Mineral geological exploration system based on remote sensing image texture analysis

Abstract

The present application relates to remote sensing geological information processing technical field, disclose a kind of mineral geological exploration system based on remote sensing image texture analysis, comprising: time series baseline library construction unit, for generating the time series statistical baseline of normal fluctuation range definition, ecological proxy texture channel processing module, it is purified according to time series statistical baseline to the normalized vegetation index texture of current image, to generate ecological proxy texture anomaly map, construct texture channel processing module, for generating construction texture anomaly map, and collaborative verification engine module, before executing spatial coupling determination, first according to image information entropy evaluation two channel information quality and dynamically select verification logic, the information of surface covering in traditional exploration, which is regarded as interference, is converted into an independent verification dimension, and through the double constraints of space and time and the adaptive evaluation of information quality, an internal cross-validation evidence chain is constructed.

Description

Technical Field

[0001] This invention relates to a mineral geological exploration system based on remote sensing image texture analysis, belonging to the field of remote sensing geological information processing technology. Background Technology

[0002] In the field of mineral geological exploration services using high technology, remote sensing technology has become a commonly used technical means in the early stages of general survey and target area screening due to its macroscopic, rapid and economical characteristics. Its main application is to identify geological structures, lithological anomalies or alteration information related to mineralization by interpreting spectral or texture information in images.

[0003] However, in most areas of the Earth's landmass that are covered by vegetation, soil, and other cover, the application of the above-mentioned technologies faces a fundamental dilemma. In order to obtain information about the covered bedrock, the current mainstream technical methods generally follow a direct reduction logic, that is, to try to strip away the influence of the surface cover through complex image algorithms or by relying on high-cost sensors with certain penetration capabilities in order to directly observe the geological information underneath. In exploration practice, especially in the general survey stage where economic requirements are stringent, this approach has become a technical trade-off. Either higher exploration costs are invested in pursuit of direct information, which violates the economic principle of early exploration, or when using universally available remote sensing data, the unavoidable introduction of artifacts and uncertainties in the stripping process leads to a decrease in the reliability of the interpretation results.

[0004] To overcome this predicament, the industry has continuously strived to develop more advanced sensors or more sophisticated algorithms. However, these efforts are essentially improvements on the existing direct reduction approach, failing to break free from its inherent limitations. This technological inertia has led the industry to overlook a crucial possibility: the surface cover itself, which acts as an obstacle to exploration, actually exhibits a surface ecological response under the long-term, comprehensive influence of the deep geological environment, containing rich and interpretable indirect information. This response mechanism is mainly established through two pathways: 1. Deep mineralized bodies, through geological processes such as weathering and leaching, alter the geochemical environment of the overlying soil, creating elemental anomalies that are stressful or selective for plant growth. This long-term geochemical stress can lead to observable systematic differences in species composition, growth density, and health status of surface vegetation; 2. Geological structures such as faults and fracture zones associated with mineralization directly affect the hydrological conditions and physical properties of local soils, thereby regulating vegetation type and growth, forming linear or ring-shaped ecological anomalies along the structure. Therefore, the technical problem to be solved by this invention is how to create an information processing method that no longer attempts to peel off or penetrate the surface cover, but instead uses the texture features of the cover itself as a proxy information channel that is independent of geological structure information, and achieves reliable identification of hidden geological anomalies by relying solely on standard, universally available remote sensing data through the collaborative verification of information from the two channels. Summary of the Invention

[0005] This invention provides a mineral geological exploration system based on remote sensing image texture analysis. Its main purpose is to solve the dilemma of cost and reliability in the processing of remote sensing images of vegetated areas by adhering to the direct reconstruction logic. In particular, it addresses how to use universally accessible remote sensing data to achieve high-reliability and low-cost identification of hidden geological anomalies through an information processing method that transforms surface cover information from noise into proxy signals.

[0006] To achieve the above objectives, this invention provides a mineral geological exploration system based on remote sensing image texture analysis, the system comprising:

[0007] A time-series baseline library construction unit acquires multiple historical remote sensing images of the survey area and calculates the statistical parameters of the normalized vegetation index texture of the historical remote sensing images to generate a time-series statistical baseline that defines the normal fluctuation range.

[0008] An ecological proxy texture channel processing module calculates the normalized vegetation index map of the current remote sensing image, performs texture analysis on the normalized vegetation index map, and compares the texture analysis results with the time-series statistical baseline to generate an ecological proxy texture anomaly map that only contains areas where texture values ​​exceed the normal fluctuation range.

[0009] A texture channel processing module is used to acquire shortwave infrared images that are insensitive to vegetation and perform texture analysis on the images to generate texture anomaly maps.

[0010] A collaborative verification engine module receives an ecological proxy texture anomaly map and a constructed texture anomaly map, calculates the image information entropy value of the two maps, and compares the information entropy value with a benchmark interval obtained from multi-sample statistics. When the information entropy values ​​of the two maps are both within the benchmark interval, strong verification logic is executed to identify the spatially coupled anomaly region in the two maps as the target area for mineral exploration.

[0011] Preferably, when the collaborative verification engine module executes strong verification logic, it uses a coupling index C determined by both spatial overlap and spatial proximity. index To determine whether anomaly regions are spatially coupled, the coupling index C is used. index The calculation rules are as follows: Among them, A overlap To construct the spatial overlap area between the anomalous connected clusters in the texture anomaly map and the anomalous connected clusters in the ecological proxy texture anomaly map, A T d is the sum of the areas of two connected clusters. min Let d be the minimum edge distance between two connected clusters. max To find the maximum search distance, w1 and w2 are non-negative weight coefficients that satisfy w1 + w2 = 1. The collaborative verification engine module will use the coupling index C. index Regions exceeding the judgment threshold are identified as spatially coupled abnormal regions.

[0012] Preferably, both the ecological proxy texture channel processing module and the constructed texture channel processing module use the gray-level co-occurrence matrix algorithm to perform texture analysis. Specifically, the constructed texture channel processing module calculates at least one texture parameter among the contrast or heterogeneity of the shortwave infrared band image, and the ecological proxy texture channel processing module calculates at least one texture parameter among the variance or entropy of the normalized vegetation index map.

[0013] Preferably, when either of the two graphs' information entropy values ​​is not within the baseline range, the collaborative verification engine module executes weak verification logic. It takes the abnormal region in the abnormal graph generated by the channel whose information entropy value is within the baseline range as the main target, and searches for edge responses in the other channel only within the bandwidth range of the main target to confirm the main target.

[0014] Preferably, when the exploration area is determined to be a homogeneous lithological cover area, the structural texture channel processing module selects remote sensing images with low solar altitude angles acquired at sunrise or sunset in that area, and performs edge detection on the remote sensing images to extract the shadow distribution formed by the slight topographic undulations to generate a structural texture anomaly map.

[0015] Preferably, the system also includes a terrain context modulation module, which calculates the terrain humidity index map of the survey area based on the digital elevation model; and the ecological proxy texture channel processing module spatially modulates the normal fluctuation range according to the terrain humidity index map when generating the time-series statistical baseline, expanding the boundary value of the normal fluctuation range in areas with high terrain humidity index values ​​and narrowing the boundary value of the normal fluctuation range in areas with low terrain humidity index values.

[0016] Preferably, the system also includes a structural permeability detection module. This module automatically acquires normalized difference water index images of the area at two different times—when the soil is near saturation and when it is dry—after a rainfall event in the exploration area. By subtracting these two normalized difference water index images pixel by pixel, an image representing the spatial difference in soil moisture decay rate is generated. The collaborative verification engine module uses the area with abnormal moisture decay rate in this image as a third information layer and performs collaborative verification with the structural texture anomaly map and the ecological proxy texture anomaly map to determine the target area for mineral exploration.

[0017] Preferably, the system further includes a residual information interpretation module, which performs a spatial logical XOR operation on the constructed texture anomaly map and the ecological proxy texture anomaly map to extract uncoupled anomaly regions that appear only in a single channel, and marks the regions from the constructed texture channel processing module as potential deep structural regions and the regions from the ecological proxy texture channel processing module as shallow environmental anomaly regions according to the source channel of the uncoupled anomaly regions, so as to generate auxiliary decision-making information.

[0018] Preferably, the temporal baseline library construction unit acquires all historical images of the same month as the current image within the past ten years of the exploration area. The ecological proxy texture channel processing module uses the Ojin method or a statistical method based on the mean and standard deviation of historical samples to automatically determine the threshold used to segment the texture analysis results, so as to initially identify abnormal areas and compare the initially identified abnormal areas with the temporal statistical baseline.

[0019] Compared with the prior art, the beneficial effects of the present invention are:

[0020] 1. This invention provides an information processing method for mineral geological exploration. It constructs a structural texture channel reflecting the physical topography of the substrate and an ecological proxy texture channel representing the state of the surface ecosystem in parallel. Then, a collaborative verification engine is used to identify and determine the spatially coupled anomalous areas in the two channels. This processing method transforms surface cover information, which is considered interference in traditional exploration, into an independent verification dimension that is related to the geoscientific origin of structural information. When the system makes target judgments, it no longer relies on a single-dimensional texture anomaly signal, but on coupled signals that are simultaneously reflected at both the physical structure and ecological response levels. Thus, the identification of real anomalies caused by deep geological reasons no longer depends on improving the signal-to-noise ratio of a single signal, but on confirming an evidence chain with an inherent causal relationship. This provides a high-reliability information screening path for early exploration using universal remote sensing data in a wide coverage area.

[0021] 2. This invention further introduces a dynamic anomaly determination mechanism based on time-series statistical baselines. When processing the ecological proxy texture channel, this mechanism does not use a fixed global threshold, but automatically acquires historical images of the same period over many years in the exploration area. It establishes the normal texture fluctuation range of each plot under a specific season. The texture state of the current image is only identified as abnormal when it deviates from its own historical normal trajectory. This process enables the system to distinguish between periodic and seasonal surface changes and long-term continuous geological influences. Thus, before entering the dual-channel collaborative verification, pseudo-anomaly signals caused by climate or routine human activities that are not directly related to deep mineralization have been filtered out, ensuring the ecological proxy texture information sent to the final verification stage.

[0022] 3. In the processing of the structural texture channel, this invention also provides a working mode for information enhancement using micro-topographic shadows. That is, in areas with homogeneous lithology and weak differences in spectral characteristics, the system will prioritize the use of low solar altitude angle images acquired at sunrise or sunset. By analyzing the distribution of shadows that are elongated due to slight topographic undulations, the potential macro-geological structural patterns can be identified and delineated. This makes the extraction of structural information no longer completely limited to the intrinsic spectral properties of surface materials, but transforms the external natural condition of sunlight into a means of amplifying micro-topographic morphology. This allows the entire exploration system of this invention to still obtain a structural texture input in areas where traditional optical remote sensing methods are scarce, such as areas covered by large areas of single lithology or aeolian sediments, ensuring the operation of the dual-channel collaborative verification mechanism. Attached Figure Description

[0023] Figure 1 This is a flowchart of the dual-channel information processing and collaborative verification technology of the present invention.

[0024] Figure 2 This is a performance calibration curve of the spatial coupling index weighting parameter of the present invention;

[0025] Figure 3 This is a flowchart illustrating the dynamic decision-making logic of the collaborative verification engine based on information entropy in this invention.

[0026] Figure 4 This is a sequence diagram showing the module collaboration of the collaborative verification engine of this invention in executing dual verification logic. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this invention, not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0028] This invention provides a mineral geological exploration system based on remote sensing image texture analysis, comprising a temporal baseline database construction unit, an ecological proxy texture channel processing module, a structural texture channel processing module, and a collaborative verification engine module. These components work collaboratively, processing remote sensing data in parallel to generate two independent texture anomaly information layers. Based on a preset collaborative verification logic, they identify spatially coupled anomalies in the two information layers to determine the exploration target area. In a specific application scenario, such as conducting mineral prospecting in areas with extensive vegetation and laterite coverage, the system's temporal baseline database construction unit is activated first. This unit is configured to access a publicly available remote sensing data archive and automatically obtain data on the user-specified exploration area over the past ten years, comparing it with the current data. All historical images of the same month for the image to be processed show that the surface condition of a specific area in a specific season follows a statistical periodic pattern. Therefore, by acquiring long-term contemporaneous images, a quantitative reference benchmark can be established for the normal texture fluctuation range of the area. For each acquired historical image, the ecological proxy texture channel processing module calculates its Normalized Difference Vegetation Index (NDVI) image and uses the Gray-Level Co-occurrence Matrix (GLCM) algorithm to calculate texture parameters such as variance or entropy of the NDVI image to characterize the uniformity or complexity of vegetation cover. The temporal baseline library construction unit statistically analyzes the texture parameters of all historical contemporaneous images, calculates and stores the mean and standard deviation of the texture parameters for each image unit, thereby generating a temporal statistical baseline that defines the normal fluctuation range.

[0029] After the time-series statistical baseline is established, the ecological proxy texture channel processing module processes the current remote sensing image. It calculates the NDVI map and texture parameter map of the current image, and compares the calculated texture parameter values ​​with the normal fluctuation range of the corresponding geographical location and month in the time-series baseline library. Only when the texture state of a region deviates significantly from its own historical normal trajectory, such as when its texture value exceeds two standard deviations from the historical average for the same period, is it identified as an ecological proxy texture anomaly and recorded in an ecological proxy texture anomaly map. This processing method aims to filter out texture changes caused by regular seasonal variations or periodic human activities, thereby improving the purity of geological information in subsequent analyses. In parallel, the construction texture channel processing module works to extract data reflecting the basement structure. Regarding the information on geological patterns, this module is configured to avoid interference from vegetation and soil moisture. To this end, it automatically selects shortwave infrared (SWIR) band images that are not sensitive to vegetation, such as Band 6 images from Landsat 8 or 9, or Band 11 images from Sentinel-2. In these bands, the image brightness values ​​mainly respond to the mineral composition of rocks and soil, reducing the interference of surface cover on the interpretation of underlying geological information. This module also uses the Gray-Level Co-occurrence Matrix (GLCM) algorithm to calculate the contrast or heterogeneity texture parameters of the SWIR images. These parameters are sensitive to linear or annular brightness abrupt changes caused by fault structures, lithological contact zones, etc. The resulting structural texture anomaly map identifies the potential regional-scale geological structural patterns within the area.

[0030] When the exploration area is identified as a homogeneous lithological cover zone, such as a large desert, the intrinsic texture contrast in the SWIR band is weak. In this case, the structural texture channel processing module is configured in a special working mode. In this mode, the system prioritizes remote sensing images acquired at sunrise or sunset in the area with low solar altitude angles. Utilizing the phenomenon that weak topographic relief casts elongated shadows under low-angle illumination, the system extracts shadows with specific morphological distributions by performing edge detection algorithms on the image. The extracted shadow patterns are then used as structural texture anomaly maps. This approach uses sunlight as a means of analyzing micro-topographical morphology. Enhanced detection methods ensure that the collaborative verification mechanism can obtain structural information input under different surface conditions. Ecological proxy texture anomaly maps and structural texture anomaly maps are fed into the collaborative verification engine module. Before performing spatial coupling determination, this module first calculates the image information entropy of the two anomaly maps to quantify the complexity of the image information. The calculated information entropy value is compared with a benchmark interval obtained from multi-sample statistics. When the information entropy values ​​of both maps are within this benchmark interval, the engine executes strong verification logic, using a coupling index C determined by both spatial overlap and spatial proximity. index To determine whether anomaly regions are spatially coupled, the calculation rule for this index is as follows: Among them, Aoverlap Let A be the spatial overlap area of ​​two abnormally connected clusters. T d is the sum of the areas of two connected clusters. min Let d be the minimum edge distance between two connected clusters. max The maximum search distance is defined by the user based on the geological characteristics of the exploration area. w1 and w2 are non-negative weight coefficients that satisfy w1 + w2 = 1. index When the value is greater than the judgment threshold, the abnormal area is identified as a high-confidence mineral exploration target area. If the information entropy value of any input map is not within the benchmark range, it indicates that the quality of the information source has decreased. The engine then switches to weak verification logic, takes the abnormal map generated by the channel with normal information entropy value as the main target, and searches for edge response in another channel only within a certain bandwidth range of the main target, as an auxiliary confirmation of the main target.

[0031] To correct for the modulation effect of topography on surface ecological response signals, the system can also integrate a topographic context modulation module. This module calculates the topographic humidity index (TWI) map of the exploration area based on the digital elevation model (DEM). During the generation of the time-series statistical baseline in the ecological proxy texture channel processing module, it spatially modulates the boundary values ​​of the normal fluctuation range based on this TWI map. In areas with high topographic humidity index values, the boundary values ​​of the normal fluctuation range are expanded; conversely, in areas with low index values, they are reduced. This modulation allows the system to adjust the judgment criteria according to local topographic conditions to adapt to the differentiated expression of signals in different topographic units. To evaluate the water-conducting and mineral-conducting potential of structures, the system can be equipped with a structural permeability detection module. This module automatically acquires normalized difference water index (NDWI) images of the area at two different times: when the soil is near saturation and after experiencing dryness, using a regional rainfall event. By analyzing the two NDWI images... The WI image is subtracted pixel by pixel to generate an image representing the spatial differences in soil moisture decay rate. On this difference map, areas with abnormal moisture decay rate indicate zones with higher underground permeability. This map is used as the third information layer and fed into the collaborative verification engine module for three-way collaborative verification with the structural texture anomaly map and the ecological proxy texture anomaly map to identify functionally active exploration target areas. In addition, the system also includes a residual information interpretation module, which performs a spatial logical XOR operation on the structural texture anomaly map and the ecological proxy texture anomaly map to extract uncoupled anomaly areas that appear only in a single channel. Based on their source channel, the modules are classified, with areas originating from the structural texture channel being marked as potential deep structural areas and areas originating from the ecological proxy texture channel being marked as shallow environmental anomaly areas. This module finally generates an auxiliary decision information layer, which is presented to the user along with the core mineral exploration target area.

[0032] Example 1: In a survey project targeting gold-bearing quartz veins covering thousands of square kilometers in the tropical hilly region of Southeast Asia, the objective obstacle faced by the exploration work was the extensive surface coverage by vegetation and laterite layers, resulting in weak geochemical anomalies in the overlying soil caused by deep structures. This limited the application of interpretation methods relying on the spectral characteristics of surface materials. If only single-period remote sensing images of the area were used for texture analysis, a large number of texture anomaly patches would be generated due to seasonal changes in surface conditions such as vegetation growth and agricultural activities, making subsequent field verification work infeasible due to the large target base. In this application scenario, the system of this invention was deployed to perform target area screening. The system's structural texture channel processing module first acquired and processed the Landsat 8 Band 6 shortwave infrared image covering the exploration area, calculated its contrast texture parameters, and generated a structural texture anomaly map, which identified several northeast-trending... The linear structural zones and several ring structures are structural units with mineralization potential within the regional geological context. Simultaneously, the system's ecological proxy texture channel processing module processes the Normalized Difference Vegetation Index (NDVI) map acquired concurrently with shortwave infrared imagery. The temporal baseline library construction unit automatically processes historical images of the same months over the past ten years in the exploration area, establishing a temporal statistical baseline for the NVI texture of the region. When the ecological proxy texture channel processing module compares the texture analysis results of the current image with this temporal statistical baseline, a large number of texture variation areas caused by seasonal factors are filtered out. The final generated ecological proxy texture anomaly map retains only a few spatially clustered areas that deviate from their historical normal fluctuation range in the temporal dimension. This temporally purified ecological proxy texture anomaly map provides a filtered input for subsequent collaborative verification.

[0033] This processing approach does not attempt to strip away surface cover to explore underlying geological information. Instead, it treats the spatial texture features of the surface cover itself as proxy signals. By introducing a time dimension constraint, it distinguishes between periodic surface changes and persistent anomalies that may be caused by long-term influences from the deep geological environment. Therefore, the core of the exploration problem shifts from how to eliminate surface interference to how to verify whether these persistent surface ecological anomalies are spatially coupled with deep geological structures. Finally, the collaborative verification engine module receives the structural texture anomaly map and the ecological proxy texture anomaly map. After confirming that the information entropy values ​​of both maps are within the baseline range, it executes strong verification logic. The analysis results show that among several northeast-trending linear structural zones, only two specific segments of structural zones exhibit spatial overlap or proximity with the anomalous clusters in the ecological proxy texture anomaly map, with a coupling index C. indexIf the value exceeds the preset judgment threshold, other structural texture anomalies and ecological proxy texture anomalies will appear as isolated entities with no spatial correlation. The mineral exploration target areas output by the system are only spatially coupled areas that are simultaneously reflected in both physical structure and ecological response dimensions. The number of exploration target areas has been reduced from thousands to single digits. Subsequent field geological surveys can focus on these high-potential target areas and have discovered alteration phenomena related to gold mineralization. This approach provides a high-reliability information screening path for early exploration in wide-area coverage areas, improving the efficiency of subsequent exploration work, under the condition of relying solely on universal remote sensing data.

[0034] Example 2: To quantitatively evaluate the effectiveness of the system of the present invention in screening exploration targets in complex covered areas, the following comparative experiment was conducted. The experiment selected an area of ​​100 km² within the survey area. 2 The block contains a surface map of eight known mineralization points, established based on detailed geological surveys and engineering verification data, which serves as a benchmark for evaluating the experimental results. The experiment consists of one experimental group and three control groups, all using the same Landsat 8 imagery covering the experimental area and its historical archives as data sources. The experimental group employs the complete technical solution of this invention; control group A uses only the output of the structural texture channel processing module; control group B uses only the output of the ecological proxy texture channel processing module purified by the time-series baseline library; and control group C uses an alteration mineral extraction method based on spectral angle mapping. For the experimental group, the coupling index C in the collaborative verification engine module... index The judgment threshold was set at 0.7. This value was chosen to balance recall and accuracy in the early stages of exploration. This value was determined through parameter testing in other known areas, and it is an engineering setting that can better balance the two. After parallel processing of each group of experiments, the control group C failed to identify any effective targets due to spectral interference from vegetation and laterite. The control group A output 25 structural texture anomalies, which, after comparison with the ground condition map, were confirmed to include 5 real targets and 20 false targets, with an accuracy rate of 20.0%. The control group B output 12 ecological proxy texture anomalies after time-series purification, which included 4 real targets and 8 false targets, with an accuracy rate of 33.3%. The experimental group, through collaborative verification of the structural texture anomaly map and the ecological proxy texture anomaly map, finally output 7 mineral exploration target areas, which included 6 real targets and 1 false target, with an accuracy rate of 85.7%.

[0035] Experimental data show that although control groups A and B could identify some real targets, they relied on information from only one dimension and could not exclude a large number of anomalies caused by non-mineralized geological structures or non-tectonic environmental factors, resulting in a large number of false targets and low accuracy. The dual-channel collaborative verification method adopted by the experimental group spatially couples and constrains structural information with time-series purified ecological anomaly information. Only anomalies that respond in both independent information dimensions are confirmed as the final target. This cross-verification mechanism eliminates the ambiguity of single-source information, intercepting most real targets while keeping the number of false targets at a low level. The results of this experiment show that compared with exploration methods that rely on a single information source, the collaborative verification method adopted in this invention can improve the accuracy of identifying concealed mineralized targets in vegetated areas, providing a technical path for dual-channel collaborative verification in remote sensing geological exploration.

[0036] Example 3: This example combines Figures 1 to 4 A description of a mineral geological exploration system based on remote sensing image texture analysis, such as... Figure 1 As shown, the data input includes multiple historical remote sensing images from the same period used to establish historical reference benchmarks, a digital elevation model (DEM) used to correct for topographic effects, current remote sensing images used to generate ecological proxy information, and shortwave infrared images used to extract tectonic information. The temporal baseline library construction unit processes historical images to calculate and output temporal statistical baselines, while the topographic context modulation module spatially modulates the baseline based on the topographic humidity index generated by the DEM. The modulated baseline is then sent to the ecological proxy texture channel processing module, which simultaneously processes the normalized difference vegetation index (NDVI) of the current image to purify and identify deviations from historical data. In the region of the trajectory, an ecological proxy texture anomaly map is generated. In parallel, a texture channel processing module analyzes shortwave infrared (SWIR) band images that are insensitive to vegetation to extract basement structural pattern information and generate a structural texture anomaly map. These two independent anomaly maps are simultaneously sent to a collaborative verification engine module. After evaluating the information quality of the two channels, this module performs strong or weak verification logic to determine spatial coupling and finally delineates high-confidence mineral exploration target areas. At the same time, a residual information interpretation module can extract uncoupled anomaly areas from the two anomaly maps to generate auxiliary decision-making information that includes potential deep structural areas and shallow environmental anomaly areas.

[0037] like Figure 2As shown in the figure, the horizontal axis represents the value of the spatial overlap weight w1, and the vertical axis represents the index value. The figure shows the trend of three curves with the value of w1, namely the average coupling index of positive samples, the average coupling index of negative samples, and the comprehensive evaluation index F1 score. It can be seen from the figure that when the value of w1 is around 0.65, the average coupling index of positive samples and the F1 score both reach their peak, while the average coupling index of negative samples is at a relatively low level. This indicates that the weight configuration can most effectively identify real targets and suppress false targets.

[0038] like Figure 3 As shown, after receiving dual-channel anomaly maps in standby / waiting-for-input mode, the engine first enters the information quality evaluation stage, calculating the information entropy values ​​of the two anomaly maps. Then, it enters a core decision node. If it is determined that the information entropy is within the baseline range, the process switches to executing strong verification logic, calculating the coupling index by judging spatial overlap and proximity. If it is determined that either information entropy is not within the baseline range, it indicates that the quality of at least one information source has deteriorated, and the process switches to executing weak verification logic. At this time, the high-quality channel is the main target, and only the edge response of the other channel is searched. Regardless of the logic executed, the results are eventually incorporated into the target generation stage to delineate the target area or confirm the main target. After the judgment is completed, the engine outputs the results and resets, returning to the standby state.

[0039] like Figure 4 As shown, the process begins with the collaborative verification engine acquiring data from the ecological anomaly map and the constructed anomaly map respectively. Then, it calculates the information entropy of the two maps and returns the entropy value. In an alt selection combo box, two different paths are displayed based on the entropy value judgment result: If the entropy values ​​of both maps are within the baseline range, strong verification logic is triggered. The engine requests the analyzer to calculate the spatial overlap and spatial proximity, and the analyzer returns the generated coupling index. If either entropy value is abnormal, weak verification logic is triggered. After selecting the main target, the engine performs edge response search and returns the verification result. Finally, the engine determines the exploration target area based on the result returned by either path and calls the output module to generate a result report.

[0040] Example 4: When applying the system of the present invention to a province with geological and ecological environmental characteristics different from those of previous areas, in order to adapt the decision logic of the system's core algorithm module to the specific conditions of the region, a systematic offline parameter calibration procedure needs to be performed. This procedure aims to establish a statistical deviation threshold for determining ecological anomalies, a benchmark interval for image information entropy to assess the quality of input information, and a coupling index C for defining spatial coupling strength. index The calibration procedure first selects several areas with a total area of ​​approximately 100 km² within the province to be explored, based on existing regional geological data, using key parameters such as weighting coefficients. 2The selection of training areas must meet two conditions: first, the geological and surface cover types must be representative of the general characteristics of the entire province to be explored; second, the area must contain known positive samples that have been verified by previous exploration work and whose locations are clearly defined, such as small mineral deposits or fault fracture zones, as well as known negative samples, such as large areas or areas with simple structures and no mineralization records.

[0041] The first step of the procedure is to establish an image information entropy benchmark interval applicable to the province. The system automatically acquires and processes all available images covering all training areas from the past 15 years, and runs the construction texture channel processing module and the ecological proxy texture channel processing module to generate a sample library containing hundreds of construction texture anomaly maps and ecological proxy texture anomaly maps. The system calculates the image information entropy of each image in the sample library and performs statistical analysis on all entropy values. The range defined by the 10th percentile and 90th percentile of this statistical distribution is determined as the information entropy benchmark interval applicable to the surveyed province. The second step of the procedure is to calibrate the judgment threshold for ecological proxy texture anomalies. The system uses... Using known positive and negative samples in the training area as references, the statistical deviation threshold in the ecological proxy texture channel processing module is iteratively tested. The test process iterates through the standard deviation multiples from 1.0 to 3.0 in steps of 0.1. At each step, the system calculates its recognition results for positive and negative samples in the training area and evaluates its overall classification performance. Ultimately, the threshold that maximizes the sum of the true positive and true negative rates is determined as the optimal judgment threshold. In this specific province, the calibration result is 2.1 times the standard deviation, and this value is set as the business parameter for the system's ecological proxy texture anomaly judgment in this region. The third step of the procedure is to adjust the coupling index C. index The weight coefficients w1 and w2 are optimized to balance the relative importance of spatial overlap and spatial proximity in determining coupling relationships. Using the anomaly map calibrated in the previous steps, the system iterates through the values ​​of w1 from 0 to 1 in the training region with a step size of 0.05, and sets w2 = 1 - w1. For each weight combination, the C0 value of all known positive sample regions is calculated. index The average value can make the positive sample region C index The set of weight coefficients that maximizes the average value, w1 = 0.65 and w2 = 0.35, is determined as the optimal configuration and applied to subsequent processing.

[0042] By executing the above procedures, all key algorithm parameters of the system have been locally calibrated before large-scale exploration operations are carried out. This ensures that every aspect of its information processing and decision-making logic has a basis for setting based on objective data verification in the local area. The system then enters standby mode, ready to screen mineral exploration target areas in the region.

[0043] Example 5: When the system of the present invention is applied to areas with a single surface material composition, such as Gobi desert areas covered by large areas of aeolian sand, the application of shortwave infrared band texture analysis that relies on lithological differences will be limited. In such cases, before executing the main process, the system first performs a global variance calculation on the standard shortwave infrared band image covering the exploration area. If the calculated variance value is lower than a low variance threshold obtained through statistical calibration of multiple geomorphic samples, the system determines the area as a homogeneous lithological coverage area and adjusts the working mode of the structural texture channel processing module. In this mode, the system no longer performs the gray-level co-occurrence matrix algorithm on the original image, but automatically accesses the satellite data archive, filters and calls remote sensing images acquired at sunrise or sunset in the area with a solar altitude angle of less than 15 degrees, and performs an edge detection algorithm on the image. By extracting the shadow distribution formed by the slight topographic undulations, a structural texture anomaly map is generated.

[0044] To evaluate the tectonic permeability of the target area, a tectonic permeability detection module is activated to analyze the region's hydrological dynamic response after rainfall events. This module monitors the survey area by accessing meteorological data services. When a rainfall event occurs with more than 20 mm of rainfall within 24 hours and covering more than 80% of the area, the module is triggered. The system then retrieves and acquires the first cloudless remote sensing image sensitive to soil moisture after the rainfall ends as the saturation phase image. Simultaneously, the system accesses a geoscience database to obtain the dominant soil type of the target area and calculates an optimal observation time window for soil moisture decay rate differences based on the built-in soil moisture evapotranspiration model and the average meteorological data of the current season. For example, for sandy loam soils in Central Asia, this window is determined to be 72 to 96 hours after the rainfall stops in summer. The system then acquires a second cloudless remote sensing image as the decay phase image within this dynamically calculated time window. By processing these two images, an information layer characterizing tectonic permeability is generated for subsequent collaborative verification.

[0045] Example 6: Before deploying the system of the present invention in a new exploration project area, a standardized localization adaptation procedure must be executed to ensure that its internal key parameters match the geoscientific characteristics and data environment of the area. The first step of this procedure is to set the geometric parameters in the collaborative verification engine module. The procedure requires inputting the average influence radius of typical geochemical halos or alteration zones that are effective in the area and are related to the target mineral type. Based on this, the system sets the maximum search distance d. max The bandwidth range in the weak verification logic is set to 1.5 times this radius value. Its function is to correlate the search scale of the spatial coupling analysis with the objective geological influence range of mineralization in the region.

[0046] The second step of the procedure is to adaptively configure the algorithm for initially identifying abnormal regions in the ecological proxy texture channel processing module. When processing the normalized vegetation index texture map, the system first analyzes the grayscale histogram of the image. If the histogram shows a bimodal shape, it indicates that there are two distinguishable populations in the image: background and potential anomalies. In this case, the system uses the Ozin method for threshold segmentation. If the histogram shows a complex shape with one or more peaks, it automatically switches to processing based on the statistical methods of historical sample mean and standard deviation. This data-driven algorithm selection mechanism aims to call the model that best matches the statistical distribution characteristics of the input data. The processing path; the third step of the procedure is to perform self-checking and fault-tolerant processing of the historical data environment. The time-series baseline library construction unit counts the number of available historical images within the target month and year range. If the number of images is less than 10, the system determines that the area is a data sparse area. Under this condition, the system automatically relaxes the data retrieval conditions to adjacent months and performs a weighted average of images from different months to construct a time-series statistical baseline. The weight of the images in the target month is set to the highest. After the procedure is completed, all core parameters of the system are configured locally and enter the optimized operation state for this specific project area.

[0047] Example 7: When surveying a subalpine area with complex terrain that had recently experienced localized wildfires, some functional modules of the system faced the challenge of asymmetric degradation in data source quality. Specifically, the large-scale fire-affected areas resulted in the surface vegetation being covered by homogenized carbon, causing the ecological proxy texture channel processing module to output anomaly maps containing large areas of low information when processing images of this area, with image information entropy values ​​lower than the preset lower limit of the baseline range.

[0048] Under this condition, the collaborative verification engine module automatically switches to weak verification logic and uses the constructed texture anomaly map with normal information entropy as the main basis for the recognition process. Simultaneously, the terrain context modulation module spatially modulates the anomaly judgment logic of the ecological proxy texture channel based on the digital elevation model of the area with significant elevation differences. This modulation enables the system to identify several weak but persistent ecological proxy texture anomalies caused by groundwater seeping along structural fissures with higher sensitivity in steep, shady slopes that have not been burned but have a high terrain humidity index. When executing the weak verification logic, the system will construct the texture... Several northwest-trending structural zones marked in the anomaly map were identified as the main targets. Within the specified bandwidth of these structural zones, the weak ecological responses identified after topographic modulation were successfully searched as confirmatory evidence, thus delineating the core exploration target area. After the exploration task was completed, the residual information interpretation module of the system processed the uncoupled anomalies in the two channels. A linear structural texture anomaly completely covered by the burned area could not be included in the collaborative verification because it lacked any ecological response signal. Therefore, it was marked as a potential deep structural area, providing independent auxiliary decision-making information for subsequent deep physical exploration work.

[0049] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0050] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A mineral geological exploration system based on remote sensing image texture analysis, characterized in that, The system includes: A time-series baseline library construction unit acquires multiple historical remote sensing images of the survey area and calculates the statistical parameters of the normalized vegetation index texture of the historical remote sensing images to generate a time-series statistical baseline that defines the normal fluctuation range. An ecological proxy texture channel processing module calculates the normalized vegetation index map of the current remote sensing image, performs texture analysis on the normalized vegetation index map, and compares the texture analysis results with the time-series statistical baseline to generate an ecological proxy texture anomaly map that only contains areas where texture values ​​exceed the normal fluctuation range. A texture channel processing module is used to acquire shortwave infrared images that are insensitive to vegetation and perform texture analysis on the images to generate texture anomaly maps. A collaborative verification engine module receives an ecological proxy texture anomaly map and a constructed texture anomaly map, calculates the image information entropy value of the two maps, and compares the information entropy value with a benchmark interval obtained from multi-sample statistics. When the information entropy values ​​of the two maps are both within the benchmark interval, strong verification logic is executed to identify the spatially coupled anomaly region in the two maps as the target area for mineral exploration.

2. The mineral geological exploration system based on remote sensing image texture analysis according to claim 1, characterized in that, When the collaborative verification engine module executes strong verification logic, it uses a coupling index determined by both spatial overlap and spatial proximity. To determine whether anomaly regions are spatially coupled, the coupling index is used. The calculation rules are as follows: ,in, To construct the spatial overlap area between anomalous connected clusters in the texture anomaly map and anomalous connected clusters in the ecological proxy texture anomaly map. The sum of the areas of two connected clusters. Let be the minimum edge distance between two connected clusters, when the two connected clusters have spatial overlap. =0; For a maximum search distance, It was set to be 1.5 times the average radius of influence of a typical geochemical halo or alteration zone within the exploration area. and To meet The non-negative weighting coefficients, the collaborative verification engine module will couple the index Regions exceeding the judgment threshold are identified as spatially coupled abnormal regions.

3. The mineral geological exploration system based on remote sensing image texture analysis according to claim 1, characterized in that, Both the ecological proxy texture channel processing module and the constructed texture channel processing module use the gray-level co-occurrence matrix algorithm to perform texture analysis. The constructed texture channel processing module calculates at least one texture parameter, either contrast or heterogeneity, of the shortwave infrared band image, while the ecological proxy texture channel processing module calculates at least one texture parameter, either variance or entropy, of the normalized vegetation index map.

4. A mineral geological exploration system based on remote sensing image texture analysis according to claim 1, characterized in that, When either of the two graphs' entropy values ​​is not within the baseline range, the collaborative verification engine module executes weak verification logic. It takes the abnormal region in the abnormal graph generated by the channel whose entropy value is within the baseline range as the primary target, and searches for the existence of edge response in the other channel only within the bandwidth range of the primary target.

5. A mineral geological exploration system based on remote sensing image texture analysis according to claim 1, characterized in that, When the exploration area is identified as a homogeneous lithological cover area, the structural texture channel processing module selects remote sensing images with low solar altitude angles acquired at sunrise or sunset in that area, and performs edge detection on the remote sensing images to extract the shadow distribution formed by the slight topographic undulations to generate a structural texture anomaly map.

6. A mineral geological exploration system based on remote sensing image texture analysis according to claim 1, characterized in that, The system also includes a terrain context modulation module, which calculates the terrain humidity index map of the survey area based on the digital elevation model; and the ecological proxy texture channel processing module spatially modulates the normal fluctuation range according to the terrain humidity index map when generating the time-series statistical baseline. In areas with high terrain humidity index values, the limit value of the normal fluctuation range is expanded, and in areas with low terrain humidity index values, the limit value of the normal fluctuation range is narrowed.

7. A mineral geological exploration system based on remote sensing image texture analysis according to claim 1, characterized in that, The system also includes a structural permeability detection module. After a rainfall event in the exploration area, the module automatically acquires normalized difference water index images of the area at two different times: when the soil is near saturation and when it has experienced dryness. By subtracting these two normalized difference water index images pixel by pixel, an image representing the spatial difference in soil moisture decay rate is generated. The collaborative verification engine module uses the area with abnormal moisture decay rate in this image as the third information layer and performs collaborative verification with the structural texture anomaly map and the ecological proxy texture anomaly map to determine the target area for mineral exploration.

8. A mineral geological exploration system based on remote sensing image texture analysis according to claim 1, characterized in that, The system also includes a residual information interpretation module, which performs a spatial logical XOR operation on the constructed texture anomaly map and the ecological proxy texture anomaly map to extract uncoupled anomaly regions that appear only in a single channel. Based on the source channel of the uncoupled anomaly region, the region originating from the constructed texture channel processing module is marked as a potential deep structural region, and the region originating from the ecological proxy texture channel processing module is marked as a shallow environmental anomaly region, thereby generating auxiliary decision-making information.

9. A mineral geological exploration system based on remote sensing image texture analysis according to claim 1, characterized in that, The temporal baseline library construction unit acquires historical images of the same month as the current image within the past ten years in the exploration area. The ecological proxy texture channel processing module uses a statistical method based on the historical sample mean and standard deviation to automatically determine the threshold used to segment the texture analysis results, so as to initially identify abnormal areas and compare the initially identified abnormal areas with the temporal statistical baseline.

Images