A geological remote sensing interpretation method and system for mineral resource exploration

Abstract

The application relates to the technical field of image data processing, in particular to a geological remote sensing interpretation method and system for mineral resource exploration, which comprises the following steps: acquiring a geological remote sensing image to be analyzed in a mineral resource area; analyzing the uniformity of the distribution of the pixel gray value of all local areas in the image to be analyzed to obtain the image smoothness; acquiring the LBP value of all pixel points in the image to be analyzed and a historical standard sample image and performing clustering; using a seed growing algorithm to obtain each growth area of each image; determining an image detail evaluation index of the image to be analyzed through the overall distribution of the edge complexity of the growth area of each cluster to which the image to be analyzed and the historical standard sample image belong, obtaining an image enhancement effect quality evaluation, adjusting the wavelet threshold in combination with the image detail evaluation index, and interpreting the enhanced image. The application aims to improve the accuracy of geological remote sensing image interpretation in mineral resource exploration.

Description

Technical Field

[0001] This application relates to the field of image data processing technology, specifically to a geological remote sensing interpretation method and system for mineral resource exploration. Background Technology

[0002] Traditional mineral exploration mainly relies on methods such as surface geological surveys, geophysical exploration, and geochemical sampling. Although these methods are highly accurate, they generally suffer from problems such as long cycles, high costs, high labor intensity, and significant environmental disturbance.

[0003] When using remote sensing technology to acquire images of survey targets, external factors such as atmospheric water vapor and aerosol scattering can cause random noise from the sensor and high-frequency surface background interference during the remote sensing imaging process. This can blur image details and make it difficult to effectively distinguish similar targets. Therefore, it is necessary to enhance remote sensing images to improve image quality. However, wavelet transform image enhancement algorithms typically use a fixed threshold. For constantly changing external interference factors, using such a fixed threshold may not be effective in handling changes, resulting in image enhancement effects that do not meet expectations, thus increasing the difficulty of mineral resource exploration and interpretation based on these images. Summary of the Invention

[0004] In view of the above, it is necessary to provide a geological remote sensing interpretation method and system for mineral resource exploration to solve the above problems.

[0005] The first aspect of this application provides a geological remote sensing interpretation method for mineral resource exploration, the method comprising:

[0006] Obtain geological remote sensing images of the mineral resource area to be explored and analyzed;

[0007] The acquired image to be analyzed is divided into regions according to a preset size, and preliminary enhancement is performed based on the dispersion of gray values ​​in the local regions after division and the number of image pixels; the uniformity of the distribution of gray values ​​of pixels in all local regions of the image to be analyzed to obtain the image smoothness.

[0008] The LBP values ​​of all pixels in the image to be analyzed and historical standard sample images are obtained and clustered. For each pixel belonging to each cluster in each image, a seed growth algorithm is used to obtain each growth region of each image. The number and shape distribution characteristics of edges in each growth region are analyzed to determine the edge complexity of each growth region. By the overall distribution of edge complexity of the growth regions belonging to each cluster in the image to be analyzed and historical standard sample images, the historical complex texture features and current complex texture features of each cluster are obtained, and the image detail evaluation index of the image to be analyzed is determined.

[0009] Based on the image detail evaluation index and the image smoothness, the image enhancement effect quality evaluation is obtained. The wavelet threshold is adjusted in combination with the image detail evaluation index until the image enhancement effect quality evaluation is greater than the preset threshold, and the enhanced image is interpreted.

[0010] The determination of the edge complexity of each growth region is specifically as follows:

[0011] For each edge line in each growth region, the average angle difference between the tangent angles of the j-th pixel on edge line i and all pixels in its neighborhood that are on the same edge line is denoted as . The maximum angular difference value is denoted as ;

[0012] The specific method for calculating edge complexity is as follows:

[0013]

[0014] in, Let S represent the edge complexity of each growth region, and let S represent the number of edge lines in each growth region. This represents the number of pixels on the i-th edge line. This represents the absolute value of the average difference between the length of edge line i and the lengths of other edge lines in the growth region. This represents the maximum length difference between edge line i and other edge lines in the growth region. Represents the normalization function. This indicates the preset value.

[0015] Preferably, the preliminary enhancement process is as follows:

[0016] Calculate the standard deviation of the gray values ​​of each pixel in each local region, and use the average of the standard deviations obtained from all local regions as the noise level of the image;

[0017] The formula for setting the wavelet threshold is: Where G is the preset wavelet threshold, The noise level of the image; The number of pixels in the image to be analyzed; Represents the logarithmic function with the natural constant as the base;

[0018] A wavelet transform image enhancement algorithm with a preset wavelet threshold is used to perform preliminary enhancement on the image to be analyzed.

[0019] Preferably, the process of obtaining image smoothness is as follows:

[0020] The grayscale values ​​of all pixels in each local region are divided into grayscale levels; the local region is evenly divided into a preset number of sub-regions with the geometric center of each local region as the center and in a centrally symmetrical distribution; the difference in the number of pixels of each grayscale level in different sub-regions in each local region is analyzed to determine the local region uniformity of each local region.

[0021] The average uniformity of all local regions in the image to be analyzed is used as the image smoothness.

[0022] Preferably, determining the local region uniformity of each local region specifically involves:

[0023] For each local region, its sub-regions are combined in pairs, and the difference in the number of pixels at each gray level in the combination is calculated. The difference in the number of pixels at all gray levels is accumulated and normalized. The difference between the natural number 1 and the normalized result is taken as the local region uniformity of each local region.

[0024] Preferably, the historical complex texture features are obtained through the following process:

[0025] Calculate the average edge complexity of all growth regions belonging to each cluster in each historical standard sample image; obtain the average absolute value of the difference between the average edge complexity of each historical standard sample image and the average edge complexity of each cluster in all other historical standard sample images, and perform normalization processing.

[0026] The difference between the natural number 1 and the obtained normalization result is used as the historical complex texture feature of each cluster.

[0027] Preferably, the specific process for obtaining the current complex texture features is as follows:

[0028] The absolute value of the difference between the average edge complexity of all growth regions belonging to each cluster in the image to be analyzed and the average edge complexity of all growth regions belonging to each cluster in the historical standard sample image is obtained and normalized. The difference between the natural number 1 and the obtained normalization result is taken as the current complex texture feature of each cluster.

[0029] Preferably, the specific process for determining the image detail evaluation index of the image to be analyzed is as follows:

[0030] The difference between the current complex texture features and the historical complex texture features of each cluster is obtained, and the average of the differences obtained from all clusters is calculated to obtain the image detail evaluation index.

[0031] Preferably, the specific process of obtaining the image enhancement effect quality evaluation and adjusting the wavelet threshold in conjunction with the image detail evaluation index is as follows:

[0032] The specific formula for evaluating the quality of image enhancement is as follows:

[0033]

[0034] in, This indicates the quality evaluation of the image enhancement effect. This represents an image detail evaluation metric. Indicates image smoothness. Indicates the noise level of the image. Represents the normalization function. Indicates the preset value;

[0035] When the image enhancement effect quality evaluation is greater than the preset threshold, the wavelet threshold is not adjusted; otherwise, when the image detail evaluation index is less than the preset historical detail reference value, the adjustment is made using the formula: Decrease the wavelet threshold, where, Preset adjustment step size; This represents the original wavelet threshold; This represents the adjusted wavelet threshold;

[0036] If the image detail evaluation index is greater than or equal to the preset historical detail reference value, the formula is used: Increase the wavelet threshold.

[0037] Secondly, embodiments of this application also provide a geological remote sensing interpretation system for mineral resource exploration, including a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor executes the computer program to implement the steps of any of the methods described above.

[0038] This application has at least the following beneficial effects:

[0039] This application first acquires geological remote sensing images of the mineral resources to be explored, ensuring data coverage of the target mining area and reflecting geological features, texture distribution, and spatial information. The images are then divided into segments of a preset size, and preliminary enhancement is performed based on local gray-level dispersion and the number of pixels. Local segmentation allows for analysis of geological features in small areas, avoiding the loss of local information caused by large-scale enhancement. Preliminary enhancement improves local contrast, making potential mineral or geological boundaries more visible. The uniformity of pixel gray-level distribution in local areas is analyzed to obtain image smoothness. Measuring the smoothness and texture preservation after image enhancement helps determine whether the enhancement is excessive or introduces noise, ensuring that geological texture is not lost during enhancement and making the interpretation results closer to the actual geological structure. The LBP values ​​of the image to be analyzed and historical standard sample images are acquired and clustered. A seed growth algorithm is used to obtain growth regions, extracting and segmenting regions with similar textures, which helps... To improve the accuracy of interpretation in the discovery of ore bodies or important geological units; to analyze the quantity and shape distribution of growth area edges and determine edge complexity, and to quantify edge features to help identify potential ore body boundaries and complex geological structures; to compare the distribution of edge complexity in cluster growth areas with historical samples and the image to be analyzed to obtain image detail evaluation indicators, which help determine whether the image has reached an interpretable quality level and improve the reliability of ore body identification and resource assessment; to evaluate the image enhancement effect based on image detail evaluation indicators and image smoothness, and to adjust the wavelet threshold until the quality reaches the preset threshold, which helps to obtain high-quality enhanced images, making geological interpretation more accurate and reliable, and improving exploration efficiency and accuracy; finally, to interpret the enhanced image and use the optimized image for mineral resource exploration, ensuring that geological textures, faults, veins and other features are clearly visible, improving the interpreter's ability to identify geological bodies, assisting decision-making and improving the success rate of exploration. Attached Figure Description

[0040] Figure 1 A flowchart illustrating the steps of a geological remote sensing interpretation method for mineral resource exploration, provided as an embodiment of this application;

[0041] Figure 2 This is a schematic diagram illustrating the division of sub-regions according to one embodiment of this application;

[0042] Figure 3 This is a schematic diagram illustrating the process of enhancing geological remote sensing images according to one embodiment of this application. Detailed Implementation

[0043] In the description of the embodiments in this application, the words "exemplary," "or," and "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design scheme described as "exemplary" or "for example" in the embodiments of this application should not be construed as being more preferred or advantageous than other embodiments or design schemes. Specifically, the use of the words "exemplary," "or," and "for example" is intended to present the relevant concepts in a specific manner.

[0044] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in this application's specification is for the purpose of describing particular embodiments only and is not intended to be limiting of the application.

[0045] It should also be noted that the terms "first" and "second" in this application and its accompanying drawings are used to distinguish similar objects, rather than to describe a specific order or sequence. The methods disclosed in the embodiments of this application or the methods shown in the flowcharts include one or more steps for implementing the method. Without departing from the scope of protection of this application, the execution order of multiple steps can be interchanged, and some steps can also be deleted.

[0046] The following, in conjunction with the accompanying drawings, details the specific scheme of the geological remote sensing interpretation method and system for mineral resource exploration provided in this application.

[0047] Please see Figure 1 The diagram illustrates a flowchart of a geological remote sensing interpretation method for mineral resource exploration according to an embodiment of this application. The method includes the following steps:

[0048] The first step: Obtain geological remote sensing images of the mineral resource area to be explored for analysis.

[0049] Geological remote sensing images of the mineral resources to be explored were obtained using SPOT 6 satellite imagery, ASTER imagery, and Landsat 8 imagery data, and the obtained images were then processed into grayscale.

[0050] The second step is to divide the acquired image to be analyzed into segments according to a preset size, and to perform preliminary enhancement based on the dispersion of gray values ​​in the local regions after segmentation and the number of image pixels; and to analyze the uniformity of the distribution of gray values ​​of pixels in all local regions of the image to be analyzed to obtain the image smoothness.

[0051] This application first uses a wavelet transform image enhancement algorithm with a preset threshold to enhance the obtained image, resulting in an initial enhanced image. The image is then divided into sections of size [size missing]. In this embodiment, n=21, the standard deviation of the grayscale value of each pixel in the local region is calculated, and the average of the standard deviations obtained from all local regions is used as the image noise level; the formula for the preset wavelet threshold in this application is: Where G is the preset wavelet threshold, The noise level of the image; The number of pixels in the image to be analyzed; This represents the logarithmic function with the natural constant as the base.

[0052] If the wavelet threshold is set too high, the image may be over-smoothed, resulting in the loss of high-frequency details (such as lithological boundaries, linear structures, and small alteration outcrops), reducing the spatial resolution of the image and thus affecting mineral exploration, especially since structural and alteration information is usually found in high-frequency subbands. If the threshold is set too low, the image may not be completely denoised, and the residual noise may interfere with subsequent classification or index calculations (for example, the iron staining index may be biased under the influence of salt-and-pepper noise), thus producing false anomalies and leading to misjudgment of mineralized target areas.

[0053] Based on the above analysis, it is known that the threshold change affects the image quality. Therefore, this application analyzes the image smoothness, image details and other information of the initial enhanced image to determine the enhancement effect of the image under the current threshold, thereby obtaining an image quality evaluation, and then adjusting the wavelet threshold according to the evaluation results.

[0054] The smoother an image is, the less information it contains, making it more difficult to obtain effective information for mineral resource exploration. This application describes the smoothness of the obtained image by combining the grayscale value information of pixels in the image with the analysis results of grayscale value levels. The specific method is as follows:

[0055] First, considering that the physical resources acquired during image capture may include various objects such as mountains, buildings, trees, and rivers, directly describing the smoothness of the entire image might yield unreliable results. Therefore, this application obtains the smoothness of local regions by analyzing the uniformity of pixel grayscale value distribution within each local region of the image. Specifically, the grayscale values ​​of pixels within each local region are clustered using the DBSCAN clustering algorithm, with clustering parameters: neighborhood radius r=3 and minimum neighbor number minpts=3, resulting in K grayscale level clusters. These clusters are then evenly divided into a predetermined number of sub-regions, centered on the geometric center of each local region, in a centrally symmetrical distribution. A schematic diagram of the sub-region division in this embodiment is shown below. Figure 2As shown, the local uniformity of a local area is obtained by measuring the difference in the number of pixels of each category in different sub-regions. In this embodiment, the preset number is 4, and the implementer can adjust it according to the actual situation. In DBSCAN, clustering is performed based on the image gray value. If the image gray value distribution is uniform, the difference in the number of pixels of each category in each local region is small. Conversely, if the image gray value distribution is uneven, the difference in the number of pixels is large.

[0056] Specifically, the number difference of pixels in each cluster is obtained in any pairwise combination of sub-regions within each local region. The number differences of all clusters are accumulated and normalized. The difference between the natural number 1 and the normalized result is used as the local region uniformity of each local region. In this embodiment, the number difference is calculated using the absolute value of the difference, and the normalization method is the maximum-minimum normalization method.

[0057] It should be understood that the smaller the difference in the number of pixels of each category in each sub-region of a local area, the higher the uniformity of the grayscale distribution of pixels in that local area.

[0058] Furthermore, the average of the local region uniformity of all local regions in the image is used as the image smoothness. The higher the resulting image smoothness, the less information the image contains, and the more necessary it is to lower the threshold.

[0059] The third step is to obtain the LBP values ​​of all pixels in the image to be analyzed and the historical standard sample images, and to perform clustering. For each pixel in each image belonging to each cluster, a seed growth algorithm is used to obtain each growth region of each image. The number and shape distribution characteristics of edges in each growth region are analyzed to determine the edge complexity of each growth region. By analyzing the overall distribution of edge complexity of the growth regions belonging to each cluster in the image to be analyzed and the historical standard sample images, the historical complex texture features and current complex texture features of each cluster are obtained, and the image detail evaluation index of the image to be analyzed is determined.

[0060] To effectively support geological feature identification in mineral resource exploration, remote sensing image enhancement processes need to preserve spatial details related to geological processes as much as possible while removing noise and improving contrast. Given that the texture features of geological bodies are primarily determined by their microstructure and edge features, image enhancement methods should ensure consistency between texture and detail.

[0061] Therefore, the texture of the image to be analyzed can be compared and analyzed using historical standard samples (in this embodiment, high-resolution images of verified mining areas). When processing new images, if the texture type of a certain area after enhancement is consistent with historical samples, but the edge richness deviates significantly from expectations, it indicates that the enhancement process may have been over-smoothed or introduced distortion, and the algorithm parameters need to be adjusted. Therefore, this application performs texture analysis based on the acquired geological remote sensing images, performs clustering based on texture features, and analyzes the clusters to obtain the correlation between the texture features of each cluster and the complexity of its neighboring edges, thereby obtaining the enhancement effect on image details in the image to be analyzed.

[0062] The LBP values ​​of all pixels in the image to be analyzed and historical standard sample images are obtained and clustered. The texture features corresponding to the pixels in each cluster of data in each image are analyzed. For pixels belonging to each cluster in each image, a seed growth algorithm is used. Using the pixels in the cluster as seeds, their 8-neighbor pixels are checked. If the difference between the LBP value of the neighboring pixels and the LBP value of the seed point is less than the preset growth threshold (5 in this embodiment, which can be adjusted by the implementer according to the actual situation), they are included in the growth region. This process is repeated until no new pixels are added, thus obtaining the growth regions of each image. The Canny operator is used to extract the edge lines of each growth region, and the edge complexity is analyzed. Specifically, in each growth region, the tangent angle of the j-th pixel on the edge line i is obtained. At the same time, the j-th pixel on edge line i is obtained. The tangent angle corresponding to the c-th edge pixel on the same edge line within the neighborhood. Let their angular differences be denoted as Using this method, we can obtain the j-th pixel on edge line i and its... The angular difference among all edge pixels on the same edge line within the neighborhood is denoted as the average angular difference value of the j-th pixel on edge line i. Let the maximum angular difference value of the j-th pixel on edge line i be denoted as The specific method for calculating edge complexity is as follows:

[0063]

[0064] in, Let S represent the edge complexity of each growth region, and let S represent the number of edge lines in each growth region. This represents the number of pixels on the i-th edge line. This represents the absolute value of the average difference between the length of edge line i and the lengths of other edge lines in the growth region. This represents the maximum length difference between edge line i and other edge lines in the growth region. To prevent A value of 0 results in a calculation result of 0. This represents the average angle difference value corresponding to the j-th pixel on edge line i. This represents the maximum angular difference value corresponding to the j-th pixel on edge line i. This represents the normalization function. In this embodiment, the maximum-minimum value normalization method is used. This represents a preset value. The tangent angle of a pixel on the edge line refers to the normal angle of the gradient direction of that pixel calculated by the edge detection operator. That is, the greater the difference in edge line length in the growth region, and the greater the change in the angle between the pixel on the edge line and the tangent of the edge line, the greater the complexity of the edge information in the growth region.

[0065] It should be noted that when calculating edge complexity, if the number of edge lines S in the current growth region is determined to be 1, then it is forcibly set to... and It equals 0.

[0066] Furthermore, by comparing the overall distribution of edge complexity of the growth regions belonging to each cluster in each historical standard sample image with other historical standard sample images, the historical complex texture features of each cluster are obtained. The specific calculation method is as follows.

[0067]

[0068] in, This represents the historical complex texture features of each cluster; T represents the numerical value of the historical standard sample image. This represents the absolute value of the difference between the average edge complexity of all growth regions belonging to each cluster in the u-th historical sample image and the average edge complexity of all growth regions belonging to each cluster in other historical sample images. In other words, the smaller the overall difference in edge complexity of regions composed of the same texture descriptor pixels corresponding to different historical sample images, the more complex the texture corresponding to the pixels in each cluster of the image.

[0069] The absolute value of the difference between the average edge complexity of all growth regions belonging to each cluster in the image to be analyzed and the average edge complexity of all growth regions belonging to each cluster in the historical standard sample image is obtained and normalized. The difference between the natural number 1 and the obtained normalization result is taken as the current complex texture feature of each cluster.

[0070] By comprehensively comparing the current and historical complex texture features of each cluster, a detail evaluation index for the image to be analyzed is obtained. The specific calculation method is as follows:

[0071]

[0072] in, This represents an image detail evaluation metric. Indicates the number of clusters. This represents the difference between the current complex texture features and historical complex texture features of the v-th cluster. If the difference is positive, it indicates that the current image has high complexity and can reflect the detailed information of the corresponding geological structure. Conversely, if the difference is negative, it indicates that the complexity is low and cannot reflect the detailed information of the corresponding geological structure. In this case, the threshold needs to be lowered to avoid the loss of high-frequency information.

[0073] The fourth step: Based on the image detail evaluation index and the image smoothness, obtain the image enhancement effect quality evaluation, adjust the wavelet threshold in combination with the image detail evaluation index until the image enhancement effect quality evaluation is greater than the preset threshold, and interpret the enhanced image.

[0074] Based on the image detail evaluation index and the previously obtained image smoothness, the quality evaluation of the image enhancement effect is obtained.

[0075]

[0076] in, This indicates the quality evaluation of the image enhancement effect. This represents an image detail evaluation metric. Indicates image smoothness. Indicates the noise level of the image. Represents the normalization function. This represents a preset value used to prevent the denominator from being 0; in this embodiment, the value is 0.01. The higher the detail evaluation index in the desired image, and the lower the image noise level and image smoothness, the better the image quality.

[0077] Acquire and calculate several historical images that have been historically assessed as high quality by those skilled in the art. The values ​​are taken as the average value as the preset threshold P. If so, the image is directly output for interpretation; if Then, feedback adjustment is initiated to adjust the wavelet threshold. Re-enhance and calculate ,until Alternatively, the maximum number of iterations can be reached before interpretation. In this embodiment, the maximum number of iterations is 10. It should be noted that the historical sample images must be acquired using the same type of sensor as the images to be analyzed, and the corresponding geographical areas must have similar geological structures.

[0078] The specific process for adjusting the wavelet threshold in this embodiment is as follows:

[0079] The average detail evaluation index of historically rated high-quality images is selected as the historical detail reference value. If the image detail evaluation index is lower than the historical detail reference value, it indicates that the wavelet threshold is too large, and texture details have been mistakenly deleted. This is determined by the formula: Decrease the wavelet threshold, where, The preset adjustment step size is set to a value of [value to be filled in]. ; This represents the original wavelet threshold; This represents the adjusted wavelet threshold.

[0080] If the image detail evaluation index is greater than or equal to the historical detail reference value, it indicates that the detail preservation is acceptable, but the noise level is low. or smoothness Excessive noise (i.e., noise not properly cleared), addressed by the formula: Increase the wavelet threshold.

[0081] The obtained remote sensing images are enhanced using the methods described above, and then interpreted according to existing interpretation techniques to complete the geological remote sensing interpretation for mineral resource exploration. A flowchart illustrating the enhancement process for geological remote sensing images is shown below. Figure 3 As shown.

[0082] First, the acquired impact data is preprocessed using the methods described above, specifically including atmospheric correction, geometric correction, image enhancement, and image cropping as described in this application.

[0083] The processed image is then interpreted using methods such as direct interpretation.

[0084] The direct interpretation method involves establishing remote sensing geological interpretation markers based on the shape, color, and other image features of geological bodies of different properties on remote sensing images, directly extracting geological information such as rocks and structures, and directly determining the attributes of geological bodies and land features.

[0085] Finally, a comprehensive verification was conducted by combining geophysical and geochemical data analysis, and the results were analyzed.

[0086] Based on the same inventive concept as the above method, this application embodiment also provides a geological remote sensing interpretation system for mineral resource exploration, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of any one of the above-described geological remote sensing interpretation methods for mineral resource exploration.

[0087] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than that shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. In the descriptions corresponding to the flowcharts and block diagrams in the accompanying drawings, the operations or steps corresponding to different blocks may also occur in a different order than disclosed in the description; sometimes there is no specific order between different operations or steps. For example, two consecutive operations or steps may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. Each block in a block diagram and / or flowchart, and combinations of blocks in a block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0088] It will be apparent to those skilled in the art that this application is not limited to the details of the exemplary embodiments described above, and that this application can be implemented in other specific forms without departing from its essential characteristics. Therefore, the embodiments described above should be considered exemplary and non-limiting in all respects; modifications to the technical solutions described in the foregoing embodiments, or equivalent substitutions of some technical features, without causing the essence of the corresponding technical solutions to deviate from the scope of the technical solutions in the embodiments of this application, should all be included within the protection scope of this application.

Claims

1. A geological remote sensing interpretation method for mineral resource exploration, characterized in that, The method includes the following steps: Obtain geological remote sensing images of the mineral resource area to be explored and analyzed; The acquired image to be analyzed is divided into regions according to a preset size, and preliminary enhancement is performed based on the dispersion of gray values ​​in the local regions after division and the number of image pixels; the uniformity of the distribution of gray values ​​of pixels in all local regions of the image to be analyzed to obtain the image smoothness. The LBP values ​​of all pixels in the image to be analyzed and historical standard sample images are obtained and clustered. For each pixel belonging to each cluster in each image, a seed growth algorithm is used to obtain each growth region of each image. The number and shape distribution characteristics of edges in each growth region are analyzed to determine the edge complexity of each growth region. By the overall distribution of edge complexity of the growth regions belonging to each cluster in the image to be analyzed and historical standard sample images, the historical complex texture features and current complex texture features of each cluster are obtained, and the image detail evaluation index of the image to be analyzed is determined. Based on the image detail evaluation index and the image smoothness, the image enhancement effect quality evaluation is obtained. The wavelet threshold is adjusted in combination with the image detail evaluation index until the image enhancement effect quality evaluation is greater than the preset threshold, and the enhanced image is interpreted. The determination of the edge complexity of each growth region is specifically as follows: For each edge line in each growth region, the average angle difference between the tangent angles of the j-th pixel on edge line i and all pixels in its neighborhood that are on the same edge line is denoted as . The maximum angular difference value is denoted as ; The specific method for calculating edge complexity is as follows: in, Let S represent the edge complexity of each growth region, and let S represent the number of edge lines in each growth region. This represents the number of pixels on the i-th edge line. This represents the absolute value of the average difference between the length of edge line i and the lengths of other edge lines in the growth region. This represents the maximum length difference between edge line i and other edge lines in the growth region. Represents the normalization function. This indicates the preset value.

2. The geological remote sensing interpretation method for mineral resource exploration as described in claim 1, characterized in that, The initial enhancement process is as follows: Calculate the standard deviation of the gray values ​​of each pixel in each local region, and use the average of the standard deviations obtained from all local regions as the noise level of the image; The formula for setting the wavelet threshold is: Where G is the preset wavelet threshold, The noise level of the image; The number of pixels in the image to be analyzed; Represents the logarithmic function with the natural constant as the base; A wavelet transform image enhancement algorithm with a preset wavelet threshold is used to perform preliminary enhancement on the image to be analyzed.

3. The geological remote sensing interpretation method for mineral resource exploration as described in claim 1, characterized in that, The process of obtaining image smoothness is as follows: The grayscale values ​​of all pixels in each local region are divided into grayscale levels; the local region is evenly divided into a preset number of sub-regions with the geometric center of each local region as the center and in a centrally symmetrical distribution; the difference in the number of pixels of each grayscale level in different sub-regions in each local region is analyzed to determine the local region uniformity of each local region. The average uniformity of all local regions in the image to be analyzed is used as the image smoothness.

4. The geological remote sensing interpretation method for mineral resource exploration as described in claim 3, characterized in that, The determination of the local region uniformity of each local region specifically involves: For each local region, its sub-regions are combined in pairs, and the difference in the number of pixels at each gray level in the combination is calculated. The difference in the number of pixels at all gray levels is accumulated and normalized. The difference between the natural number 1 and the normalized result is taken as the local region uniformity of each local region.

5. The geological remote sensing interpretation method for mineral resource exploration as described in claim 1, characterized in that, The specific process for obtaining the historical complex texture features is as follows: Calculate the average edge complexity of all growth regions belonging to each cluster in each historical standard sample image; obtain the average absolute value of the difference between the average edge complexity of each historical standard sample image and the average edge complexity of each cluster in all other historical standard sample images, and perform normalization processing. The difference between the natural number 1 and the obtained normalization result is used as the historical complex texture feature of each cluster.

6. The geological remote sensing interpretation method for mineral resource exploration as described in claim 1, characterized in that, The specific process for obtaining the current complex texture features is as follows: The absolute value of the difference between the average edge complexity of all growth regions belonging to each cluster in the image to be analyzed and the average edge complexity of all growth regions belonging to each cluster in the historical standard sample image is obtained and normalized. The difference between the natural number 1 and the obtained normalization result is taken as the current complex texture feature of each cluster.

7. The geological remote sensing interpretation method for mineral resource exploration as described in claim 1, characterized in that, The specific process for determining the image detail evaluation index of the image to be analyzed is as follows: The difference between the current complex texture features and the historical complex texture features of each cluster is obtained, and the average of the differences obtained from all clusters is calculated to obtain the image detail evaluation index.

8. The geological remote sensing interpretation method for mineral resource exploration as described in claim 2, characterized in that, The specific process of obtaining the image enhancement effect quality evaluation and adjusting the wavelet threshold in conjunction with the image detail evaluation index is as follows: The specific formula for evaluating the quality of image enhancement effects is as follows: in, This indicates the quality evaluation of the image enhancement effect. This represents an image detail evaluation metric. Indicates image smoothness. Indicates the noise level of the image. Represents the normalization function. Indicates the preset value; When the image enhancement effect quality evaluation is greater than the preset threshold, the wavelet threshold is not adjusted; otherwise, when the image detail evaluation index is less than the preset detail history reference value, the threshold is adjusted using the formula: Decrease the wavelet threshold, where, Preset adjustment step size; This represents the original wavelet threshold; This represents the adjusted wavelet threshold; If the image detail evaluation index is greater than or equal to the preset historical detail reference value, the formula is used: Increase the wavelet threshold.

9. A geological remote sensing interpretation system for mineral resource exploration, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1-8.

Images