and reorganizing the preprocessed multi-source data set into a first data set, a second data set, and a third data set;

[0019] The first data set includes a spectral reflectance index A, a terrain deformation rate B, and a lidar canopy height C;

[0020] The second data set includes terrain complexity index D, vegetation coverage E, and rainfall erosivity index F;

[0021] The third data set includes the shortwave infrared absorption depth G, the thermal infrared emissivity anomaly index H, and the radar polarization scattering entropy I.

[0022] Preferably, the step S2 completes the calculation and analysis of the first data set through the following steps:

[0023] Step S2.1, performing integration calculation on the first data set to generate a remote sensing bias analysis coefficient YGF and a remote sensing correction coefficient YGJ;

[0024] Step S2.2: Analyze the calculated remote sensing deviation analysis coefficient YGF to generate a first analysis result, as follows:

[0025] when When , it means that the current remote sensing data is normal and the remote sensing correction coefficient YGJ does not need to be used in subsequent model analysis;

[0026] when When , it means that the current remote sensing data fluctuates, and the remote sensing correction coefficient YGJ needs to be used in subsequent model analysis.

[0027] Preferably, in step S2.1, the remote sensing deviation analysis coefficient YGF and the remote sensing correction coefficient YGJ are calculated by the following formula:

[0028] ;

[0029] ;

[0030] Where: A is the spectral reflectance index, B is the terrain deformation rate, C is the lidar canopy height, a1 and a2 are weight values, and the values ​​of a1 and a2 are adjusted by the user, exp is the exponential function, max is the maximum value function, and ln is the natural logarithm function.

[0031] Preferably, the step S3 completes the calculation and analysis of the second data set through the following steps:

[0032] Step S3.1, performing integration calculation on the second data set to generate a geological deviation analysis coefficient DZF and a geological correction coefficient DZJ;

[0033] Step S3.2: Analyze the calculated geological deviation analysis coefficient DZF to generate a second analysis result, as follows:

[0034] when When , it means that the current geological data is normal and the geological correction coefficient DZJ does not need to be used in subsequent model analysis;

[0035] when When , it means that the current geological data fluctuates, and the geological correction coefficient DZJ needs to be used in subsequent model analysis.

[0036] Preferably, in step S3.1, the geological deviation analysis coefficient DZF and the geological correction coefficient DZJ are calculated by the following formula:

[0037] ;

[0038] ;

[0039] Where: D is the terrain complexity index, E is the vegetation coverage, F is the rainfall erosivity index, b1, b2 and b3 are proportional coefficients, and b1≠b2≠b3≠0, b1≠b2≠b3≠1, a3 and a4 are weight values, and the values ​​of b1, b2, b3, a3 and a4 are adjusted by the user.

[0040] Preferably, the step S4 completes the calculation and analysis of the third data set through the following steps:

[0041] Step S4.1, performing integration calculation on the third data set to generate a mineral reaction deviation analysis coefficient KWF and a mineral reaction correction coefficient KWJ;

[0042] Step S4.2: Analyze the calculated mineral reaction deviation analysis coefficient KWF to generate a third analysis result, as follows:

[0043] when When , it means that the current mineral data is normal and the mineral reaction correction coefficient KWJ does not need to be used in subsequent model analysis;

[0044] when When , it means that the current mineral data fluctuates, and the mineral reaction correction coefficient KWJ needs to be used in subsequent model analysis.

[0045] Preferably, in step S4.1, the mineral reaction deviation analysis coefficient KWF and the mineral reaction correction coefficient KWJ are calculated by the following formula:

[0046] ;

[0047] ;

[0048] Where: G is the shortwave infrared absorption depth, H is the thermal infrared emissivity anomaly index, I is the radar polarization scattering entropy, exp is the exponential function, max is the maximum value function, a5 is the weight value, K is the corrected attenuation coefficient, and a5 and K values ​​are adjusted by the user.

[0049] Preferably, the step S5 completes the calculation and analysis of the third data set through the following steps:

[0050] Step S4.1, integrating and calculating the remote sensing deviation analysis coefficient YGF, the remote sensing correction coefficient YGJ, the geological deviation analysis coefficient DZF, the geological correction coefficient DZJ, the mineral reaction deviation analysis coefficient KWF, and the mineral reaction correction coefficient KWJ, thereby generating a plurality of mineral marking guidance coefficients KWX;

[0051] Step S4.2: Analyze the calculated data to generate a first analysis result, as follows:

[0052] when When , it means the current mineral data is abnormal and no guidance mark is given;

[0053] when When , it means the current mineral data is normal and it is marked as a guide;

[0054] when When , it means the current mineral data is abnormal and no guidance mark is given;

[0055] when When , it means the current mineral data is normal and it is marked as a guide;

[0056] when When , it means the current mineral data is abnormal and no guidance mark is given;

[0057] when When , it means the current mineral data is normal and it is marked as a guide;

[0058] when When , it means the current mineral data is abnormal and no guidance mark is given;

[0059] when When , it means the current mineral data is normal and it is marked as a guide.

[0060] Preferably, in step S5.1, multiple mineral marking index coefficients KWX are obtained by calculation using the following formula:

[0061] ;

[0062] ;

[0063] ;

[0064] ;

[0065] ;

[0066] Where: YGJ is the remote sensing deviation analysis coefficient, YGJ is the remote sensing correction coefficient, DZF is the geological deviation analysis coefficient, DZJ is the geological correction coefficient, KWF is the mineral reaction deviation analysis coefficient, KWJ is the mineral reaction correction coefficient, exp is the exponential function, and K is the correction attenuation coefficient.

[0067] Compared with the prior art, the present invention has the following beneficial effects:

[0068] 1. This method integrates multi-source remote sensing data, geological data, and mineral reaction characteristic signals to construct an intelligent identification model based on deviation analysis and correction mechanisms. This model can dynamically adapt to different environmental conditions and significantly improve the accuracy and reliability of mineral resource exploration. Through dynamic correction and multi-coefficient integration, this method effectively reduces the misjudgment rate, improves the accuracy and precision of mineral target identification, and can quickly screen potential mining areas across a large area, reducing manual intervention and significantly improving exploration efficiency.

[0069] 2. Multiple mineral identification guidance coefficients KWX are generated by integrating the remote sensing bias analysis coefficient YGF, remote sensing correction coefficient YGJ, geological bias analysis coefficient DZF, geological correction coefficient DZJ, mineral reaction bias analysis coefficient KWF, and mineral reaction correction coefficient KWJ. By using weighted calculations of multiple parameters, this method comprehensively considers remote sensing, geological, and mineral reaction characteristics, comprehensively assessing mineral resource potential from multiple perspectives, and significantly improving the accuracy and stability of mineral identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 A diagram showing the steps of the method of the present invention. DETAILED DESCRIPTION

[0071] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0072] Example 1: Please refer to Figure 1 ,The intelligent identification method of mineral resources based on multi-source remote sensing data, the ,specific steps are as follows:

[0073] S1. Data collection: collecting multi-source data in the mineral identification process and preprocessing them to generate a first data set, a second data set, and a third data set;

[0074] S2. Remote sensing analysis: performing calculation and analysis on the first data set to generate a remote sensing deviation analysis coefficient YGF, a remote sensing correction coefficient YGJ, and a first analysis result;

[0075] S3, geological analysis, performing calculation and analysis on the second data set to generate a geological deviation analysis coefficient DZF and a geological correction coefficient DZJ and a second analysis result;

[0076] S4, mineral characteristic signal analysis, performing calculation and analysis on the third data set, thereby generating a mineral response deviation analysis coefficient KWF and a mineral response correction coefficient KWJ and a third analysis result;

[0077] S5. Integrate and calculate the remote sensing deviation analysis coefficient YGF, the remote sensing correction coefficient YGJ, the geological deviation analysis coefficient DZF, the geological correction coefficient DZJ, the mineral reaction deviation analysis coefficient KWF, and the mineral reaction correction coefficient KWJ to generate multiple mineral marking guidance coefficients KWX. Analyze the multiple mineral marking guidance coefficients KWX separately to generate a fourth analysis result. Based on the fourth analysis result, execute the mineral marking instruction on the target.

[0078] In this embodiment, in step S1, during the mineral resource identification process, multi-source data collection is first performed, including optical remote sensing data, radar remote sensing data, and lidar data. The collected data undergoes necessary preprocessing, specifically including radiometric correction, geometric registration, and noise filtering, to generate the first, second, and third datasets. This multi-source data preprocessing and standardization eliminates spatiotemporal inconsistencies and dimensional differences between the data, ensuring high data quality and providing a reliable foundation for subsequent analysis. This step effectively reduces potential noise and errors in the raw data, improves overall data quality, and lays the foundation for subsequent precise analysis.

[0079] In step S2, in the remote sensing analysis step, the remote sensing deviation analysis coefficient YGF and the remote sensing correction coefficient YGJ are generated by calculating and analyzing the first data set. YGF reflects the deviation between the remote sensing data and the actual observation value, and can reveal data quality problems. If YGF exceeds the set threshold, the data is corrected according to the remote sensing correction coefficient YGJ. Through this step, the remote sensing data can be denoised and corrected to eliminate interference such as atmospheric and sensor errors. This step significantly improves the reliability and accuracy of the remote sensing data by dynamically adjusting the deviation correction of the remote sensing data. The corrected remote sensing data can more accurately reflect the actual surface conditions, provide more accurate input data for the intelligent identification of mineral resources, and thus effectively reduce the risk of misidentification.

[0080] In step S3, in the geological analysis step, the second data set is calculated and analyzed to generate a geological deviation analysis coefficient DZF and a geological correction coefficient DZJ. DZF quantifies the impact of the geological environment on mineral identification by analyzing geological parameters such as terrain complexity, vegetation coverage, and rainfall erosion. When the DZF value exceeds the set threshold, the system generates a corresponding geological correction coefficient DZJ to optimize the model recognition capability. Through geological analysis, it is possible to accurately identify areas with complex geological environments, such as landslides, karst and other interference areas, to avoid misjudgment. In complex geological environments, this step can effectively eliminate the interference of unfavorable factors, making the mineral identification process more accurate and stable.

[0081] In step S4, mineral characteristic signal analysis is performed by calculating and analyzing the third data set to generate the mineral response deviation analysis coefficient KWF and the mineral response correction coefficient KWJ. KWF primarily measures the strength of the mineral characteristic signal and quantitatively analyzes the mineral response. If the KWF value exceeds the set threshold, KWJ will be corrected according to the specific situation. Through this step, the characteristic signal of the mineral can be more accurately identified, eliminating the interference of surface noise. This step can help accurately identify the mineral characteristic signal, eliminate interference from surface cover and climatic factors, and improve the accuracy of mineral classification and identification, especially in environments with low signal-to-noise ratios, and still achieve efficient identification.

[0082] In step S5, the aforementioned remote sensing deviation analysis coefficient YGF, remote sensing correction coefficient YGJ, geological deviation analysis coefficient DZF, geological correction coefficient DZJ, mineral reaction deviation analysis coefficient KWF, and mineral reaction correction coefficient KWJ are integrated and calculated to generate multiple mineral marking guidance coefficients KWX. Based on these mineral marking guidance coefficients, the target area is analyzed to generate the final mineral marking instructions. This step integrates multi-source data and correction information, providing efficient and accurate decision support for the intelligent identification of mineral resources. Through the fusion and analysis of multiple coefficients, it can integrate information from various data sources and improve the accuracy of target area identification. This method can effectively identify the spatial distribution of mineral resources and potential mineral deposits, greatly improving the efficiency and accuracy of mineral exploration.

[0083] Compared with traditional mineral resource identification methods, this method has made significant improvements in many aspects. Traditional methods often rely on a single data source and mostly use static models for mineral identification. They are easily affected by data quality issues and the complexity of the geological environment, resulting in unstable identification results. In contrast, this method integrates multi-source remote sensing data, geological data and mineral reaction characteristic signals to construct an intelligent identification model based on deviation analysis and correction mechanism. It can dynamically adapt to different environmental conditions and significantly improve the accuracy and reliability of mineral resource exploration. Through dynamic correction and multi-coefficient integration, this method effectively reduces the misjudgment rate, improves the identification accuracy and precision of mineral target areas, and can quickly screen out potential mining areas in a large area, reducing human intervention and significantly improving exploration efficiency.

[0084] Example 2: Please refer to Figure 1 ,Step S1 completes data collection and preprocessing through the following steps;

[0085] Step S1.1, identifying and extracting remote sensing data in the mineral identification process, geomorphological environment data collected by remote sensing, and mineral reaction data;

[0086] The remote sensing data include spectral reflectance index, terrain deformation rate, and lidar canopy height;

[0087] Geomorphic environmental data include terrain complexity index, vegetation coverage and rainfall erosivity index water;

[0088] Mineral reaction data include shortwave infrared absorption depth, thermal infrared emissivity anomaly index, and radar polarimetric scattering entropy;

[0089] The multi-source dataset consists of spectral reflectance index, terrain deformation rate, lidar canopy height, terrain complexity index, vegetation coverage, rainfall erosivity index, shortwave infrared absorption depth, thermal infrared emissivity anomaly index, and radar polarization scattering entropy;

[0090] Step S1.2, preprocessing and dimensionlessizing the multi-source data set, and reorganizing the preprocessed multi-source data set into a first data set, a second data set, and a third data set;

[0091] The first dataset includes the spectral reflectance index A, terrain deformation rate B, and lidar canopy height C;

[0092] The second data set includes terrain complexity index D, vegetation coverage E, and rainfall erosivity index F;

[0093] The third data set includes the shortwave infrared absorption depth G, the thermal infrared emissivity anomaly index H, and the radar polarization scattering entropy I.

[0094] In this embodiment, traditional methods typically rely on a single data source for mineral resource identification, which is susceptible to local data bias or noise. In step S1, by integrating remote sensing data such as spectral reflectance index, terrain deformation rate, and lidar canopy height, combined with geomorphic environment data and mineral reaction data, this method provides multi-dimensional, multi-source data support, effectively improving the accuracy and reliability of the identification process.

[0095] The processed data is reorganized into three independent, functionally specific datasets—the first, second, and third datasets—making data analysis more organized and structured. This division of the datasets not only facilitates subsequent calculations and analysis but also allows each dataset to focus on distinct mineral identification features, improving the specificity and accuracy of the analysis. For example, the first dataset focuses on remote sensing signals and topographic changes, the second on geological and vegetation characteristics, and the third on mineral response characteristics. This categorization allows for more efficient feature extraction and data processing.

[0096] By rationally partitioning and accurately preprocessing multi-source datasets, this step provides a high-quality data foundation for subsequent remote sensing, geological, and mineral characterization analyses. The processing and optimization of each dataset ensures the accuracy and relevance of each data source during subsequent analysis, enhancing the robustness of the overall identification method.

[0097] Example 3: Please refer to Figure 1 , step S2 completes the calculation and analysis of the first data set through the following steps;

[0098] Step S2.1, performing integration calculation on the first data set to generate a remote sensing bias analysis coefficient YGF and a remote sensing correction coefficient YGJ;

[0099] Step S2.2: Analyze the calculated remote sensing deviation analysis coefficient YGF to generate a first analysis result, as follows:

[0100] when When , it means that the current remote sensing data is normal and the remote sensing correction coefficient YGJ does not need to be used in subsequent model analysis;

[0101] when When , it means that the current remote sensing data fluctuates, and the remote sensing correction coefficient YGJ needs to be used in subsequent model analysis.

[0102] In step S2.1, the remote sensing deviation analysis coefficient YGF and the remote sensing correction coefficient YGJ are calculated using the following formulas;

[0103] ;

[0104] ;

[0105] Where: A is the spectral reflectance index, B is the terrain deformation rate, C is the lidar canopy height, a1 and a2 are weight values, and the values ​​of a1 and a2 are adjusted by the user, exp is the exponential function, max is the maximum value function, and ln is the natural logarithm function.

[0106] In this embodiment: Traditional methods usually use fixed thresholds or simple linear correction methods to deal with deviations in remote sensing data, which may lead to unstable model effects in complex geological environments. By introducing the remote sensing deviation analysis coefficient YGF and the remote sensing correction coefficient YGJ, this method can automatically determine whether correction is needed based on the quality fluctuations of the remote sensing data itself. Specifically, when YGF≤0.3, it indicates that the data quality is good and the model does not need additional correction; when YGF>0.3, it indicates that there are fluctuations in the data, and the model automatically uses the remote sensing correction coefficient YGJ for correction. This dynamic correction mechanism improves the intelligence and accuracy of data processing.

[0107] In step S2.1, the remote sensing bias analysis factor (YGF) is calculated by combining a polynomial calculation with an exponential function, taking into account multiple remote sensing parameters. This multi-dimensional, comprehensive calculation method enables a more comprehensive assessment of remote sensing data quality, particularly identifying biases caused by varying geological conditions and observation environments. The combined calculation based on spectral, topographic, and lidar data enables YGF to more accurately reflect data bias, providing a more precise basis for subsequent correction steps.

[0108] By combining the exponential function exp and the maximum function max, the correction coefficient YGJ can be dynamically adjusted based on the most critical parameters of the remote sensing data. This makes the correction coefficient more sensitive to important parameters and can be adjusted accordingly based on different data conditions. Compared with traditional linear correction schemes, this method can more flexibly cope with extreme fluctuations in the data, further improving the adaptability and accuracy of the model.

[0109] When calculating YGF and YGJ, the weights a1 and a2 are user-configurable, providing flexibility and personalized optimization options for practical applications. Users can customize weights and adjust the model's sensitivity based on the specific needs of different regions, mineral types, and exploration targets. This flexible adjustment makes this method widely applicable to different geological environments and mineral resource types, further enhancing the broad applicability of intelligent mineral resource identification.

[0110] This method effectively handles fluctuations or outliers that may occur in remote sensing data, especially when data is collected under complex geological conditions or inclement weather. The correction coefficient YGJ eliminates interference from unstable factors, making the final analysis results more robust and reliable. This intelligent correction of fluctuating data significantly improves the robustness and accuracy of the system, reducing misjudgments and missed detections caused by fluctuating data quality.

[0111] Example 4: Please refer to Figure 1 , step S3 completes the calculation and analysis of the second data set through the following steps;

[0112] Step S3.1, performing integration calculation on the second data set to generate a geological deviation analysis coefficient DZF and a geological correction coefficient DZJ;

[0113] Step S3.2: Analyze the calculated geological deviation analysis coefficient DZF to generate a second analysis result, as follows:

[0114] when When , it means that the current geological data is normal and the geological correction coefficient DZJ does not need to be used in subsequent model analysis;

[0115] when When , it means that the current geological data fluctuates, and the geological correction coefficient DZJ needs to be used in subsequent model analysis.

[0116] In step S3.1, the geological deviation analysis coefficient DZF and the geological correction coefficient DZJ are calculated using the following formulas;

[0117] ;

[0118] ;

[0119] Where: D is the terrain complexity index, E is the vegetation coverage, F is the rainfall erosivity index, b1, b2 and b3 are proportional coefficients, and b1≠b2≠b3≠0, b1≠b2≠b3≠1, a3 and a4 are weight values, and the values ​​of b1, b2, b3, a3 and a4 are adjusted by the user.

[0120] In this embodiment, traditional mineral resource identification methods may rely solely on a single geological data source and often fail to dynamically correct for data fluctuations. This step dynamically assesses the quality of geological data by calculating the geological deviation analysis coefficient, DZF. If DZF ≤ 0.7, the geological data is relatively stable, and subsequent models do not need to use the geological correction coefficient, DZJ. If DZF > 0.7, the geological data fluctuates and requires the introduction of the geological correction coefficient, DZJ. This intelligent assessment and correction mechanism not only effectively improves the quality control of geological data, but also reduces the instability caused by data fluctuations, enhancing the reliability of identification results.

[0121] In step S3.1, the geological deviation analysis coefficient DZF is derived through a comprehensive calculation of the terrain complexity index D, vegetation cover E, and rainfall erosivity index F. This comprehensive calculation method for multidimensional geological data considers multiple factors, including topography, vegetation, and rainfall. It can more comprehensively reflect the actual deviation of geological data and more accurately assess data quality, especially in complex geological environments. By setting appropriate coefficient ratios, this method ensures a comprehensive analysis of various factors and is more adaptable to the characteristics of different geological environments.

[0122] The scaling and weighting factors in this step are user-configurable. This customizable feature provides flexibility for exploration in different regions or mineral resource types. Users can adjust these factors to tailor the sensitivity and correction strategy of geological deviation analysis. This feature makes this method widely applicable, meeting data analysis requirements under various geological conditions.

[0123] When the DZF value exceeds 0.7, the geological correction factor (DZJ) is activated and adjusted based on the DZF deviation. Using a quadratic correction formula, DZJ can make more precise adjustments to the DZF deviation, ensuring that the correction factor more accurately reflects data fluctuations. Compared with traditional linear correction, this polynomial correction method can provide greater adaptability and correction accuracy in the face of large data fluctuations, thereby improving the reliability and accuracy of the entire model in complex geological environments.

[0124] This step, through comprehensive analysis of geological data, effectively identifies potential mineral resources in complex geological environments. In particular, when geological data fluctuates, the correction factor DZJ effectively eliminates the impact of these fluctuations on mineral resource identification. This adaptive correction mechanism significantly improves identification accuracy in complex geological conditions and avoids misidentification caused by variations in geological factors.

[0125] Example 5: Please refer to Figure 1, step S4, completing the calculation and analysis of the third data set through the following steps;

[0126] Step S4.1, performing integration calculation on the third data set to generate a mineral reaction deviation analysis coefficient KWF and a mineral reaction correction coefficient KWJ;

[0127] Step S4.2: Analyze the calculated mineral reaction deviation analysis coefficient KWF to generate a third analysis result, as follows:

[0128] when When , it means that the current mineral data is normal and the mineral reaction correction coefficient KWJ does not need to be used in subsequent model analysis;

[0129] when When , it means that the current mineral data fluctuates, and the mineral reaction correction coefficient KWJ needs to be used in subsequent model analysis.

[0130] In step S4.1, the mineral reaction deviation analysis coefficient KWF and the mineral reaction correction coefficient KWJ are calculated by the following formula;

[0131] ;

[0132] ;

[0133] Where: G is the shortwave infrared absorption depth, H is the thermal infrared emissivity anomaly index, I is the radar polarization scattering entropy, exp is the exponential function, max is the maximum value function, a5 is the weight value, K is the corrected attenuation coefficient, and a5 and K values ​​are adjusted by the user.

[0134] In this embodiment, traditional methods may fail to fully account for fluctuations and anomalies in mineral reaction data and often rely on static correction coefficients. By introducing the mineral reaction deviation analysis coefficient KWF and the mineral reaction correction coefficient KWJ, this method can dynamically assess mineral data quality and determine whether to enable correction based on fluctuations. Specifically, when KWF ≤ 0.4, the mineral data is considered normal and no correction is required for subsequent analysis. When KWF > 0.4, it indicates fluctuations in the mineral data, and the system automatically enables KWJ correction. This intelligent assessment and correction mechanism improves adaptability to fluctuations in mineral reaction data and ensures data quality control during the identification process.

[0135] In step S4.1, the calculation of the mineral response deviation analysis coefficient KWF involves a comprehensive analysis of the shortwave infrared absorption depth G, the thermal infrared emissivity anomaly index H, and the radar polarization scattering entropy I. This comprehensive calculation method of multi-dimensional mineral characteristics fully considers the reaction characteristics of different minerals, and improves the sensitivity of KWF to key mineral reaction characteristics through the maximum function max and the exponential decay exp function, ensuring a more accurate deviation assessment. This method effectively captures the changes and fluctuations in mineral data and can better cope with complex mineral reaction characteristics. By using the exponential decay correction mechanism, the mineral response correction coefficient KWJ can be adjusted according to the fluctuation of the KWF value, thereby achieving effective correction when the data fluctuates greatly. This exponential decay method is more flexible and accurate than the traditional linear correction method, especially for strongly fluctuating mineral data, and can provide stronger adaptability.

[0136] The weight coefficient a5 and the modified attenuation coefficient K in this step are user-adjustable. This custom setting allows the mineral resource identification process to be optimized based on the reaction characteristics of different minerals in different regions. Users can adjust these coefficients to better adapt to various mineral types and geological environments, improving the applicability and accuracy of the model.

[0137] During the mineral resource identification process, mineral reaction data can be affected by a variety of factors, leading to data fluctuations or anomalies. By introducing a dynamic correction mechanism, the system can automatically identify these fluctuations and take corrective measures to ensure the stability and accuracy of the analysis results. Especially under complex geological conditions, where mineral data fluctuations are often significant, this method can effectively improve the robustness and stability of the identification system, preventing data fluctuations from unnecessarily impacting the final results.

[0138] Example 6: Please refer to Figure 1 , step S5, completing the calculation and analysis of the third data set through the following steps;

[0139] Step S4.1, integrating and calculating the remote sensing deviation analysis coefficient YGF, the remote sensing correction coefficient YGJ, the geological deviation analysis coefficient DZF, the geological correction coefficient DZJ, the mineral reaction deviation analysis coefficient KWF, and the mineral reaction correction coefficient KWJ, thereby generating a plurality of mineral marking guidance coefficients KWX;

[0140] Step S4.2: Analyze the calculated data to generate a first analysis result, as follows:

[0141] when When , it means the current mineral data is abnormal and no guidance mark is given;

[0142] when When , it means the current mineral data is normal and it is marked as a guide;

[0143] when When , it means the current mineral data is abnormal and no guidance mark is given;

[0144] when When , it means the current mineral data is normal and it is marked as a guide;

[0145] when When , it means the current mineral data is abnormal and no guidance mark is given;

[0146] when When , it means the current mineral data is normal and it is marked as a guide;

[0147] when When , it means the current mineral data is abnormal and no guidance mark is given;

[0148] when When , it means the current mineral data is normal and it is marked as a guide.

[0149] In step S5.1, multiple mineral marking index coefficients KWX are calculated using the following formula;

[0150] ;

[0151] ;

[0152] ;

[0153] ;

[0154] ;

[0155] Where: YGJ is the remote sensing deviation analysis coefficient, YGJ is the remote sensing correction coefficient, DZF is the geological deviation analysis coefficient, DZJ is the geological correction coefficient, KWF is the mineral reaction deviation analysis coefficient, KWJ is the mineral reaction correction coefficient, exp is the exponential function, and K is the correction attenuation coefficient.

[0156] In this embodiment, this step generates multiple mineral identification guidance coefficients KWX by integrating the remote sensing bias analysis coefficient YGF, the remote sensing correction coefficient YGJ, the geological bias analysis coefficient DZF, the geological correction coefficient DZJ, the mineral reaction bias analysis coefficient KWF, and the mineral reaction correction coefficient KWJ. By using a weighted calculation of multiple parameters, this method comprehensively considers remote sensing, geological, and mineral reaction characteristics, comprehensively assessing mineral resource potential from multiple perspectives, and significantly improving the accuracy and stability of mineral identification.

[0157] The formula for calculating the mineral marker index KWX utilizes an exponential decay function and a maximum function (max), enabling more flexible and accurate corrections for diverse data. In particular, the introduction of the exponential correction (exp) effectively mitigates the adverse effects of biases in remote sensing, geological, and mineralogical reaction data, thereby avoiding overcorrection or neglect of key factors. The maximum function ensures that the most significant factor among multiple input conditions dominates the correction process, thereby more accurately reflecting the characteristics of the mineral data.

[0158] After generating the mineral marking guidance coefficients KWX, the system performs marking analysis based on the set thresholds. When KWX1, KWX2, KWX3, KWX4, and KWX5 are less than 0.4, the current mineral data is abnormal and no guidance marking is performed. However, when the coefficients are greater than or equal to 0.4, the mineral data is normal and guidance marking is required. This dynamic threshold determination mechanism effectively avoids mislabeling or misleading, ensuring a high degree of confidence in the final identified mineral target area.

[0159] Multiple attenuation coefficients, K, and correction parameters are used to calculate the mineral identification index. These attenuation coefficients can be flexibly configured by the user based on actual needs. This provides powerful flexibility and adaptability for intelligent identification of different mineral types and geological environments. By adjusting the attenuation coefficients, the system can be optimized based on the characteristics of different regions or mineral resources, improving the accuracy of mineral identification.

[0160] This step provides more comprehensive data processing capabilities by comprehensively considering remote sensing data, geological environment data, and mineral reaction characteristics. Each mineral identification index is generated through a weighted calculation of multiple data sources, comprehensively covering the multidimensional characteristics of the mineral resource. This not only improves the accuracy of mineral identification, but also enhances the system's recognition capabilities in complex geological conditions and diverse mineral characteristics.

[0161] Through the calculations and analysis in this step, the system can accurately identify potential areas for mineral resources in a relatively short period of time. This optimization process helps prospectors quickly identify important target areas, improves resource exploration efficiency and accuracy, and reduces the impact of human judgment errors.

[0162] The contents not described in detail in this specification belong to the prior art known to those skilled in the art.

[0163] Although the present invention has been described in detail with reference to the aforementioned embodiments, it is still possible for those skilled in the art to modify the technical solutions described in the aforementioned embodiments, or to make equivalent substitutions for some of the technical features therein. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent identification method for mineral resources based on multi-source remote sensing data, characterized by: The specific steps are as follows: S1. Data collection: collecting multi-source data in the mineral identification process and preprocessing them to generate a first data set, a second data set, and a third data set; Step S1 completes data collection and preprocessing through the following steps; Step S1.1, identifying and extracting remote sensing data in the mineral identification process, geomorphological environment data collected by remote sensing, and mineral reaction data; The remote sensing data include spectral reflectance index, terrain deformation rate, and lidar canopy height; Geomorphic environmental data include terrain complexity index, vegetation coverage and rainfall erosivity index water; Mineral reaction data include shortwave infrared absorption depth, thermal infrared emissivity anomaly index, and radar polarimetric scattering entropy; The multi-source dataset consists of spectral reflectance index, terrain deformation rate, lidar canopy height, terrain complexity index, vegetation coverage, rainfall erosivity index, shortwave infrared absorption depth, thermal infrared emissivity anomaly index, and radar polarization scattering entropy; Step S1.2, preprocessing and dimensionlessizing the multi-source data set, and reorganizing the preprocessed multi-source data set into a first data set, a second data set, and a third data set; The first data set includes a spectral reflectance index A, a terrain deformation rate B, and a lidar canopy height C; The second data set includes terrain complexity index D, vegetation coverage E, and rainfall erosivity index F; The third data set includes shortwave infrared absorption depth G, thermal infrared emissivity anomaly index H, and radar polarization scattering entropy I; S2. Remote sensing analysis: performing calculation and analysis on the first data set to generate a remote sensing deviation analysis coefficient YGF, a remote sensing correction coefficient YGJ, and a first analysis result; The step S2 completes the calculation and analysis of the first data set through the following steps; Step S2.1, performing integration calculation on the first data set to generate a remote sensing bias analysis coefficient YGF and a remote sensing correction coefficient YGJ; Step S2.2: Analyze the calculated remote sensing deviation analysis coefficient YGF to generate a first analysis result, as follows: when When , it means that the current remote sensing data is normal and the remote sensing correction coefficient YGJ does not need to be used in subsequent model analysis; when When , it means that the current remote sensing data fluctuates, and the remote sensing correction coefficient YGJ needs to be used in subsequent model analysis; S3, geological analysis, performing calculation and analysis on the second data set to generate a geological deviation analysis coefficient DZF and a geological correction coefficient DZJ and a second analysis result; S4, mineral characteristic signal analysis, performing calculation and analysis on the third data set, thereby generating a mineral response deviation analysis coefficient KWF and a mineral response correction coefficient KWJ and a third analysis result; S5. Integrate and calculate the remote sensing deviation analysis coefficient YGF, the remote sensing correction coefficient YGJ, the geological deviation analysis coefficient DZF, the geological correction coefficient DZJ, the mineral reaction deviation analysis coefficient KWF, and the mineral reaction correction coefficient KWJ to generate multiple mineral marking guidance coefficients KWX. Analyze the multiple mineral marking guidance coefficients KWX separately to generate a fourth analysis result. Based on the fourth analysis result, execute the mineral marking instruction on the target.

2. The method for intelligent identification of mineral resources based on multi-source remote sensing data according to claim 1, characterized in that: In step S2.1, the remote sensing deviation analysis coefficient YGF and the remote sensing correction coefficient YGJ are calculated by the following formula; ； ； Where: A is the spectral reflectance index, B is the terrain deformation rate, C is the lidar canopy height, a1 and a2 are weight values, and the values ​​of a1 and a2 are adjusted by the user, exp is the exponential function, max is the maximum value function, and ln is the natural logarithm function.

3. The method for intelligent identification of mineral resources based on multi-source remote sensing data according to claim 2, characterized in that: The step S3 completes the calculation and analysis of the second data set through the following steps; Step S3.1, performing integration calculation on the second data set to generate a geological deviation analysis coefficient DZF and a geological correction coefficient DZJ; Step S3.2: Analyze the calculated geological deviation analysis coefficient DZF to generate a second analysis result, as follows: when When , it means that the current geological data is normal and the geological correction coefficient DZJ does not need to be used in subsequent model analysis; when When , it means that the current geological data fluctuates, and the geological correction coefficient DZJ needs to be used in subsequent model analysis.

4. The method for intelligent identification of mineral resources based on multi-source remote sensing data according to claim 3 is characterized in that: In step S3.1, the geological deviation analysis coefficient DZF and the geological correction coefficient DZJ are calculated by the following formula; ； ； Where: D is the terrain complexity index, E is the vegetation coverage, F is the rainfall erosivity index, b1, b2 and b3 are proportional coefficients, and b1≠b2≠b3≠0, b1≠b2≠b3≠1, a3 and a4 are weight values, and the values ​​of b1, b2, b3, a3 and a4 are adjusted by the user.

5. The method for intelligent identification of mineral resources based on multi-source remote sensing data according to claim 4 is characterized in that: The step S4 is to complete the calculation and analysis of the third data set through the following steps; Step S4.1, performing integration calculation on the third data set to generate a mineral reaction deviation analysis coefficient KWF and a mineral reaction correction coefficient KWJ; Step S4.2: Analyze the calculated mineral reaction deviation analysis coefficient KWF to generate a third analysis result, as follows: when When , it means that the current mineral data is normal and the mineral reaction correction coefficient KWJ does not need to be used in subsequent model analysis; when When , it means that the current mineral data fluctuates, and the mineral reaction correction coefficient KWJ needs to be used in subsequent model analysis.

6. The method for intelligent identification of mineral resources based on multi-source remote sensing data according to claim 5, characterized in that: In step S4.1, the mineral reaction deviation analysis coefficient KWF and the mineral reaction correction coefficient KWJ are calculated by the following formula; ； ； Where: G is the shortwave infrared absorption depth, H is the thermal infrared emissivity anomaly index, I is the radar polarization scattering entropy, exp is the exponential function, max is the maximum value function, a5 is the weight value, K is the corrected attenuation coefficient, and a5 and K values ​​are adjusted by the user.

7. The method for intelligent identification of mineral resources based on multi-source remote sensing data according to claim 6, characterized in that: The step S5 completes the calculation and analysis of the third data set through the following steps; Step S4.1, integrating and calculating the remote sensing deviation analysis coefficient YGF, the remote sensing correction coefficient YGJ, the geological deviation analysis coefficient DZF, the geological correction coefficient DZJ, the mineral reaction deviation analysis coefficient KWF, and the mineral reaction correction coefficient KWJ, thereby generating a plurality of mineral marking guidance coefficients KWX; Step S4.2: Analyze the calculated data to generate a first analysis result, as follows: when When , it means the current mineral data is abnormal and no guidance mark is given; when When , it means the current mineral data is normal and it is marked as a guide; when When , it means the current mineral data is abnormal and no guidance mark is given; when When , it means the current mineral data is normal and it is marked as a guide; when When , it means the current mineral data is abnormal and no guidance mark is given; when When , it means the current mineral data is normal and it is marked as a guide; when When , it means the current mineral data is abnormal and no guidance mark is given; when When , it means the current mineral data is normal and it is marked as a guide.

8. The method for intelligent identification of mineral resources based on multi-source remote sensing data according to claim 7, characterized in that: In step S5.1, multiple mineral marking index coefficients KWX are calculated using the following formula; ； ； ； ； ； Where: YGJ is the remote sensing deviation analysis coefficient, YGJ is the remote sensing correction coefficient, DZF is the geological deviation analysis coefficient, DZJ is the geological correction coefficient, KWF is the mineral reaction deviation analysis coefficient, KWJ is the mineral reaction correction coefficient, exp is the exponential function, and K is the correction attenuation coefficient.

Images