A method for ore prospecting based on local sensitive hashing

Abstract

This application provides a mineral exploration method based on locality-sensitive hashing (LSH). The method includes: acquiring and interpreting the geological structure of remote sensing image information; marking areas in the remote sensing structural interpretation map that meet preset reference geological information as alteration anomaly areas; compiling a remote sensing alteration anomaly map; identifying the geological information corresponding to the locations of multiple known mineral deposits in the remote sensing alteration anomaly map; creating mineral exploration prediction feature models corresponding to various mineral deposit types based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit; calculating the locality-sensitive hash set of the mineral exploration prediction feature model for each mineral deposit type; calculating the locality-sensitive hash of multiple target locations in the remote sensing alteration anomaly map; calculating the similarity between the locality-sensitive hash of each target location and the locality-sensitive hash sets of various mineral deposits; and determining areas with a similarity greater than a preset threshold as mineral exploration prediction areas, which can improve mineral exploration efficiency.

Description

Technical Field

[0001] This application relates to the field of mineral exploration technology, and in particular to a mineral exploration method based on locality-sensitive hashing. Background Technology

[0002] In recent years, with the advancement of science and technology, many advanced technologies have yielded fruitful results in the field of mineral exploration. For example, geophysical, geochemical, and remote sensing technologies have been widely applied in mineral exploration. However, the inventors have found in practice that although existing technologies can improve mineral exploration efficiency to some extent, the delineation of mineral prediction zones still relies on human experience after obtaining the exploration results, resulting in low overall delineation efficiency. Summary of the Invention

[0003] The purpose of this application is to provide a mineral exploration method based on locality-sensitive hashing (LSH) to solve the technical problem that the delineation of mineral prediction zones still relies on manual experience, resulting in low overall delineation efficiency. The specific technical solution is as follows: A first aspect of this application provides a mining method based on locality-sensitive hashing, the method comprising: Acquire remote sensing image information of the mineral cluster area to be identified; interpret the geological structure of the remote sensing image information to obtain a remote sensing structural interpretation map of the mineral cluster area to be identified; The areas in the remote sensing structural interpretation map that meet the preset reference geological information are marked as alteration anomaly areas; a remote sensing alteration anomaly map is compiled based on the identified alteration anomaly areas; Based on the location information of multiple known mineral deposits obtained in advance, the geological information corresponding to the locations of the multiple known mineral deposits in the remote sensing alteration anomaly map is identified to obtain the geological information of multiple known mineral deposits; based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit, mineral exploration prediction feature models corresponding to various mineral deposit types are created. Calculate the local sensitive hash set of the prospecting prediction feature model for each type of mineral deposit; calculate the local sensitive hash of multiple target locations in the remote sensing alteration anomaly map; calculate the similarity between the local sensitive hash of each target location and the local sensitive hash set of various mineral deposits; and determine the regions with a similarity greater than a preset threshold as prospecting prediction areas.

[0004] In one possible implementation, the step of acquiring remote sensing image information of the mineral cluster area to be identified and interpreting the geological structure of the remote sensing image information to obtain a remote sensing structural interpretation map of the mineral cluster area to be identified includes: The remote sensing image information of the mineral cluster area to be identified is obtained from multiple sources, including: remote sensing image information collected by Pleiades and image information of multiple spectral bands collected by the Landsat 8 land imager. Based on the pre-defined correspondence between different element features and geological structures, the remote sensing image information from multiple sources is interpreted to obtain the element features corresponding to each geological structure in the remote sensing image information from multiple sources. The element features include one or more of the following: linear structure, ring structure, zone feature, color feature, and block feature. In the pre-defined correspondence between different element features and geological structures, the linear structure is used to represent faults, the ring structure is used to represent arcuate faults or folds, the zone feature is used to represent alteration zones, the color feature is used to represent hydrothermal alteration, and the block feature is used to represent intrusive rock masses. According to the actual location of each geological structure in the remote sensing image information from each source, the feature characteristics corresponding to each geological structure are arranged to obtain the feature arrangement map corresponding to the remote sensing image information from each source; the feature arrangement maps corresponding to the remote sensing image information from each source are merged according to their actual locations to obtain the output structural interpretation map of the mineral cluster area to be identified. Based on the content of each element in different regions identified through the remote sensing image information, the content of each element in different regions is marked according to the actual location in the structural interpretation map to be output, thereby obtaining the remote sensing structural interpretation map; wherein, the content of each element in different regions identified through the remote sensing image information is calculated by using the spectral matching method to analyze the image information of multiple spectral bands acquired by the land imager.

[0005] In one possible implementation, the preset reference geological information includes: preset abnormal content ranges corresponding to different elements, and preset abnormal geological structure sets; The process of marking areas in the remote sensing structural interpretation map that meet preset reference geological information as alteration anomaly areas; and compiling a remote sensing alteration anomaly map based on the identified alteration anomaly areas, including: Based on the preset abnormal content ranges corresponding to different elements and the content of each element in different regions of the remote sensing construction interpretation map, identify the abnormal regions of any element whose content satisfies the corresponding abnormal content range. Based on the preset set of anomalous geological structures, geological anomalous regions in the remote sensing structural interpretation map that satisfy the preset set of anomalous geological structures are identified, wherein the preset set of anomalous geological structures includes: hydrothermal alteration, alteration zone, arcuate fault, and fold. Identify the merged region of the elemental anomaly region and the geological anomaly region in the remote sensing structural interpretation map, and take the merged region as the alteration anomaly region; The alteration anomaly region is marked in the remote sensing structural interpretation map, and the marking result is used as the remote sensing alteration anomaly map.

[0006] In one possible implementation, based on the pre-acquired location information of multiple known mineral deposits, the geological information corresponding to the locations of the multiple known mineral deposits in the remote sensing alteration anomaly map is identified to obtain the geological information of the multiple known mineral deposits; based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit, prospecting prediction feature models corresponding to various mineral deposit types are created, including: Based on the location information of multiple known mineral deposits, the actual content of each element and the actual geological structure at the location of each known mineral deposit in the remote sensing alteration anomaly map are identified; for any known mineral deposit, the identified actual content of each element and the actual geological structure are marked as the mineral deposit characteristics of that known mineral deposit. Based on the mineral deposit characteristics and mineral deposit types of each known mineral deposit, mineral exploration prediction feature models corresponding to different mineral deposit types are created. The mineral exploration prediction feature models corresponding to different mineral deposit types include mineral deposit characteristics of one or more known mineral deposits belonging to that mineral deposit type.

[0007] In one possible implementation, calculating the local sensitive hash set of the prospecting prediction feature model for each type of mineral deposit includes: Using a word embedding algorithm, the mapping vector of each mineral deposit feature in the mineral exploration prediction feature model for each type of mineral deposit is calculated, and the calculated mapping vector is used as the feature vector of that type of mineral deposit; using a hash algorithm, the hash value of the feature vector of each type of mineral deposit is calculated. A random projection matrix is ​​created using the random projection method; the hash values ​​of the feature vectors of various deposit types are input into the random projection matrix, and the output projection result is used as the local sensitive hash set of the prospecting prediction feature model for that deposit type.

[0008] In one possible implementation, calculating the hash value of the feature vectors for various mineral deposit types using a hash algorithm includes: The hash value of the mapping vector corresponding to multiple known mineral deposits in the feature vector of various mineral deposit types is calculated using a hash algorithm. The process involves creating a random projection matrix using a random projection method; inputting the hash values ​​of the feature vectors for various mineral deposit types into the random projection matrix; and using the output projection results as the local sensitive hash set of the mineral exploration prediction feature model for that type of mineral deposit, including: Randomly generate multiple unit vectors of preset dimensions; merge the multiple unit vectors of preset dimensions to obtain the random projection matrix; The hash values ​​of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types are input into the random projection matrix to obtain the local sensitive hashes of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types. The local sensitivity hashes of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types are merged separately to obtain the local sensitivity hash set of the mineral exploration prediction feature model for each mineral deposit type.

[0009] In one possible implementation, the step of calculating the local sensitive hashes of multiple target locations in the remote sensing alteration anomaly map; calculating the similarity between the local sensitive hash of each target location and the local sensitive hash sets of various mineral deposits; and determining areas with a similarity greater than a preset threshold as mineral exploration prediction areas includes: Extract the elemental content and geological structure of multiple target locations from the remote sensing alteration anomaly map to obtain the target element content and target geological structure; calculate the target vector of each element target content and target geological structure using a word embedding algorithm; calculate the hash value of the target vector using a hash algorithm; input the hash value of the target vector into the random projection matrix to obtain the local sensitive hash of multiple target locations; wherein, the target locations are manually selected locations or randomly selected locations; Calculate the local sensitivity hash of each target location and the cosine similarity of the local sensitivity hash sets of various mineral deposits; If the cosine similarity of any target location is greater than a preset threshold, then the target location is marked as the mineral exploration prediction area.

[0010] A second aspect of this application provides a mining apparatus based on locality-sensitive hashing, the apparatus comprising: The interpretation map creation module is used to acquire remote sensing image information of the mineral cluster area to be identified; and to interpret the geological structure of the remote sensing image information to obtain the remote sensing structural interpretation map of the mineral cluster area to be identified. Anomaly map creation module is used to mark areas in the remote sensing structural interpretation map that meet preset reference geological information as alteration anomaly areas; and to compile remote sensing alteration anomaly maps based on the identified alteration anomaly areas. The model creation module is used to identify the geological information corresponding to the locations of the multiple known mineral deposits in the remote sensing alteration anomaly map based on the location information of multiple known mineral deposits obtained in advance, and to obtain the geological information of multiple known mineral deposits; based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit, to create mineral exploration prediction feature models corresponding to various mineral deposit types respectively; The prediction area identification module is used to calculate the local sensitive hash set of the prospecting prediction feature model for each type of mineral deposit; calculate the local sensitive hash of multiple target locations in the remote sensing alteration anomaly map; calculate the similarity between the local sensitive hash of each target location and the local sensitive hash set of various mineral deposits; and identify the areas with a similarity greater than a preset threshold as prospecting prediction areas.

[0011] In one possible implementation, the step of acquiring remote sensing image information of the mineral cluster area to be identified and interpreting the geological structure of the remote sensing image information to obtain a remote sensing structural interpretation map of the mineral cluster area to be identified includes: The remote sensing image information of the mineral cluster area to be identified is obtained from multiple sources, including: remote sensing image information collected by Pleiades and image information of multiple spectral bands collected by the Landsat 8 land imager. Based on the pre-defined correspondence between different element features and geological structures, the remote sensing image information from multiple sources is interpreted to obtain the element features corresponding to each geological structure in the remote sensing image information from multiple sources. The element features include one or more of the following: linear structure, ring structure, zone feature, color feature, and block feature. In the pre-defined correspondence between different element features and geological structures, the linear structure is used to represent faults, the ring structure is used to represent arcuate faults or folds, the zone feature is used to represent alteration zones, the color feature is used to represent hydrothermal alteration, and the block feature is used to represent intrusive rock masses. According to the actual location of each geological structure in the remote sensing image information from each source, the feature characteristics corresponding to each geological structure are arranged to obtain the feature arrangement map corresponding to the remote sensing image information from each source; the feature arrangement maps corresponding to the remote sensing image information from each source are merged according to their actual locations to obtain the output structural interpretation map of the mineral cluster area to be identified. Based on the content of each element in different regions identified through the remote sensing image information, the content of each element in different regions is marked according to the actual location in the structural interpretation map to be output, thereby obtaining the remote sensing structural interpretation map; wherein, the content of each element in different regions identified through the remote sensing image information is calculated by using the spectral matching method to analyze the image information of multiple spectral bands acquired by the land imager.

[0012] In one possible implementation, the preset reference geological information includes: preset abnormal content ranges corresponding to different elements, and preset abnormal geological structure sets; The process of marking areas in the remote sensing structural interpretation map that meet preset reference geological information as alteration anomaly areas; and compiling a remote sensing alteration anomaly map based on the identified alteration anomaly areas, including: Based on the preset abnormal content ranges corresponding to different elements and the content of each element in different regions of the remote sensing construction interpretation map, identify the abnormal regions of any element whose content satisfies the corresponding abnormal content range. Based on the preset set of anomalous geological structures, geological anomalous regions in the remote sensing structural interpretation map that satisfy the preset set of anomalous geological structures are identified, wherein the preset set of anomalous geological structures includes: hydrothermal alteration, alteration zone, arcuate fault, and fold. Identify the merged region of the elemental anomaly region and the geological anomaly region in the remote sensing structural interpretation map, and take the merged region as the alteration anomaly region; The alteration anomaly region is marked in the remote sensing structural interpretation map, and the marking result is used as the remote sensing alteration anomaly map.

[0013] In one possible implementation, based on the pre-acquired location information of multiple known mineral deposits, the geological information corresponding to the locations of the multiple known mineral deposits in the remote sensing alteration anomaly map is identified to obtain the geological information of the multiple known mineral deposits; based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit, prospecting prediction feature models corresponding to various mineral deposit types are created, including: Based on the location information of multiple known mineral deposits, the actual content of each element and the actual geological structure at the location of each known mineral deposit in the remote sensing alteration anomaly map are identified; for any known mineral deposit, the identified actual content of each element and the actual geological structure are marked as the mineral deposit characteristics of that known mineral deposit. Based on the mineral deposit characteristics and mineral deposit types of each known mineral deposit, mineral exploration prediction feature models corresponding to different mineral deposit types are created. The mineral exploration prediction feature models corresponding to different mineral deposit types include mineral deposit characteristics of one or more known mineral deposits belonging to that mineral deposit type.

[0014] In one possible implementation, calculating the local sensitive hash set of the prospecting prediction feature model for each type of mineral deposit includes: Using a word embedding algorithm, the mapping vector of each mineral deposit feature in the mineral exploration prediction feature model for each type of mineral deposit is calculated, and the calculated mapping vector is used as the feature vector of that type of mineral deposit; using a hash algorithm, the hash value of the feature vector of each type of mineral deposit is calculated. A random projection matrix is ​​created using the random projection method; the hash values ​​of the feature vectors of various deposit types are input into the random projection matrix, and the output projection result is used as the local sensitive hash set of the prospecting prediction feature model for that deposit type.

[0015] In one possible implementation, calculating the hash value of the feature vectors for various mineral deposit types using a hash algorithm includes: The hash value of the mapping vector corresponding to multiple known mineral deposits in the feature vector of various mineral deposit types is calculated using a hash algorithm. The process involves creating a random projection matrix using a random projection method; inputting the hash values ​​of the feature vectors for various mineral deposit types into the random projection matrix; and using the output projection results as the local sensitive hash set of the mineral exploration prediction feature model for that type of mineral deposit, including: Randomly generate multiple unit vectors of preset dimensions; merge the multiple unit vectors of preset dimensions to obtain the random projection matrix; The hash values ​​of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types are input into the random projection matrix to obtain the local sensitive hashes of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types. The local sensitivity hashes of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types are merged separately to obtain the local sensitivity hash set of the mineral exploration prediction feature model for each mineral deposit type.

[0016] In one possible implementation, the step of calculating the local sensitive hashes of multiple target locations in the remote sensing alteration anomaly map; calculating the similarity between the local sensitive hash of each target location and the local sensitive hash sets of various mineral deposits; and determining areas with a similarity greater than a preset threshold as mineral exploration prediction areas includes: Extract the elemental content and geological structure of multiple target locations from the remote sensing alteration anomaly map to obtain the target element content and target geological structure; calculate the target vector of each element target content and target geological structure using a word embedding algorithm; calculate the hash value of the target vector using a hash algorithm; input the hash value of the target vector into the random projection matrix to obtain the local sensitive hash of multiple target locations; wherein, the target locations are manually selected locations or randomly selected locations; Calculate the local sensitivity hash of each target location and the cosine similarity of the local sensitivity hash sets of various mineral deposits; If the cosine similarity of any target location is greater than a preset threshold, then the target location is marked as the mineral exploration prediction area.

[0017] Another aspect of the embodiments of this application also provides an electronic device, including: Memory, used to store computer programs; The processor, when executing a program stored in memory, implements any of the above-mentioned locality-sensitive hashing-based mining methods.

[0018] In another aspect of the embodiments of this application, a computer-readable storage medium is provided, wherein a computer program is stored therein, and when the computer program is executed by a processor, it implements any of the above-described mining methods based on locality-sensitive hashing.

[0019] In another aspect of the embodiments of this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the above-described mining methods based on locality-sensitive hashing.

[0020] Beneficial effects of the embodiments in this application: This application provides a mineral exploration method based on Locality Sensitive Hash (LSH). The method includes: acquiring remote sensing image information of a mineral cluster to be identified; interpreting the geological structure of the remote sensing image information to obtain a remote sensing structural interpretation map of the mineral cluster to be identified; marking areas in the remote sensing structural interpretation map that meet preset reference geological information as alteration anomaly areas; compiling a remote sensing alteration anomaly map based on the identified alteration anomaly areas; identifying the geological information corresponding to the locations of multiple known mineral deposits in the remote sensing alteration anomaly map based on the location information of multiple known mineral deposits obtained in advance, to obtain the geological information of multiple known mineral deposits; creating mineral exploration prediction feature models corresponding to multiple mineral deposit types based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit; calculating the locality sensitive hash set of the mineral exploration prediction feature model for each mineral deposit type; calculating the locality sensitive hash of multiple target locations in the remote sensing alteration anomaly map; calculating the similarity between the locality sensitive hash of each target location and the locality sensitive hash sets of various mineral deposits; and determining areas with a similarity greater than a preset threshold as mineral exploration prediction areas. The scheme proposed in this application allows for the creation of a remote sensing structural interpretation map after acquiring remote sensing image information. Based on the characteristics of known mineral deposits in the remote sensing structural interpretation map, a local sensitive hash set is created. Furthermore, by using the local sensitive hash set and the local sensitive hash of each target location, similarity is calculated to identify the mineral exploration prediction area. This eliminates the need for manual delineation of prediction areas based on experience, avoiding human error and improving mineral exploration efficiency.

[0021] Of course, implementing any product or method of this application does not necessarily require achieving all of the advantages described above at the same time. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other embodiments can be obtained based on these drawings.

[0023] Figure 1A schematic flowchart of a mineral exploration method based on locality-sensitive hashing provided in an embodiment of this application; Figure 2 A schematic diagram of a mineral exploration device based on locality-sensitive hashing provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art based on this application are within the scope of protection of this application.

[0025] The first aspect of this application's embodiments provides a mining method based on locality-sensitive hashing, see [link to relevant documentation]. Figure 1 The method includes: Step S11: Obtain remote sensing image information of the mineral cluster area to be identified; interpret the geological structure of the remote sensing image information to obtain the remote sensing structural interpretation map of the mineral cluster area to be identified; Step S12: Mark the areas in the remote sensing structural interpretation map that meet the preset reference geological information as alteration anomaly areas; compile a remote sensing alteration anomaly map based on the identified alteration anomaly areas; Step S13: Based on the location information of multiple known mineral deposits obtained in advance, identify the geological information corresponding to the locations of the multiple known mineral deposits in the remote sensing alteration anomaly map, and obtain the geological information of multiple known mineral deposits; based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit, create mineral exploration prediction feature models corresponding to multiple mineral deposit types respectively; Step S14: Calculate the local sensitive hash set of the prospecting prediction feature model for each type of mineral deposit; calculate the local sensitive hash of multiple target locations in the remote sensing alteration anomaly map; calculate the similarity between the local sensitive hash of each target location and the local sensitive hash set of various mineral deposits; and determine the regions with a similarity greater than a preset threshold as prospecting prediction areas.

[0026] Corresponding to step S11 above, remote sensing image information of the mineral cluster to be identified is obtained. This can include publicly available satellite imagery, digital elevation models, geographical features such as roads, water systems, and place names, as well as topographic maps. For example, the remote sensing image information includes: remote sensing imagery acquired by Pleiades satellites and imagery of multiple spectral bands acquired by the Landsat 8 land imager. The geological structure of the remote sensing image information is then interpreted to obtain a remote sensing structural interpretation map of the mineral cluster to be identified. Referring to existing technologies, specifically, the geological structure can be interpreted from the remote sensing image information from multiple sources according to a pre-defined correspondence between different feature elements and geological structures. For example, at the corresponding locations of geological structures, the geological structures can be replaced with different feature elements to obtain a remote sensing structural interpretation map of the mineral cluster to be identified.

[0027] Corresponding to step S12 above, the areas in the remote sensing structural interpretation map that meet the preset reference geological information are marked as alteration anomaly areas. This can be achieved by setting the geological type and element range that meet the anomaly conditions, thereby marking the locations in the remote sensing structural interpretation map that meet the anomaly conditions as alteration anomaly areas, and thus determining the remote sensing structural interpretation map marked with alteration anomaly areas as a remote sensing alteration anomaly map.

[0028] Corresponding to step S13 above, based on the pre-acquired location information of multiple known mineral deposits, the geological information corresponding to the locations of the multiple known mineral deposits in the remote sensing alteration anomaly map is identified, thus obtaining the geological information of the multiple known mineral deposits. Specifically, the geological information corresponding to these locations in the remote sensing alteration anomaly map can be identified based on the location information of the known mineral deposits, such as coordinate ranges. Based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit, prospecting prediction feature models corresponding to various mineral deposit types are created. Alternatively, the geological information corresponding to the same mineral deposit type can be treated as a whole, and this whole can be used as the prospecting prediction feature model corresponding to each mineral deposit type.

[0029] Corresponding to step S14 above, the local sensitivity hash set of the prospecting prediction feature model for each type of mineral deposit is calculated. This can be done using existing algorithms for calculating local sensitivity hashes, such as the projection method. Specifically, the calculation process may include vectorizing the prospecting prediction feature model for each type of mineral deposit and then calculating the local sensitivity hash of the vectorized vector. Since the prospecting prediction feature model for each type of mineral deposit includes multiple prospecting prediction feature models corresponding to multiple known mineral deposits, multiple local sensitivity hashes are also obtained, so these multiple hashes are named as a local sensitivity hash set. Calculating the local sensitivity hashes of multiple target locations in the remote sensing alteration anomaly map can be done by randomly selecting multiple locations and using these selected locations as target locations, or by selecting multiple locations based on manual experience and using them as target locations. The similarity between the local sensitive hash of each target location and the local sensitive hash set of various mineral deposits can be calculated using multiple similarity calculation methods. Areas with similarity greater than a preset threshold are identified as mineral exploration prediction areas, which facilitate further exploration of minerals in these prediction areas by human intervention.

[0030] As can be seen, the scheme of this application can create a remote sensing structural interpretation map after acquiring remote sensing image information, and then create a local sensitive hash set based on the characteristics of known mineral deposits in the remote sensing structural interpretation map. Furthermore, by using the local sensitive hash set and the local sensitive hash of each target location, the similarity is calculated to identify the mineral exploration prediction area. This achieves the delineation of the prediction area without relying on experience by humans, which not only avoids human error, but also improves the efficiency of mineral exploration.

[0031] In one possible implementation, acquiring remote sensing image information of the mineral cluster area to be identified; interpreting the geological structure of the remote sensing image information to obtain a remote sensing structural interpretation map of the mineral cluster area to be identified includes: acquiring remote sensing image information of the mineral cluster area from multiple sources, wherein the remote sensing image information from multiple sources includes: remote sensing image information acquired by Pleiades and image information of multiple spectral bands acquired by the Landsat 8 land imager; interpreting the geological structure of the remote sensing image information from multiple sources according to a pre-defined correspondence between different element characteristics and geological structures to obtain a remote sensing structural interpretation map of the mineral cluster area to be identified. The document describes the feature characteristics of geological structures from various sources of remote sensing image information; according to the actual location of each geological structure in the remote sensing image information from each source, the feature characteristics of each geological structure are arranged to obtain the feature arrangement map corresponding to each source of remote sensing image information; the feature arrangement maps corresponding to each source of remote sensing image information are merged according to the actual location to obtain the output structural interpretation map of the mineral cluster area to be identified; based on the content of each element in different regions identified by the remote sensing image information, the content of each element in different regions is marked according to the actual location in the output structural interpretation map to obtain the remote sensing structural interpretation map.

[0032] The spectral bands in the images acquired by the Landsat 8 land imager can be selected according to the actual situation. In practical use, publicly available maps and survey information can also be selected, such as Baidu Maps, Google Maps, and regional survey information. The feature elements include one or more of the following: linear structures, ring structures, zonal features, color features, and block features. In the pre-defined correspondence between different feature elements and geological structures, linear structures represent faults, ring structures represent arcuate faults or folds, zonal features represent alteration zones, color features represent hydrothermal alteration, and block features represent intrusive rock masses. The aforementioned linear structures, ring structures, zonal features, color features, and block features are all features of existing technology in the interpretation maps, and therefore will not be elaborated upon in this application. Specifically, the pre-defined correspondence between different feature elements and geological structures can be set according to the actual situation, and this application does not limit this. Based on the actual location of each geological structure in the remote sensing image information from each source, the corresponding feature elements of each geological structure are arranged to obtain a feature arrangement map corresponding to each source of remote sensing image information. This arrangement can be based on the actual coordinates of each geological structure's actual location. The feature arrangement maps corresponding to each source of remote sensing image information are then merged according to their actual locations to obtain the output structural interpretation map of the mineralized area to be identified. The feature arrangement maps corresponding to each source of remote sensing image information can be fused, and the elements in this fused map should include those in the feature arrangement maps corresponding to each source of remote sensing image information. Based on the content of each element in different regions identified through the remote sensing image information, the content of each element in different regions is marked according to their actual locations in the output structural interpretation map, resulting in the remote sensing structural interpretation map. The content of each element in different regions identified through the remote sensing image information is calculated using a spectral matching method on image information from multiple spectral bands acquired by the land imager. The method for calculating the content of each element in different regions by using spectral matching to obtain image information of multiple spectral bands acquired by the land imager can be found in existing technologies.

[0033] In one possible implementation, the preset reference geological information includes: preset anomalous content ranges corresponding to different elements, and a preset set of anomalous geological structures; the region in the remote sensing structural interpretation map that satisfies the preset reference geological information is marked as an alteration anomaly region; compiling a remote sensing alteration anomaly map based on the identified alteration anomaly regions includes: identifying elemental anomaly regions where the content of any element satisfies the corresponding anomalous content range based on the preset anomalous content ranges corresponding to different elements and the content of each element in different regions of the remote sensing structural interpretation map; and identifying geological anomaly regions in the remote sensing structural interpretation map where the geological structures satisfy the preset set of anomalous geological structures based on the preset set of anomalous geological structures.

[0034] The preset abnormal content ranges corresponding to different elements and the preset abnormal geological structure set can be manually set according to actual conditions. In one example, the preset abnormal content ranges corresponding to different elements and the preset abnormal geological structure set should cover the geological structures and element content ranges of known mineral deposits in the region. The preset abnormal geological structure set includes: hydrothermal alteration, alteration zones, arcuate faults, and folds. The merged area of ​​the elemental anomaly region and the geological anomaly region in the remote sensing structural interpretation map is identified, and this merged area is taken as the alteration anomaly region. Here, "merging" means that the alteration anomaly region should include all elemental anomaly regions and geological anomaly regions. The alteration anomaly region is marked in the remote sensing structural interpretation map, and the marking result is taken as the remote sensing alteration anomaly map. It should be noted that in actual use, the content of different elements varies greatly among different mineral deposits; therefore, key elements, such as those in the main ores or compounds, can be selected.

[0035] In one possible implementation, based on the location information of multiple known mineral deposits obtained in advance, the geological information corresponding to the locations of the multiple known mineral deposits in the remote sensing alteration anomaly map is identified to obtain the geological information of the multiple known mineral deposits; based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit, prospecting prediction feature models corresponding to various mineral deposit types are created, including: based on the location information of the multiple known mineral deposits, identifying the actual content of each element and the actual geological structure at the location of each known mineral deposit in the remote sensing alteration anomaly map; for any known mineral deposit, marking the identified actual content of each element and the actual geological structure as the mineral deposit characteristics of that known mineral deposit; and based on the mineral deposit characteristics and the mineral deposit type of each known mineral deposit, creating prospecting prediction feature models corresponding to different mineral deposit types.

[0036] The prospecting prediction feature models corresponding to different mineral deposit types include the mineral deposit characteristics of one or more known mineral deposits belonging to that type. In actual prospecting, known mineral deposits can be large-scale deposits; the specific criteria for large-scale deposits can be set according to actual conditions. Screening for large-scale deposits can prevent interference from small deposits. When identifying the actual element content and geological structure at the location of each known mineral deposit in the remote sensing alteration anomaly map based on the location information of multiple known mineral deposits, the remote sensing alteration anomaly map also includes geological element content information and geological type information at the location of known mineral deposits, since the preset abnormal content ranges corresponding to different elements and the preset abnormal geological structure set should cover known mineral deposits. Therefore, in practical use, the actual element content and actual geological structure at the location of each known mineral deposit can be directly identified through the remote sensing alteration anomaly map. The identified actual element content and actual geological structure are then marked as the mineral deposit characteristics. Further, based on the mineral deposit characteristics and mineral deposit types of each known mineral deposit, prospecting prediction feature models corresponding to different mineral deposit types are identified. For example, if deposit A ​​corresponds to deposit feature a, deposit B corresponds to deposit feature b, and deposit C corresponds to deposit feature c, and deposit A ​​and deposit B are copper deposits and deposit C is a coal deposit, then when identifying the prospecting prediction feature models corresponding to different deposit types, the prospecting prediction feature model for copper deposits can be a+b, and the prospecting prediction feature model for coal deposits can be c.

[0037] In one possible implementation, the calculation of the local sensitive hash set of the prospecting prediction feature model for each type of mineral deposit includes: calculating the mapping vector of each mineral deposit feature in the prospecting prediction feature model for each type of mineral deposit using a word embedding algorithm, and using the calculated mapping vector as the feature vector of that type of mineral deposit; calculating the hash value of the feature vector of each type of mineral deposit using a hash algorithm; creating a random projection matrix using a random projection method; inputting the hash value of the feature vector of each type of mineral deposit into the random projection matrix, and using the output projection result as the local sensitive hash set of the prospecting prediction feature model for that type of mineral deposit. In one possible implementation, calculating the hash values ​​of feature vectors for various mineral deposit types using a hash algorithm includes: calculating the hash values ​​of mapping vectors corresponding to multiple known mineral deposits within the feature vectors of various mineral deposit types using a hash algorithm; creating a random projection matrix using a random projection method; inputting the hash values ​​of feature vectors for various mineral deposit types into the random projection matrix, and using the output projection result as the local sensitive hash set of the prospecting prediction feature model for that mineral deposit type, includes: randomly generating multiple unit vectors of preset dimensions; merging the multiple unit vectors of preset dimensions to obtain the random projection matrix; inputting the hash values ​​of mapping vectors corresponding to multiple known mineral deposits for various mineral deposit types into the random projection matrix to obtain the output local sensitive hashes of mapping vectors corresponding to multiple known mineral deposits for various mineral deposit types; and merging the local sensitive hashes of mapping vectors corresponding to multiple known mineral deposits for various mineral deposit types separately to obtain the local sensitive hash set of the prospecting prediction feature model for various mineral deposit types.

[0038] Specifically, a word embedding algorithm is used to calculate the mapping vector of each mineral deposit feature in the prospecting prediction feature model for each type of mineral deposit. This calculated mapping vector is then used as the feature vector for that type of mineral deposit. The word embedding algorithm can be used with mapping tools such as Word2Vec (a correlation model used to generate word vectors) to input the mineral deposit features into the prospecting prediction feature model for each type of mineral deposit, resulting in the output feature vector for that type of mineral deposit. A hash algorithm is then used to calculate the hash value of one or more feature vectors corresponding to each prospecting prediction feature model. Specific hash algorithms can be found in existing technologies. Multiple unit vectors of preset dimensions are randomly generated. These unit vectors are then merged to obtain the random projection matrix. The hash values ​​of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types are input into the random projection matrix to obtain the local sensitivity hashes of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types. Specifically, a random projection matrix can be created using a random projection method, generating k random unit vectors r1, r2, ..., rk. These k random unit vectors are merged to obtain the random projection matrix. This projection matrix can be used to project the aligned hash array into a binary string, achieving dimensionality reduction and obtaining the local sensitivity hashes. Then, the local sensitivity hashes of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types are merged separately to obtain the local sensitivity hash set of the prospecting prediction feature model for each mineral deposit type. The preset dimensions can be set according to the situation, for example, it can be 8 dimensions.

[0039] In one possible implementation, the step of calculating the local sensitive hashes of multiple target locations in the remote sensing alteration anomaly map; calculating the similarity between the local sensitive hash of each target location and the local sensitive hash sets of various mineral deposits; and identifying areas with a similarity greater than a preset threshold as mineral exploration prediction areas includes: extracting the element content and geological structure of multiple target locations in the remote sensing alteration anomaly map to obtain the target content of each element and the target geological structure; calculating the target vector of each element target content and the target geological structure using a word embedding algorithm; calculating the hash value of the target vector using a hash algorithm; inputting the hash value of the target vector into the random projection matrix to obtain the local sensitive hashes of multiple target locations; wherein the target locations are manually selected locations or randomly selected locations; calculating the cosine similarity between the local sensitive hash of each target location and the local sensitive hash sets of various mineral deposits; and marking the target location as the mineral exploration prediction area if the cosine similarity corresponding to any target location is greater than a preset threshold.

[0040] This process involves extracting the elemental content and geological structure of multiple target locations from the remote sensing alteration anomaly map to obtain the target element content and target geological structure. A word embedding algorithm is used to calculate the target vectors for each element's target content and target geological structure, which can be performed using the same algorithm as described in the previous embodiment. Then, a hash algorithm is used to calculate the hash value of the target vectors. Finally, the hash value of the target vectors is input into the random projection matrix to obtain the local sensitive hashes of multiple target locations. The cosine similarity between the local sensitive hash of each target location and the local sensitive hash sets of various mineral deposits is calculated. If the cosine similarity of the local sensitive hash sets corresponding to any target location and any type of mineral deposit is greater than a preset threshold, then the target location is marked as the mineral exploration prediction area. In practical use, this preset threshold can be set according to the actual situation.

[0041] In this embodiment, by introducing Locality Sensitive Hashing (LSH), two adjacent data points in the original data space are highly likely to remain adjacent in the new data space after undergoing the same mapping or projection transformation, while non-adjacent data points are unlikely to be mapped to the same bucket. This principle enables the delineation of mineral exploration prediction areas. This method not only prevents the inaccuracies and inefficiencies of manually delineating prediction areas based on experience, improving the accuracy and efficiency of the delineation, but also, relying on the principle of LSH, ensures that the cosine similarity of LSH sets corresponding to similar mineral deposit types and geological conditions exceeds a preset threshold, thereby achieving automated calculation and reducing manual costs. Furthermore, those skilled in the art will recognize that LSH is suitable for finding similar data in large datasets, thus offering significant advantages when dealing with large datasets in mineral exploration.

[0042] A second aspect of this application provides a mining apparatus based on locality-sensitive hashing, see [link to relevant documentation]. Figure 2 , Figure 2 A schematic diagram of a locality-sensitive hash-based mining device provided in this application embodiment, the device comprising: The interpretation map creation module 201 is used to acquire remote sensing image information of the mineral cluster area to be identified; and to interpret the geological structure of the remote sensing image information to obtain a remote sensing structural interpretation map of the mineral cluster area to be identified. Anomaly map creation module 202 is used to mark areas in the remote sensing structural interpretation map that meet preset reference geological information as alteration anomaly areas; and to compile remote sensing alteration anomaly maps based on the identified alteration anomaly areas. The model creation module 203 is used to identify the geological information corresponding to the location of the multiple known mineral deposits in the remote sensing alteration anomaly map based on the location information of the multiple known mineral deposits obtained in advance, and to obtain the geological information of the multiple known mineral deposits; and to create mineral exploration prediction feature models corresponding to multiple mineral deposit types based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit. The prediction area identification module 204 is used to calculate the local sensitive hash set of the prospecting prediction feature model for each type of mineral deposit; calculate the local sensitive hash of multiple target locations in the remote sensing alteration anomaly map; calculate the similarity between the local sensitive hash of each target location and the local sensitive hash set of various mineral deposits; and determine the area with a similarity greater than a preset threshold as the prospecting prediction area.

[0043] In one possible implementation, the step of acquiring remote sensing image information of the mineral cluster area to be identified and interpreting the geological structure of the remote sensing image information to obtain a remote sensing structural interpretation map of the mineral cluster area to be identified includes: The remote sensing image information of the mineral cluster area to be identified is obtained from multiple sources, including: remote sensing image information collected by Pleiades and image information of multiple spectral bands collected by the Landsat 8 land imager. Based on the pre-defined correspondence between different element features and geological structures, the remote sensing image information from multiple sources is interpreted to obtain the element features corresponding to each geological structure in the remote sensing image information from multiple sources. The element features include one or more of the following: linear structure, ring structure, zone feature, color feature, and block feature. In the pre-defined correspondence between different element features and geological structures, the linear structure is used to represent faults, the ring structure is used to represent arcuate faults or folds, the zone feature is used to represent alteration zones, the color feature is used to represent hydrothermal alteration, and the block feature is used to represent intrusive rock masses. According to the actual location of each geological structure in the remote sensing image information from each source, the feature characteristics corresponding to each geological structure are arranged to obtain the feature arrangement map corresponding to the remote sensing image information from each source; the feature arrangement maps corresponding to the remote sensing image information from each source are merged according to their actual locations to obtain the output structural interpretation map of the mineral cluster area to be identified. Based on the content of each element in different regions identified through the remote sensing image information, the content of each element in different regions is marked according to the actual location in the structural interpretation map to be output, thereby obtaining the remote sensing structural interpretation map; wherein, the content of each element in different regions identified through the remote sensing image information is calculated by using the spectral matching method to analyze the image information of multiple spectral bands acquired by the land imager.

[0044] In one possible implementation, the preset reference geological information includes: preset abnormal content ranges corresponding to different elements, and preset abnormal geological structure sets; The process of marking areas in the remote sensing structural interpretation map that meet preset reference geological information as alteration anomaly areas; and compiling a remote sensing alteration anomaly map based on the identified alteration anomaly areas, including: Based on the preset abnormal content ranges corresponding to different elements and the content of each element in different regions of the remote sensing construction interpretation map, identify the abnormal regions of any element whose content satisfies the corresponding abnormal content range. Based on the preset set of anomalous geological structures, geological anomalous regions in the remote sensing structural interpretation map that satisfy the preset set of anomalous geological structures are identified, wherein the preset set of anomalous geological structures includes: hydrothermal alteration, alteration zone, arcuate fault, and fold. Identify the merged region of the elemental anomaly region and the geological anomaly region in the remote sensing structural interpretation map, and take the merged region as the alteration anomaly region; The alteration anomaly region is marked in the remote sensing structural interpretation map, and the marking result is used as the remote sensing alteration anomaly map.

[0045] In one possible implementation, based on the pre-acquired location information of multiple known mineral deposits, the geological information corresponding to the locations of the multiple known mineral deposits in the remote sensing alteration anomaly map is identified to obtain the geological information of the multiple known mineral deposits; based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit, prospecting prediction feature models corresponding to various mineral deposit types are created, including: Based on the location information of multiple known mineral deposits, the actual content of each element and the actual geological structure at the location of each known mineral deposit in the remote sensing alteration anomaly map are identified; for any known mineral deposit, the identified actual content of each element and the actual geological structure are marked as the mineral deposit characteristics of that known mineral deposit. Based on the mineral deposit characteristics and mineral deposit types of each known mineral deposit, mineral exploration prediction feature models corresponding to different mineral deposit types are created. The mineral exploration prediction feature models corresponding to different mineral deposit types include mineral deposit characteristics of one or more known mineral deposits belonging to that mineral deposit type.

[0046] In one possible implementation, calculating the local sensitive hash set of the prospecting prediction feature model for each type of mineral deposit includes: Using a word embedding algorithm, the mapping vector of each mineral deposit feature in the mineral exploration prediction feature model for each type of mineral deposit is calculated, and the calculated mapping vector is used as the feature vector of that type of mineral deposit; using a hash algorithm, the hash value of the feature vector of each type of mineral deposit is calculated. A random projection matrix is ​​created using the random projection method; the hash values ​​of the feature vectors of various deposit types are input into the random projection matrix, and the output projection result is used as the local sensitive hash set of the prospecting prediction feature model for that deposit type.

[0047] In one possible implementation, calculating the hash value of the feature vectors for various mineral deposit types using a hash algorithm includes: The hash value of the mapping vector corresponding to multiple known mineral deposits in the feature vector of various mineral deposit types is calculated using a hash algorithm. The process involves creating a random projection matrix using a random projection method; inputting the hash values ​​of the feature vectors for various mineral deposit types into the random projection matrix; and using the output projection results as the local sensitive hash set of the mineral exploration prediction feature model for that type of mineral deposit, including: Randomly generate multiple unit vectors of preset dimensions; merge the multiple unit vectors of preset dimensions to obtain the random projection matrix; The hash values ​​of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types are input into the random projection matrix to obtain the local sensitive hashes of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types. The local sensitivity hashes of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types are merged separately to obtain the local sensitivity hash set of the mineral exploration prediction feature model for each mineral deposit type.

[0048] In one possible implementation, the step of calculating the local sensitive hashes of multiple target locations in the remote sensing alteration anomaly map; calculating the similarity between the local sensitive hash of each target location and the local sensitive hash sets of various mineral deposits; and determining areas with a similarity greater than a preset threshold as mineral exploration prediction areas includes: Extract the elemental content and geological structure of multiple target locations from the remote sensing alteration anomaly map to obtain the target element content and target geological structure; calculate the target vector of each element target content and target geological structure using a word embedding algorithm; calculate the hash value of the target vector using a hash algorithm; input the hash value of the target vector into the random projection matrix to obtain the local sensitive hash of multiple target locations; wherein, the target locations are manually selected locations or randomly selected locations; Calculate the local sensitivity hash of each target location and the cosine similarity of the local sensitivity hash sets of various mineral deposits; If the cosine similarity of any target location is greater than a preset threshold, then the target location is marked as the mineral exploration prediction area.

[0049] As can be seen, the scheme of this application can create a remote sensing structural interpretation map after acquiring remote sensing image information, and then create a local sensitive hash set based on the characteristics of known mineral deposits in the remote sensing structural interpretation map. Furthermore, by using the local sensitive hash set and the local sensitive hash of each target location, the similarity is calculated to identify the mineral exploration prediction area. This achieves the delineation of the prediction area without relying on experience by humans, which not only avoids human error, but also improves the efficiency of mineral exploration.

[0050] This application also provides an electronic device, such as... Figure 3 As shown, it includes: Memory 301 is used to store computer programs; When processor 302 executes a program stored in memory 301, it performs the following steps: Acquire remote sensing image information of the mineral cluster area to be identified; interpret the geological structure of the remote sensing image information to obtain a remote sensing structural interpretation map of the mineral cluster area to be identified; The areas in the remote sensing structural interpretation map that meet the preset reference geological information are marked as alteration anomaly areas; a remote sensing alteration anomaly map is compiled based on the identified alteration anomaly areas; Based on the location information of multiple known mineral deposits obtained in advance, the geological information corresponding to the locations of the multiple known mineral deposits in the remote sensing alteration anomaly map is identified to obtain the geological information of multiple known mineral deposits; based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit, mineral exploration prediction feature models corresponding to various mineral deposit types are created. Calculate the local sensitive hash set of the prospecting prediction feature model for each type of mineral deposit; calculate the local sensitive hash of multiple target locations in the remote sensing alteration anomaly map; calculate the similarity between the local sensitive hash of each target location and the local sensitive hash set of various mineral deposits; and determine the regions with a similarity greater than a preset threshold as prospecting prediction areas.

[0051] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0052] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0053] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0054] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0055] In another embodiment provided in this application, a computer-readable storage medium is also provided, which stores a computer program that, when executed by a processor, implements the steps of any of the above-described mining methods based on locality-sensitive hashing.

[0056] In another embodiment provided in this application, a computer program product containing instructions is also provided, which, when run on a computer, causes the computer to execute any of the locality-sensitive hashing-based mining methods described in the above embodiments.

[0057] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a solid-state drive (SSD), etc.

[0058] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0059] The various embodiments in this specification are described in a related manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on its differences from other embodiments. In particular, the embodiments for apparatus, electronic devices, and storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0060] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application are included within the scope of protection of this application.

Claims

1. A mineral exploration method based on locality-sensitive hashing, characterized in that, The method includes: Acquire remote sensing image information of the mineral cluster area to be identified; interpret the geological structure of the remote sensing image information to obtain a remote sensing structural interpretation map of the mineral cluster area to be identified; The areas in the remote sensing structural interpretation map that meet the preset reference geological information are marked as alteration anomaly areas; a remote sensing alteration anomaly map is compiled based on the identified alteration anomaly areas; Based on the location information of multiple known mineral deposits obtained in advance, the geological information corresponding to the locations of the multiple known mineral deposits in the remote sensing alteration anomaly map is identified to obtain the geological information of multiple known mineral deposits; based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit, mineral exploration prediction feature models corresponding to various mineral deposit types are created. Calculate the local sensitive hash set of the prospecting prediction feature model for each type of mineral deposit; calculate the local sensitive hash of multiple target locations in the remote sensing alteration anomaly map; calculate the similarity between the local sensitive hash of each target location and the local sensitive hash set of various mineral deposits; and determine the regions with a similarity greater than a preset threshold as prospecting prediction areas.

2. The method according to claim 1, characterized in that, The process of acquiring remote sensing image information of the mineral cluster to be identified, and interpreting the geological structure of the remote sensing image information to obtain a remote sensing structural interpretation map of the mineral cluster to be identified, includes: The remote sensing image information of the mineral cluster area to be identified is obtained from multiple sources, including: remote sensing image information collected by the Pleiades satellite and image information of multiple spectral bands collected by the Landsat 8 land imager. Based on the pre-defined correspondence between different element features and geological structures, the remote sensing image information from multiple sources is interpreted to obtain the element features corresponding to each geological structure in the remote sensing image information from multiple sources. The element features include one or more of the following: linear structure, ring structure, zone feature, color feature, and block feature. In the pre-defined correspondence between different element features and geological structures, the linear structure is used to represent faults, the ring structure is used to represent arcuate faults or folds, the zone feature is used to represent alteration zones, the color feature is used to represent hydrothermal alteration, and the block feature is used to represent intrusive rock masses. According to the actual location of each geological structure in the remote sensing image information from each source, the feature characteristics corresponding to each geological structure are arranged to obtain the feature arrangement map corresponding to the remote sensing image information from each source; the feature arrangement maps corresponding to the remote sensing image information from each source are merged according to their actual locations to obtain the output structural interpretation map of the mineral cluster area to be identified. Based on the content of each element in different regions identified through the remote sensing image information, the content of each element in different regions is marked according to the actual location in the structural interpretation map to be output, thereby obtaining the remote sensing structural interpretation map; wherein, the content of each element in different regions identified through the remote sensing image information is calculated by using the spectral matching method to analyze the image information of multiple spectral bands acquired by the land imager.

3. The method according to claim 1, characterized in that, The preset reference geological information includes: preset abnormal content ranges corresponding to different elements, and preset abnormal geological structure sets; The process of marking areas in the remote sensing structural interpretation map that meet preset reference geological information as alteration anomaly areas; and compiling a remote sensing alteration anomaly map based on the identified alteration anomaly areas, including: Based on the preset abnormal content ranges corresponding to different elements and the content of each element in different regions of the remote sensing construction interpretation map, identify the abnormal regions of any element whose content satisfies the corresponding abnormal content range. Based on the preset set of anomalous geological structures, geological anomalous regions in the remote sensing structural interpretation map that satisfy the preset set of anomalous geological structures are identified, wherein the preset set of anomalous geological structures includes: hydrothermal alteration, alteration zone, arcuate fault, and fold. Identify the merged region of the elemental anomaly region and the geological anomaly region in the remote sensing structural interpretation map, and take the merged region as the alteration anomaly region; The alteration anomaly region is marked in the remote sensing structural interpretation map, and the marking result is used as the remote sensing alteration anomaly map.

4. The method according to claim 1, characterized in that, Based on the location information of multiple known mineral deposits obtained in advance, the geological information corresponding to the locations of the multiple known mineral deposits in the remote sensing alteration anomaly map is identified, and the geological information of the multiple known mineral deposits is obtained. Based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit, prospecting prediction feature models corresponding to various mineral deposit types are created, including: Based on the location information of multiple known mineral deposits, the actual content of each element and the actual geological structure at the location of each known mineral deposit in the remote sensing alteration anomaly map are identified; for any known mineral deposit, the identified actual content of each element and the actual geological structure are marked as the mineral deposit characteristics of that known mineral deposit. Based on the mineral deposit characteristics and mineral deposit types of each known mineral deposit, mineral exploration prediction feature models corresponding to different mineral deposit types are created. The mineral exploration prediction feature models corresponding to different mineral deposit types include mineral deposit characteristics of one or more known mineral deposits belonging to that mineral deposit type.

5. The method according to claim 4, characterized in that, The local sensitive hash set for calculating the prospecting prediction feature model for each type of mineral deposit includes: Using a word embedding algorithm, the mapping vector of each mineral deposit feature in the mineral exploration prediction feature model for each type of mineral deposit is calculated, and the calculated mapping vector is used as the feature vector of that type of mineral deposit; using a hash algorithm, the hash value of the feature vector of each type of mineral deposit is calculated. A random projection matrix is ​​created using the random projection method; the hash values ​​of the feature vectors of various deposit types are input into the random projection matrix, and the output projection result is used as the local sensitive hash set of the prospecting prediction feature model for that deposit type.

6. The method according to claim 5, characterized in that, The process of calculating the hash values ​​of feature vectors for various mineral deposit types using a hash algorithm includes: The hash value of the mapping vector corresponding to multiple known mineral deposits in the feature vector of various mineral deposit types is calculated using a hash algorithm. The process involves creating a random projection matrix using a random projection method; inputting the hash values ​​of the feature vectors for various mineral deposit types into the random projection matrix; and using the output projection results as the local sensitive hash set of the mineral exploration prediction feature model for that type of mineral deposit, including: Randomly generate multiple unit vectors of preset dimensions; merge the multiple unit vectors of preset dimensions to obtain the random projection matrix; The hash values ​​of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types are input into the random projection matrix to obtain the local sensitive hashes of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types. The local sensitivity hashes of the mapping vectors of multiple known mineral deposits corresponding to various mineral deposit types are merged separately to obtain the local sensitivity hash set of the mineral exploration prediction feature model for each mineral deposit type.

7. The method according to claim 5, characterized in that, The calculation of local sensitive hashes for multiple target locations in the remote sensing alteration anomaly map; and the calculation of the similarity between the local sensitive hash of each target location and the local sensitive hash sets of various mineral deposits. Regions with a similarity greater than a preset threshold are identified as mineral exploration prediction areas, including: Extract the elemental content and geological structure of multiple target locations from the remote sensing alteration anomaly map to obtain the target element content and target geological structure; calculate the target vector of each element target content and target geological structure using a word embedding algorithm; calculate the hash value of the target vector using a hash algorithm; input the hash value of the target vector into the random projection matrix to obtain the local sensitive hash of multiple target locations; wherein, the target locations are manually selected locations or randomly selected locations; Calculate the local sensitivity hash of each target location and the cosine similarity of the local sensitivity hash sets of various mineral deposits; If the cosine similarity of any target location is greater than a preset threshold, then the target location is marked as the mineral exploration prediction area.

8. A mineral exploration device based on locality-sensitive hashing, characterized in that, The device includes: The interpretation map creation module is used to acquire remote sensing image information of the mineral cluster area to be identified; and to interpret the geological structure of the remote sensing image information to obtain the remote sensing structural interpretation map of the mineral cluster area to be identified. Anomaly map creation module is used to mark areas in the remote sensing structural interpretation map that meet preset reference geological information as alteration anomaly areas; and to compile remote sensing alteration anomaly maps based on the identified alteration anomaly areas. The model creation module is used to identify the geological information corresponding to the locations of the multiple known mineral deposits in the remote sensing alteration anomaly map based on the location information of multiple known mineral deposits obtained in advance, and to obtain the geological information of multiple known mineral deposits; based on the geological information of the multiple known mineral deposits and the mineral deposit type of each known mineral deposit, to create mineral exploration prediction feature models corresponding to various mineral deposit types respectively; The prediction area identification module is used to calculate the local sensitive hash set of the prospecting prediction feature model for each type of mineral deposit; calculate the local sensitive hash of multiple target locations in the remote sensing alteration anomaly map; calculate the similarity between the local sensitive hash of each target location and the local sensitive hash set of various mineral deposits; and identify the areas with a similarity greater than a preset threshold as prospecting prediction areas.

9. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, when executing a program stored in memory, implements the method described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium contains a computer program that, when executed by a processor, implements the method described in any one of claims 1-7.

Images