Method for carrying out mineral and element relation analysis on gold-bearing mineral based on image matching and data fusion and application thereof
By integrating AMICS and EPMA, the occurrence state analysis of trace elements in high-arsenic and high-antimony gold ores and the visualization reconstruction of mineral-element relationships were realized, solving the problems of mineral identification and element distribution in existing technologies and improving the accuracy and efficiency of mineral processing.
Patent Information
- Application Number
- CN202511242812.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-09-02
- Publication Date
- 2026-01-20
AI Technical Summary
Existing technologies struggle to accurately identify the occurrence and distribution of trace elements in high-arsenic and high-antimony gold ores, hindering the rapid and accurate characterization of minerals with a wide field of view. In particular, the identification of the symbiotic relationships of elements such as Au, As, Sb, Bi, and Te is difficult, affecting the optimization of mineral processing and the recovery of metal resources.
By employing an integrated approach combining AMICS and EPMA, and through image matching and data fusion, a multidimensional detection technique of "image-element-space" is constructed. By combining AMICS mineral atlases with EPMA elemental information, trace element relationship analysis and visualization reconstruction of mineral-element correspondence mechanisms are achieved.
It enables two-dimensional/quasi-three-dimensional distribution modeling of key elements such as Au and As, meeting the four-dimensional data requirements of digital mines for "mineral-element-space-structure", supporting the construction of intelligent mineral exploration modules, and improving the guidance of beneficiation and metallurgy processes and the efficiency of metal resource recovery.
Smart Images

Figure 1OUXOFEBNANOJBTFEPCOU45TI0EXXROUDXB42ER5 
Figure 5ZCZRBLHQZC1MBYYYWVKSQFCUQEOZK41PUFNDVLT 
Figure 8Q75AFEI3EYH1PBZGLSHISGTS4HCJVYINB5IXBXX
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the cross field of mineralogy and geochemistry, in particular to a method for analyzing the relationship between minerals and elements of gold-bearing minerals based on image matching and data fusion and application thereof. BACKGROUND
[0002] Gold is an important strategic mineral resource in China, with high economic value and strategic reserve significance. With the gradual depletion of shallow and easily processed gold mines, gold resource development and utilization is shifting towards deep and refractory gold mines.
[0003] High-arsenic and high-antimony gold ores are widely found in polymetallic complex deposits. The ore-forming material composition is complex, and the occurrence state of trace metals is diverse. In particular, elements such as Au, As, Sb, Bi, and Te are often finely disseminated, intergrown, or wrapped in a state, which is difficult to accurately and quickly identify the occurrence of minerals and element distribution using conventional methods, severely restricting the optimization of beneficiation processes and efficient recovery of metal resources.
[0004] Currently, commonly used detection methods such as inductively coupled plasma mass spectrometry (ICP-MS) and X-ray fluorescence spectroscopy (XRF) have the following technical defects: 1. Lack of spatial distribution information: ICP-MS and XRF can only provide the element content of the whole sample, and cannot provide the occurrence state and distribution of elements in different minerals; 2. Cannot be phase analyzed: cannot identify which mineral the metal element is specifically assigned to, which is not conducive to "mineral-element" integrated evaluation; 3. Difficulty in identifying complex intergrowth relationship: cannot distinguish whether the element is independently assigned or wrapped in a certain host mineral, making it difficult to guide subsequent beneficiation processes.
[0005] In recent years, automatic mineral quantitative identification system (AMICS) and electron probe microanalysis (EPMA) have been gradually applied in geology and beneficiation analysis. AMICS has the ability to automatically identify mineral types, draw mineral distribution maps, and quantitatively analyze minerals, but its quantitative accuracy of trace elements is limited. EPMA has high spatial resolution and high precision element quantification capability, but the detection range is limited, making it difficult to quickly identify target areas in large area samples. Therefore, there is currently no technology that can simultaneously achieve large field mineral identification, spatial positioning, and trace element occurrence state interpretation, which cannot meet the needs of rapid and accurate characterization of complex trace elements in high-arsenic and high-antimony gold ores.
[0006] Therefore, it is necessary to develop an integrated and intelligent "image-element-space" multi-dimensional detection and analysis technology based on AMICS and EPMA for the occurrence analysis of trace elements in high-arsenic and high-antimony gold ores and the visualization reconstruction of the "mineral-element" correspondence mechanism. SUMMARY
[0007] The present application aims at overcoming the deficiencies of the prior art, and provides a method for analyzing the mineral and element relationship of gold-bearing minerals based on image matching and data fusion, which fuses AMICS mineral atlas and EPMA element information, can realize the relationship analysis of trace elements in gold mines and the visualization reconstruction of the corresponding mechanism of "mineral-element", and provides effective data and image model for the micro-ore-forming information demand in digital mine construction.
[0008] The purpose of the present application is achieved by the following technical solutions: In one aspect, the present application provides a method for analyzing the mineral and element relationship of gold-bearing minerals based on image matching and data fusion, comprising the following steps: S11. Ore sample processing: selecting a gold-bearing ore sample, first crushing to a particle size of 2-5 mm, then second crushing to a particle size of less than 200 microns, to obtain ore powder; S12. Sample surface preparation: mixing the ore powder into epoxy resin for degassing and curing, then surface treatment to obtain mirror effect, and then carbon film sputtering to obtain the sample surface; S13. AMICS calibration of gold-bearing mineral image: using a scanning electron microscope to collect the BSE image of the sample surface, and automatically identifying the minerals through energy spectrum analysis, then marking the gold-bearing mineral types in combination with AMICS software, outputting the attribute data of different minerals, and retaining the image spatial coordinates to obtain the AMICS image; S14. Fixed-point EPMA and EDS scanning of trace elements: selecting gold-bearing mineral particles on the mineral distribution map containing the coordinate system to use an electron probe analyzer for fixed-point EPMA trace element analysis, analyzing target elements, and obtaining the distribution changes of the target elements in and between particles through EDS scanning, to obtain EPMA data and EPMA image respectively; S15. Image-level matching and spatial data fusion: spatially matching the AMICS image and the EPMA image, then data superposition and coordinate annotation on the picture after spatial matching of the EPMA data, to make the EPMA data and the EPMA image coincide with the AMICS image, to obtain a gold-bearing mineral-element fusion image; S16. Analysis of the relationship between gold-bearing minerals and elements: screening and characteristic statistics of gold-bearing minerals according to the gold-bearing mineral-element fusion image, and determining the gold occurrence type, then performing element paragenesis analysis to obtain the quantitative relationship between gold and elements such as arsenic in the gold-bearing minerals.
[0009] In the technical scheme, the method of the application upgrades the traditional point analysis to the continuous spatial expression of the mineral unit through image coordinate registration, probe data fusion and element visualization modeling, effectively realizes the two-dimensional / quasi-three-dimensional distribution modeling of key elements such as Au and As, and meets the needs of the digital mine for the four-dimensional data of “mineral-element-space-structure”; with the aid of image recognition and spatial registration technology, the AMICS mineral map, the EPMA element map and the probe chemical data are fused in the same coordinate system, which lays a data foundation for realizing the unified data structure management (such as database / platform modeling) in the digital mine platform, and is beneficial to the construction of the intelligent prospecting module.
[0010] Further, the gold-containing ore sample includes a high-arsenic and high-antimony gold ore sample.
[0011] Further, the scanning electron microscope includes a high-resolution field emission scanning electron microscope with an AMICS mineral analysis module.
[0012] Further, the gold-containing mineral types include pyrite, arsenopyrite, stibnite and native gold.
[0013] Further, the gold-containing mineral particles include pyrite, arsenopyrite, stibnite and native gold.
[0014] Further, the target elements include Au, As, Sb and Ag.
[0015] Further, the carbon film thickness of the sample surface is 15-25 nm.
[0016] Further, the sample obtained after the screening includes mineral particles with an Au content greater than 0.01 wt%.
[0017] Further, the attribute data includes spatial position, particle size and surrounding inclusion relationship.
[0018] Further, the EDS scanning includes at least one of line scanning and area scanning.
[0019] Further, the characteristic statistics include statistics of the average value of the Au content, the average value of the As content, the particle size range and the dissociation degree.
[0020] Further, the element paragenesis analysis includes linear relationship analysis and nonlinear trend analysis.
[0021] Illustratively, the linear relationship analysis employs a tool including Pearson correlation coefficient analysis.
[0022] Illustratively, the nonlinear trend analysis employs a tool including Spearman rank correlation coefficient analysis.
[0023] In the technical scheme, the method can realize quantitative relationship analysis of gold and arsenic in gold-bearing minerals through Pearson and Spearman double correlation analysis at a micro scale, can reveal enrichment rules and partition behavior of gold in different carriers, and provides a new idea for element behavior geochemistry research.
[0024] Further, the parameters of the scanning electron microscope are 15 kV of acceleration voltage, 2 nM of beam current, and high vacuum mode.
[0025] Further, the parameters of the electron probe analyzer are 15 kV of acceleration voltage, 20 nA of beam current, and LIF, TAP and PET of analysis crystal.
[0026] Further, the method for spatial matching of the AMICS image and the EPMA image comprises the following steps: S21. The image analysis software is used to open the AMICS image and the EPMA image respectively, and the same pixel resolution and canvas size are adjusted, and the sizes of the two images are made consistent; S22. The AMICS image is set as the bottom image, and the EPMA image is superimposed as a layer, and the image is fine-tuned according to the feature points in the image to realize image coincidence; S23. The opacity of the AMICS image is adjusted to ensure that the details of the AMICS image and the EPMA image can be observed simultaneously.
[0027] Illustratively, the image analysis software comprises ImageJ.
[0028] In one aspect, the application provides the above method for automatic identification and type classification of gold particles in gold-bearing ores.
[0029] In one aspect, the application provides the application of the above method in digital mine technology services, including importing at least one image of the mineral identification map, the spatial element map and the occurrence point annotation map obtained by the method into the digital mine platform to improve the richness and authenticity of digital mine modeling.
[0030] In the technical scheme, through intelligent discrimination of gold types (free gold, inclusions and solid solution gold) and distribution statistics, the method provided by the application can provide reference basis of “mineral-structure-element” ternary relationship for the beneficiation module (such as flotation simulation and process path decision) in the digital mine, realize front-end data support for mineral processing path; and the method of the application can be linked with the digital mine modeling platform to provide real image output results such as mineral identification map, spatial element map and occurrence point annotation map, enrich the model display dimension, and make the model have the visual and credible properties of “source traceable and logic verifiable”.
[0031] The beneficial effects of the present application are: 1、The present application first constructs a general registration and fusion mechanism for AMICS images and EPMA images, realizes spatial unification of images from different sources through low-threshold image superposition and coordinate calibration technology, and provides key method support for micro-scale mineral-element relationship construction.
[0032] 2、The method of the present application comprehensively combines image interpretation and verification test results, establishes a determination rule based on spatial distribution, composition characteristics and occurrence morphology, constructs a "image + probe" integrated gold-containing particle determination process, and realizes automatic identification and type classification of gold particles.
[0033] 3、The whole process of the present application is developed based on common image and data processing software (such as ImageJ, Excel, PowerPoint, etc.), does not require expensive professional systems, is convenient to use under different experimental conditions, and is suitable for exploration evaluation and scientific research of various metal ores.
[0034] 4、The data and image results generated by the method of the present application are easy to import into the digital mine platform, can realize modularized service, and have important practical significance and broad application scenarios for promoting the transformation and upgrading of digital mine high-end technical service. BRIEF DESCRIPTION OF DRAWINGS
[0035] Figure 1 The sample surface prepared for Example 1 of the present application; Figure 2 The AMICS image obtained by AMICS analysis of the ore in Example 2 of the present application; Figure 3 The EPMA image obtained after EPMA fixed-point scanning in Example 3 of the present application; Figure 4 The AMICS and EPMA superimposed image obtained after registration of the AMICS image and the EPMA image in Example 4 of the present application; Figure 5 The gold-containing mineral-element fusion image obtained after probe data superposition and coordinate annotation of the AMICS and EPMA superimposed image in Example 4 of the present application. DETAILED DESCRIPTION
[0036] The technical solutions of the present application will be described in further detail below in conjunction with the drawings, but the protection scope of the present application is not limited to the following description.
[0037] Example 1 Sample treatment The sample is the Sawayaerdun gold mine in Wujia, Xinjiang, and the multi-element analysis results are shown in Table 1.
[0038] Table 1 Multi-element analysis results of the sample ore Sample preparation: The gold ore sample of the Savayalton gold mine in Wujia, Xinjiang was coarsely crushed to about 2-5 mm using a jaw crusher, and then crushed to a particle size of less than 200 μm after being divided, to obtain a mineral powder; Sample inlay: The mineral powder was uniformly mixed into an epoxy resin, degassed and cured under vacuum to obtain a resin inlay sample block, ensuring that the surface of the resin inlay sample block was free of bubbles; Surface treatment: The surface of the resin inlay sample block was treated with precision grinding and polishing sandpaper and aluminum oxide polishing paste until the surface of the resin inlay sample block appeared a mirror effect; Carbon film sputtering: A conductive carbon film with a thickness of about 15-25 nm was sputtered using a carbon evaporation instrument to ensure uniform response of the sample surface to the electron beam, suitable for scanning electron microscopy (SEM) and EPMA observation, and the obtained sample surface is as shown in Figure 1 .
[0039] For example, in the subsequent examples, the thickness of the conductive carbon film is 20 nm.
[0040] Example 2 Image calibration of gold-bearing minerals Equipment: High-resolution field emission scanning electron microscope (Zeiss Sigma300) with AMICS mineral analysis module.
[0041] Equipment parameters: Accelerating voltage: 15 kV; beam current: 2 nA; vacuum mode: high vacuum.
[0042] Using SEM combined with energy dispersive spectroscopy (EDS) spectrum, a high-resolution BSE image of the entire sample surface prepared in Example 1 was collected, and then mineral automatic identification was performed by EDS to obtain the mineral composition and content of the ore (Table 2), and combined with the AMICS software, the types of gold-bearing minerals (pyrite, arsenopyrite, stibnite, natural gold) were marked, the spatial position, particle size, surrounding inclusion relationship, etc. attribute data of each type of mineral were output, and the image spatial coordinates were saved as the basis for subsequent fusion, to obtain the AMICS image Figure 2 .
[0043] Table 2 Mineral composition and content table of ore Example 3 EPMA fixed-point scanning and EDS scanning Equipment: Electron probe analyzer (EPMA-1720).
[0044] Equipment parameters: Accelerating voltage: 15 kV; beam current: 20 nA; analysis crystal: LIF, TAP, PE.
[0045] Based on the AMICS image obtained in Example 2, gold-bearing mineral particles were selected for point EPMA microelement scanning, and target elements Au, As, Sb, Co, Ag, Ni, etc. in different mineral particles were analyzed to obtain the analysis results as shown in Table 3, and line scanning or area scanning was performed in cooperation with EDS to obtain the distribution changes of elements in the particles and between the particles to obtain the EPMA images as shown in Figure 3 .
[0046] Table 3 Electronic probe test results of gold-bearing mineral particles (%) Example 4 Image registration and data fusion processing (1) Image preprocessing and standardization The AMICS image and the EPMA image were opened using ImageJ, and the pixel resolution was adjusted to be the same (0.1 μm per pixel), and the image canvas size was adjusted to make the size and direction of the two images consistent.
[0047] (2) Image coarse registration The AMICS image was set as the base map, and then the EPMA image was superimposed on it as a layer. Manual fine adjustment was performed using feature points such as gold particle points, arsenopyrite corner points, and fracture intersection points in the image. After the images were visually overlapped, the opacity of the AMICS image was adjusted until the details of the two images could be observed simultaneously, obtaining the AMICS and EPMA superimposed image as shown in Figure 4 , and recording the image size, scaling ratio and other parameters.
[0048] (3) Probe data superposition and coordinate labeling The probe test data was sorted using Excel software, and the fields included number (such as point 1, point 2, etc.), measured coordinates (in μm), and element concentration (Au, As, etc.), obtaining the gold-bearing mineral data integration table as shown in Table 4. Then, the positions of each probe point were labeled in the gold-bearing mineral-element fusion image according to the gold-bearing mineral data integration table, and the types of minerals where each probe point was located were marked, obtaining the gold-bearing mineral-element fusion image Figure 5 .
[0049] Table 4 Gold-bearing mineral data integration table Example 5 Mineral-element relationship and element paragenesis analysis (1) Gold-bearing mineral screening and characteristic analysis According to the data integration table of gold-bearing minerals obtained in Example 4, the mineral particle size with Au content greater than 0.01 wt% was extracted, and then classified and counted according to the mineral type, the average values of Au and As contents in different types of minerals were calculated, and the particle size range and resolution index were summarized, to obtain the gold-bearing mineral statistical table as shown in Table 5.
[0050] Table 5 Gold-bearing mineral statistical table (2) Gold element occurrence type determination According to the gold-bearing mineral-element fusion diagram obtained in Example 4, the occurrence state and occurrence type of gold element were determined.
[0051] Granular gold: Au is distributed in the form of independent particles, without significant inclusion, the image boundary is clear, and the concentration is extremely high; Inclusion gold: Au is included by pyrite or arsenopyrite, distributed in the crystal boundary or inside; Solid solution gold: Au is uniformly distributed in the lattice of pyrite or arsenopyrite, with low concentration and no obvious image signal; Fracture gold: Au occurs along the mineral fracture, commonly found in microcracks of pyrite crystal, and coexists with Fe and S anomalies.
[0052] (3) Element paragenesis analysis a. Linear relationship: Pearson correlation coefficient analysis Pearson correlation coefficient (usually represented as r) measures the degree of linear correlation between two continuous variables.
[0053] Formula: Where X̄ and Ȳ are the mean values of X and Y respectively; X i and Y i are single observation values; r value interval correlation explanation: 0.8-1 or -0.8-1 strong correlation; 0.6-0.8 or -0.6- -0.8 strong correlation; 0.4-0.6 or -0.4- -0.6 moderate correlation; 0.2-0.4 or -0.2- -0.4 weak correlation; 0-0.2 or -0.2-0 very weak or no correlation.
[0054] The Pearson correlation coefficient was calculated by Excel, and then the function =CORREL(array1, array2) was used to analyze the content of gold element and arsenic element by mineral type grouping, and the results are shown in Table 6.
[0055] Table 6 Pearson correlation coefficient analysis results It can be seen that the Au-As correlation in pyrite is weak, indicating that Au in pyrite may exist in the form of solid solution gold; the significant negative correlation in arsenopyrite reflects the influence of element partitioning or inclusion state in the crystal on Au content.
[0056] b. Nonlinear trend: Spearman rank correlation coefficient analysis Spearman rank correlation coefficient (usually denoted as ρ) measures the strength of monotonic relationship between two variables. It is based on the ranking order of variables rather than the original data values, so it is more robust to outliers and nonlinear relationships.
[0057] Formula: Where d i is the rank difference of each pair of observations; n is the total number of pairs of observations.
[0058] ρ value interval correlation explanation: 0.8-1 or -0.8-1 very strong correlation; 0.6-0.8 or -0.6- -0.8 strong correlation; 0.4-0.6 or -0.4- -0.6 moderate correlation; 0.2-0.4 or -0.2- -0.4 weak correlation; 0-0.2 or -0.2-0 very weak or no correlation.
[0059] For some small samples or nonlinear trends, Spearman correlation analysis of Au and As was performed using Excel, and the results are shown in Table 7.
[0060] Table 7 Results of Spearman rank correlation coefficient analysis Spearman analysis can identify nonlinear covariation relationships, and is more applicable to small sample data and Au abnormal concentration points. Combining the results of Spearman analysis with Pearson analysis can help determine whether there is a complex occurrence trend.
[0061] Therefore, it can be judged that the natural gold in the Sawayaerdun gold deposit in Wushi, Xinjiang, mainly exists in the free state, with very high Au content and low As content; the linear and rank correlation between pyrite and Au are both very low, indicating that Au exists in the form of solid solution; Au and As in arsenopyrite show a moderate negative correlation. Combined with the analysis of gold-bearing mineral-element fusion diagram, it can be inferred that Au mainly exists in the form of inclusion gold or fissure gold.
[0062] In summary, the method of the present application structures, images and models the micro-mineral-element relationship in the digital mine background, and promotes the "trinity of multi-source data fusion, image-data linkage and visualization expression" technology system. With its highly integrated image-data fusion mechanism, practical micro-area analysis process and good scalability, it has important practical significance and broad application prospects for promoting the transformation and upgrading of China's digital mine high-end technology services.
Claims
1. A method for analyzing the mineral-element relationship of gold-bearing minerals based on image matching and data fusion, characterized in that, Includes the following steps: S11. Mineral sample processing: Select a gold-bearing ore sample, first crush it to a particle size of 2~5 mm and then reduce it, then crush it to a particle size of less than 200 μm to obtain mineral powder, mix it into epoxy resin for degassing and curing, then perform surface treatment to obtain a mirror effect, and then perform carbon film spraying to obtain the sample surface. S12. AMICS calibration of gold-bearing mineral images: BSE images of the sample surface are acquired using a scanning electron microscope, and minerals are automatically identified by energy dispersive spectroscopy. Then, AMICS software is used to label the types of gold-bearing minerals, output the attribute data of different minerals, and retain the spatial coordinates of the images to obtain AMICS images. S13. Point-to-point EPMA and EDS scanning of trace elements: Gold-bearing mineral particles are selected on the mineral distribution map containing the coordinate system and point-to-point EPMA trace element analysis is performed using an electron probe analyzer to analyze the target elements. The distribution changes of the target elements within and between particles are obtained by the EDS scanning, and EPMA data and EPMA images are obtained respectively. S14. Image-level matching and spatial data fusion: Spatial matching is performed between the AMICS image and the EPMA image. Then, the EPMA data is superimposed and coordinate-labeled on the spatially matched image so that the EPMA data and the EPMA image coincide with the AMICS image, resulting in a gold-bearing mineral-element fusion image. S15. Analysis of the relationship between gold-bearing minerals and elements: Based on the gold-bearing mineral-element fusion diagram, gold-bearing minerals are screened and their characteristics are statistically analyzed, and the type of gold occurrence is determined. Then, elemental co-occurrence analysis is performed to obtain the quantitative relationship between gold and elements such as arsenic in the gold-bearing minerals.
2. The method according to claim 1, characterized in that, The gold-bearing ore samples include high-arsenic and high-antimony gold ore samples; And / or, the scanning electron microscope includes a high-resolution field emission scanning electron microscope with an AMICS mineral analysis module; And / or, the gold-bearing mineral types include pyrite, arsenopyrite, stibnite, and native gold; And / or, the gold-bearing mineral particles include pyrite, arsenopyrite, stibnite, and native gold; And / or, the target element includes Au, As, Sb, and Ag.
3. The method according to claim 1, characterized in that, The thickness of the carbon film on the surface of the sample is 15~25 nm; And / or, the sample obtained after screening includes mineral particles with an Au content greater than 0.01 wt%.
4. The method according to claim 1, characterized in that, The attribute data includes spatial location, granularity, and surrounding enclosing relationships; And / or, the EDS scan includes at least one of line scan and area scan; And / or, the feature statistics include statistics on features such as average Au content, average As content, particle size range, and degree of dissociation; And / or, the element symbiosis analysis includes linear relationship analysis and nonlinear trend analysis.
5. The method according to claim 4, characterized in that, The linear relationship analysis employed tools including Pearson correlation coefficient analysis. And / or, the tools used in the nonlinear trend analysis include Spearman's rank correlation coefficient analysis.
6. The method according to claim 1, characterized in that, The parameters of the scanning electron microscope are: accelerating voltage 15 kV, beam current 2 nM, and high vacuum mode. And / or, the parameters of the electron probe analyzer are: accelerating voltage 15 kV, beam current 20 nA, and analytical crystals LIF, TAP, and PET.
7. The method according to claim 1, characterized in that, The method for spatial matching between the AMICS image and the EPMA image includes: S21. Use image analysis software to open the AMICS image and the EPMA image respectively, adjust them to the same pixel resolution and canvas size, and make the size and orientation of the two images consistent; S22. Set the AMICS image as the base image, overlay the EPMA image as a layer, and make fine adjustments based on the feature points in the image to achieve image overlap; S23. Adjust the opacity of the AMICS image to ensure that the detail alignment of the AMICS image and the EPMA image can be observed simultaneously.
8. The method according to claim 6, characterized in that, The image analysis software includes ImageJ.
9. The application of the method described in any one of claims 1 to 8 in the automatic identification and classification of gold particles in gold-bearing ores.
10. The application of the method described in any one of claims 1 to 8 in digital mining technology services, characterized in that, This includes importing at least one of the mineral identification map, spatial element map, and location point annotation map obtained by the method into the digital mine platform to improve the richness and realism of digital mine modeling.
Citation Information
Cited By
Intelligent screening and conveying integrated system for rare earth ore dressing process
CN121649033A
Mineral identification method, and training method and device of deep learning semantic segmentation model
CN121708407A
