Variable rock zone identification method and device, medium and electronic equipment

Abstract

The invention relates to an altered rock zone identification method and device, a medium and electronic equipment, and relates to the technical field of geological reconnaiss.After hyperspectral data and gamma-ray energy spectrum data of a to-be-detected area are obtained, the hyperspectral data and the gamma-ray energy spectrum data are preprocessed to obtain ground reflection data of ground features and raster data; then, dimensionality reduction and / or noise reduction processing is carried out on the ground reflection data of the ground features, and first-class feature data is obtained; performing feature extraction on the raster data to obtain a second type of feature data; and obtaining at least one group of fusion data by using the first type of feature data and the second type of feature data, determining target fusion data from the at least one group of fusion data, and determining alteration rock zone information of the to-be-detected region by using the target fusion data. Therefore, feature fusion of the hyperspectral data and the gamma-ray energy spectrum data is realized, and the altered rock zone information is analyzed by using the fused data, so that the determined altered rock zone information can be more accurate.

Description

Technical Field

[0001] The present disclosure relates to the technical field of geological exploration, and in particular, to a method, device, medium, and electronic device for identifying altered rock zones. Background Art

[0002] With the development of science and technology, in the process of geological exploration, through the use of hyperspectral data or gamma-ray spectrometry data for preliminary geological exploration, it is no longer necessary for users to reach the actual exploration scene for exploration, thus improving the exploration efficiency.

[0003] Hyperspectral data has rich spectral information of ground objects, and has significant effects on the identification and classification of minerals and lithological components, as well as the delineation of altered zones. Airborne gamma-ray spectrometry measurement is an important nuclear geophysical method, which reflects the types, contents, and distributions of radioactive elements uranium (U), thorium (Th), potassium (K), and other radioactive nuclides in different rocks and ground objects. Summary of the Invention

[0004] The purpose of the present disclosure is to provide a method, device, medium, and electronic device for identifying altered rock zones. The method for identifying altered rock zones in the present disclosure focuses on the feature fusion of hyperspectral data and gamma-ray spectrometry data, and uses the fused data to analyze the information of altered rock zones, so that the determined information of altered rock zones can be more accurate.

[0005] To achieve the above purpose, in a first aspect, the present disclosure provides a method for identifying an altered rock zone, including: obtaining hyperspectral data and gamma-ray spectrometry data of a region to be detected; performing preprocessing on the hyperspectral data to obtain ground reflection data of ground objects, and performing preprocessing on the gamma-ray spectrometry data to obtain grid data; performing feature extraction processing on the ground reflection data according to a first feature extraction method to obtain first-class feature data; and performing feature extraction processing on the grid data according to a second feature extraction method to obtain second-class feature data; wherein the first feature extraction method includes dimensionality reduction and / or noise reduction; fusing the first-class feature data and the second-class feature data according to a predefined method to obtain at least one set of fused data; determining target fused data from at least one set of fused data, and using the target fused data to determine the information of the altered rock zone in the region to be detected.

[0006] Optionally, determining the target fused data from at least one set of fused data includes: determining the lithological classification accuracy corresponding to each set of fused data in at least one set of fused data; determining the target fused data from the at least one set of fused data according to the lithological classification accuracy corresponding to each set of fused data.

[0007] Optionally, the above method further includes: Obtaining rock type samples for training and rock type samples for verification from a pre-built database; Performing classification learning on a pre-selected support vector machine using the rock type samples for training, and performing lithology classification on the at least one set of fusion data using the learned support vector machine; Verifying the lithology classification results corresponding to each set of fusion data using the rock type samples for verification, and determining the lithology classification accuracy corresponding to each set of fusion data.

[0008] Optionally, after preprocessing the hyperspectral data to obtain ground reflection data of ground objects and preprocessing the gamma energy spectrum data to obtain raster data, the above method further includes: Performing interpolation registration processing on the ground reflection data and the raster data so that the spatial sizes of the ground reflection data and the raster data are the same.

[0009] Optionally, the above-mentioned feature extraction processing of the ground reflection data according to the first feature extraction method to obtain the first type of feature data includes: Performing minimum noise separation processing on the ground reflection data to obtain noise-reduced feature data; And performing principal component analysis processing on the ground reflection data to obtain spectral band dimensionality reduction feature data; Wherein, the first type of feature data includes noise-reduced feature data and spectral band dimensionality reduction feature data.

[0010] Optionally, the above-mentioned feature extraction processing of the raster data according to the second feature extraction method to obtain the second type of feature data includes: Determining key elements; Determining key element data and ratio data of key elements from the above raster data, and determining the total channel data of the above raster data; Wherein, the second type of feature data includes key element data, ratio data of key elements, and total channel data.

[0011] Optionally, the above-mentioned determination of the altered rock zone information of the area to be detected using the above target fusion data includes: Performing spectral analysis on the ground reflection data to obtain a mineral species identification result; Performing density segmentation on the second type of feature data to obtain uranium high-field data; Based on the above target fusion data, mineral species identification result, and the uranium high-field data, determining the altered rock zone information of the area to be detected.

[0012] In a second aspect, the present disclosure provides an altered rock zone identification device, including: An acquisition unit for acquiring hyperspectral data and gamma energy spectrum data of a region to be detected; A preprocessing unit for preprocessing the hyperspectral data to obtain ground reflection data of ground objects, and preprocessing the gamma energy spectrum data to obtain raster data; A feature extraction unit for performing feature extraction processing on the ground reflection data according to a first feature extraction method to obtain first-class feature data; and performing feature extraction processing on the raster data according to a second feature extraction method to obtain second-class feature data; wherein, the first feature extraction method includes dimensionality reduction and / or noise reduction; A fusion unit for fusing the first-class feature data and the second-class feature data according to a predefined method to obtain at least one set of fusion data; A determination unit for determining target fusion data from at least one set of fusion data, and using the target fusion data to determine the altered rock belt information of the region to be detected.

[0013] In a third aspect, the present disclosure provides a non-transitory computer-readable storage medium having a computer program stored thereon, and when the program is executed by a processor, the steps of the method shown in any optional manner of the first aspect are implemented.

[0014] In a fourth aspect, the present disclosure provides an electronic device, including: A memory having a computer program stored thereon; A processor for executing the computer program in the memory to implement the steps of the method shown in any optional manner of the first aspect.

[0015] Through the above technical solutions, after acquiring the hyperspectral data and gamma energy spectrum data of the region to be detected, the hyperspectral data and gamma energy spectrum data are respectively preprocessed to obtain the ground reflection data of ground objects and raster data; then, dimensionality reduction and / or noise reduction processing is performed on the ground reflection data of ground objects to obtain first-class feature data; and feature extraction is performed on the raster data to obtain second-class feature data; and at least one set of fusion data is obtained by using the first-class feature data and the second-class feature data, and target fusion data is determined from at least one set of fusion data, and the altered rock belt information of the region to be detected is determined by using the target fusion data. In this way, feature fusion of hyperspectral data and gamma energy spectrum data is realized, and the altered rock belt information is analyzed by using the fused data, so that the determined altered rock belt information can be more accurate.

[0016] Other features and advantages of the present disclosure will be described in detail in the subsequent specific implementation section. Description of the Drawings

[0017] The accompanying drawings are used to provide a further understanding of the present disclosure, and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure, but do not constitute a limitation to the present disclosure. In the accompanying drawings: Figure 1 It is a schematic flowchart of a method for identifying altered rock zones provided by an exemplary embodiment of the present disclosure.

[0018] Figure 2 It is a schematic diagram of the process of a method for identifying altered rock zones provided by an exemplary embodiment of the present disclosure.

[0019] Figure 3 It is a block diagram of an apparatus for identifying altered rock zones provided by an exemplary embodiment of the present disclosure.

[0020] Figure 4 A block diagram of an electronic device provided by an exemplary embodiment of the present disclosure.

[0021] Figure 5 A block diagram of an electronic device provided by another exemplary embodiment of the present disclosure. Specific Embodiments

[0022] The following provides a detailed description of the specific embodiments of the present disclosure with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are only used to illustrate and explain the present disclosure, and are not used to limit the present disclosure.

[0023] It should be noted that all actions of obtaining signals, information, or data in the present disclosure are carried out on the premise of complying with the corresponding data protection regulations and policies of the country where it is located, and with the authorization given by the owner of the corresponding device.

[0024] As can be seen from the description of the above background art, hyperspectral data and gamma ray spectrometry data have different advantages. Therefore, if hyperspectral data and gamma ray spectrometry data are combined for analysis, the hyperspectral data and gamma ray spectrometry data can be better utilized. For example, the information of radioactive element characteristics can be enhanced, while the influence of hyperspectral data affected by surface coverage interference can be reduced. Further, high-precision lithology classification can be achieved, and more information related to uranium mineralization can be obtained, which has special value for studying geological phenomena and laws from shallow to deep.

[0025] In the related art, the combination of remote sensing data (similar to hyperspectral data, a type of data obtained by non-contact means) and gamma ray spectrometry data for research is mainly divided into three types: one is the comprehensive analysis of multispectral data and airborne gamma ray spectrometry data; the second is the integration of multi-source geoscience information mainly based on airborne gamma ray spectrometry data; the third is to add hyperspectral data on the basis of multi-source information integration to improve the analysis accuracy. This type of combination is mainly applied to geological mapping, delimiting ore-forming structures, and optimizing ore prospecting target areas.

[0026] The alteration zone identification method provided by the present disclosure can fuse the characteristics of hyperspectral data and gamma ray spectrometry data, and use the fused data to analyze the alteration zone information, so that the determined alteration zone information can be more accurate.

[0027] Please refer to Figure 1 , Figure 1 which is a schematic flow chart of the alteration zone identification method provided by the present disclosure. As shown in Figure 1 , this alteration zone identification method may include the following steps: Step S101, obtain hyperspectral data and gamma ray spectrometry data of the area to be detected.

[0028] The hyperspectral data can be obtained using a hyperspectral instrument, which can obtain the reflection spectrum of an object within different wavelength ranges. In practical applications, remote sensing satellites or ground hyperspectral instruments can be used to obtain the data.

[0029] The gamma ray spectrometry data can be obtained through radioactive measurement. For example, in geological surveys, gamma ray spectrometers are used to measure the content of radioactive elements in rocks, soils, and water.

[0030] Of course, in the present disclosure, the specific methods for obtaining the hyperspectral data and gamma ray spectrometry data of the area to be detected are not limited, and only reasonable selection needs to be made according to the actual situation.

[0031] It should be understood that any area that needs to perform alteration zone identification and analysis can be understood as the area to be detected.

[0032] Step S102, preprocess the hyperspectral data to obtain the ground reflection data of the ground object, and preprocess the gamma ray spectrometry data to obtain raster data.

[0033] As an example, preprocessing the hyperspectral data may include removing inappropriate data and correcting the data. For example, through band-by-band inspection, bad bands and strong water vapor absorption bands can be removed, and then the data after removal can be calibrated and corrected. For example, calibration may include but is not limited to: radiometric calibration, stripe removal, Flaash atmospheric correction, orthorectification. That is, after preprocessing the hyperspectral data, the ground reflection data of the ground object is obtained. That is, after preprocessing the hyperspectral data, the obtained ground reflection data can be made more accurate.

[0034] As an example, based on the total air gamma radioactivity count and data of 3 natural radioactive elements (U, Th, K), the gamma ray spectrometry data can be converted into raster data. It should be noted that the gamma ray spectrometry data can be grid data. At this time, the grid data can be converted into raster data using Arcgis software (geographic information system software).

[0035] It should be understood that the hyperspectral data is preprocessed to obtain the ground reflection data of the ground object, and the gamma energy spectrum data is preprocessed to obtain raster data, so that it is convenient to perform feature extraction and fusion on the processed data.

[0036] Step S103: Perform feature extraction processing on the ground reflection data according to the first feature extraction method to obtain the first type of feature data; and perform feature extraction processing on the raster data according to the second feature extraction method to obtain the second type of feature data.

[0037] Here, the first type of feature extraction method includes dimensionality reduction and / or noise reduction.

[0038] As an example, although the ground reflection data is obtained by preprocessing the hyperspectral data, in the preprocessing stage, only part of the hyperspectral data is removed and corrected, so that the number of ground reflection data is still relatively large, resulting in a large number of bands and a large quantity of ground reflection data. Using such data as the data to be fused may lead to a low interpretation accuracy and poor effect of the finally obtained fused data.

[0039] Based on this, the present disclosure performs dimensionality reduction and / or noise reduction processing on the ground reflection data, so that the obtained first type of feature data not only has fewer bands, but also can contain most of the information in the ground reflection data (generally, it can contain more than 90% of the information). That is, the first type of feature extraction method can be understood as dimensionality reduction and noise reduction feature extraction.

[0040] As an example, the airborne gamma energy spectrum data collects the radioactive values of uranium (U), thorium (Th), and potassium (K). By analyzing the geochemical characteristics of these three elements and summarizing the migration and enrichment laws of the elements during the geological process, during the geological process, potassium (K) and uranium (U) are easy to migrate, while thorium (Th) has a lower migration property. The three elements and their ratios each have advantages in the analysis of rock masses, structures, and alterations. Therefore, based on this concept, feature extraction can be performed on the airborne gamma energy spectrum data to obtain the second type of feature data. That is, the second type of feature extraction method can be understood as feature extraction according to element characteristics.

[0041] Step S104: Fuse the first type of feature data and the second type of feature data according to a predefined method to obtain at least one set of fused data.

[0042] It should be understood that the first type of feature data may include multiple sub-data, and the second type of feature data may also include multiple sub-data. Fusing the first type of feature data and the second type of feature data can be understood as concatenating the sub-data in the first type of feature data with the sub-data in the second type of feature data to obtain at least one set of fused data.

[0043] It should be understood that the predefined method may include concatenating data and fusing it into an image. The specific fusion method can be defined according to the actual situation.

[0044] For example, the first type of feature data may include noise-reduced feature data obtained after noise reduction and spectral band dimensionality-reduced feature data obtained after dimensionality reduction. The second fusion data may include key element data, ratio data of key elements, and total channel data. By combining the sub-data in the first fusion data and the sub-data in the second fusion data, at least one set of fusion data can be obtained.

[0045] Step S105, determine target fusion data from at least one set of fusion data, and use the target fusion data to determine the altered rock zone information of the area to be detected.

[0046] As an example, after obtaining at least one set of fusion data, the target fusion data can be determined, and the altered rock zone information of the area to be detected can be determined using the target fusion data, so that the determined altered rock zone information can be more accurate.

[0047] As an example, each fusion data in at least one set of fusion data can be processed to determine the lithology classification accuracy corresponding to each fusion data, and the target fusion data can be determined according to the lithology classification accuracy. In this way, since the classification accuracy corresponding to the target fusion data is relatively high, when using the target fusion data to determine the altered rock zone information, the determined altered rock zone information can be more accurate.

[0048] In some implementation manners, the target fusion data may be a fusion image, and image recognition can be performed on the fusion image to determine the lithology classification. Then, the lithology classification accuracy can be determined using the verification samples.

[0049] It can be seen that in the present disclosure, after obtaining the hyperspectral data and gamma energy spectrum data of the area to be detected, the hyperspectral data and gamma energy spectrum data are respectively preprocessed to obtain the ground reflection data and raster data of the ground objects; then, the ground reflection data of the ground objects is subjected to dimensionality reduction and / or noise reduction processing to obtain the first type of feature data; and the raster data is subjected to feature extraction to obtain the second type of feature data; at least one set of fusion data is obtained using the first type of feature data and the second type of feature data, the target fusion data is determined from at least one set of fusion data, and the altered rock zone information of the area to be detected is determined using the target fusion data. In this way, the feature fusion of the hyperspectral data and the gamma energy spectrum data is realized, and the fused data is used to analyze the altered rock zone information, so that the determined altered rock zone information can be more accurate.

[0050] In some embodiments, "determining target fusion data from at least one set of fusion data" in step S105 may specifically include: Determining the lithology classification accuracy corresponding to each set of fusion data in at least one set of fusion data; Determining target fusion data from at least one set of fusion data according to the lithology classification accuracy corresponding to each set of fusion data.

[0051] It should be understood that the lithology classification accuracy corresponding to the target fusion data is the highest.

[0052] As an example, if the lithology classification accuracy is high, it can indicate that the fusion result can more clearly display the boundaries of geological bodies and the layers are more distinct; thus, the information of the original geological map can be enriched more, and it is beneficial to effectively revise the boundaries of rock masses; in this way, a more refined interpretation of geological bodies can be achieved overall.

[0053] That is, since the lithology classification accuracy corresponding to the target fusion data is the highest, it can make the determination of the altered rock zone information in the area to be detected using the target fusion data more accurate.

[0054] In some embodiments, the above-mentioned altered rock zone identification method may further include: Obtaining training rock type samples and verification rock type samples from a pre-built database; Performing classification learning on a pre-selected support vector machine using the training rock type samples, and performing lithology classification on at least one set of fusion data using the learned support vector machine; Verifying the lithology classification results corresponding to each set of fusion data using the verification rock type samples, and determining the lithology classification accuracy corresponding to each set of fusion data.

[0055] As an example, specific training samples to be selected are determined according to the actual situation. For example, the training rock type samples to be selected can be determined according to the possible rock types in the area to be measured. Correspondingly, the verification rock type samples can be determined according to the selected training rock type samples.

[0056] It should be understood that the number of each type of rock type sample in the training rock type samples is greater than the number of each type of rock type sample in the verification samples.

[0057] As an example, the rock types included in the training rock type samples may include: Caledonian diorite, late Caledonian granite, late Caledonian primary alkaline complex, two-mica quartz schist, biotite quartz schist, etc.

[0058] It should be understood that the pre-built database can also be limited according to the actual situation. For example, the pre-built database can be obtained by collecting and organizing the historical rock type image information.

[0059] It can be seen that in the method of the present disclosure, training rock samples and verification rock samples can be obtained from a pre-established database according to the possible rock types in the detection area to be measured. In this way, the support vector machine (SVM) can be used for lithology classification with the training rock samples, and the classification accuracy can be determined using the verification samples. Since the number of training rock samples can be greater than that of the verification rock samples, not only can the classification effect of the support vector machine after learning be more accurate, but also the lithology classification accuracy corresponding to each group of fusion data can be determined efficiently.

[0060] In some embodiments, after preprocessing the hyperspectral data to obtain the ground reflection data of the ground object and preprocessing the gamma energy spectrum data to obtain the grid data, the method further includes: Performing interpolation registration processing on the ground reflection data and the grid data to make the spatial sizes of the ground reflection data and the grid data the same.

[0061] It should be understood that after the spatial sizes of the ground reflection data and the grid data are the same, it is convenient to perform feature extraction on the ground reflection data and the grid data and then fuse the extracted features.

[0062] In some embodiments, the step of "performing feature extraction processing on the ground reflection data according to the first feature extraction method to obtain the first type of feature data" in step S103 may specifically include: Performing minimum noise separation processing on the above-mentioned ground reflection data to obtain noise reduction feature data; And performing principal component analysis processing on the above-mentioned ground reflection data to obtain spectral band dimensionality reduction feature data; Wherein, the first type of feature data includes noise reduction feature data and spectral band dimensionality reduction feature data.

[0063] As an example, the specific number of bands corresponding to the noise reduction feature data can be determined according to the concentration of the main signal-to-noise ratio information in the statistical results. Correspondingly, the specific number of bands corresponding to the spectral band dimensionality reduction feature data can also be determined according to the concentration of the main spectral information in the statistical results. At the same time, it should also be understood that the bands with concentrated main spectral information can be selected. In this way, the obtained noise reduction feature data and spectral band dimensionality reduction feature data can better reflect the ground reflection data, and compared with the ground reflection data, the number of bands they contain is less. That is, in the method of the present disclosure, the obtained first type of feature data can contain fewer bands and more than 90% of the information volume, so that the data can be truly denoised and dimensionally reduced.

[0064] In some embodiments, performing feature extraction processing on the above-mentioned grid data according to the second feature extraction method to obtain the second type of feature data includes: Determine the key elements; Determine the key element data and the ratio data of the key elements from the above grid data, and determine the total channel data of the above grid data; Among them, the above-mentioned second type of characteristic data includes key element data, ratio data of key elements, and total channel data.

[0065] As an example, the key elements may include uranium (U), thorium (Th), and potassium (K). The ratio data of the key elements may include potassium / thorium (K / Th) and uranium / thorium (U / Th).

[0066] Of course, in actual application scenarios, what specific key elements include can be set according to the actual situation, and the specific types of key elements are not limited here.

[0067] It should be understood that uranium and thorium are two main radioactive elements, which generate energy through radioactive decay. Potassium is a common element, widely present in rocks, and can also generate energy through radioactive decay. The total channel (Tc) is the end point of a decay chain, and its content can reflect the degree of radioactive decay. Potassium / thorium and uranium / thorium are the ratios of the contents of two elements, which can provide information about the distribution and combination of radioactive elements in geological bodies.

[0068] By analyzing these parameters, the geological structure, rock type, distribution and combination of radioactive elements in this area can be understood, so as to evaluate the mineralization potential. If these parameters show characteristics conducive to uranium mineralization, then it can be considered that this area has good uranium mineralization potential. On the contrary, if the parameter values are low or the distribution is uneven, it may be unfavorable for uranium mineralization.

[0069] In some embodiments, using the target fusion data to determine the altered rock zone information of the above-mentioned area to be detected, including: Perform spectral analysis on the ground reflection data to obtain the mineral species identification result; Perform density segmentation on the second type of characteristic data to obtain the uranium high-field data; Based on the target fusion data, the mineral species identification result, and the uranium high-field data, determine the altered rock zone information of the area to be detected.

[0070] As an example, the altered rock zone information can indicate the hematitization alkali metasomatized altered rock zone related to uranium ore prospecting. When determining the altered rock zone information of the area to be detected, the target fusion data, the mineral species identification result, and the uranium high-field data are combined at the same time. In this way, the determined altered rock zone information of the area to be detected can be more accurate.

[0071] For ease of understanding, it can be combined with Figure 2 for illustration. Figure 2It can be understood as a schematic diagram of the process for identifying altered rock zones provided by a possible implementation manner of the present disclosure. As Figure 2 shown, after obtaining hyperspectral data and gamma energy spectrum data, the hyperspectral data and gamma energy spectrum data can be analyzed for characteristic information. Then, the hyperspectral data and gamma energy spectrum data are respectively preprocessed and spatially registered. Next, feature extraction can be performed on the hyperspectral data and feature extraction can be performed on the gamma energy spectrum data, and then the features are combined and fused into a feature image. The fused feature image can be used for lithology classification, and the classification accuracy corresponding to each feature image can be determined according to training samples and verification samples, so as to determine the lithology classification map with the highest classification accuracy. Then, the altered rock zone information can be determined by combining the mineral identification results and the uranium high-field analysis results.

[0072] To better understand the idea of the present disclosure, the specific process of a possible implementation manner of the present disclosure is further described in detail below. Specifically: Taking the characteristics of Gf-5 hyperspectral data and gamma energy spectrum data as an example, multi-source feature extraction is performed on the two types of data respectively. The extracted spectral and radioactive information is combined in a tandem (stack) form and normalized to obtain a feature combination, realizing feature-level fusion. Based on the support vector machine, lithology classification in the study area is carried out. By comprehensively analyzing lithology mapping, mineral identification, and radioactive characteristic high-fields, a hematitization alkali metasomatism altered rock zone related to uranium ore prospecting is identified. The specific steps are as follows: Step 1 Feature mining of data information Hyperspectral data has rich spectral information of ground objects and is significant for the identification and classification of minerals and lithology components and the division of altered zones. Airborne gamma energy spectrum measurement is an important nuclear geophysical method, which reflects the types, contents, and distributions of radioactive elements uranium (U), thorium (Th), potassium (K), and other radioactive nuclides in different rocks and ground objects. It has high measurement efficiency, low cost, and is not restricted by factors such as terrain, and has good effects in distinguishing lithology and potassium alteration information. Hyperspectral images and gamma energy spectrum images have unique advantages and can achieve the complementary advantages of the two types of data (Table 1).

[0073] Table 1 Comparison table of information characteristics and geoscience application characteristics of hyperspectral images and gamma energy spectrum images

[0074] Step 2 Data sources and preprocessing The data sources can include Gf-5 hyperspectral data and airborne gamma energy spectrum data.

[0075] Step 2.1 Acquisition and preprocessing of hyperspectral data For example, the imaging time of Gf-5 hyperspectral image is a certain year, month and day, scene number XXXXX, data level is L1, and the image in the area is clear without cloud and snow interference. The main technical indicators of the image sensor are as follows: spatial resolution 30m, of which the visible light-near infrared spectrum range is 390~1029nm, spectral resolution 5nm, and the number of bands is 150; the short-wave infrared spectrum range is 1004~2500nm, spectral resolution 10nm, and the number of bands is 180. The preprocessing steps of Gf-5 hyperspectral data band selection are as follows: through band-by-band inspection, bad bands and water vapor strong absorption bands are eliminated, and then the eliminated data are radiated, de-striped, Flaash atmospheric corrected, and orthorectified based on ENVI5.3 software. After data preprocessing, the ground reflectivity of the ground object is obtained.

[0076] Step 2.2: Preprocessing of airborne gamma spectroscopy data The airborne gamma spectrum data is the radioactivity data of the area to be tested collected by the Nuclear Industry Aerial Survey and Remote Sensing Center from XXXX to XXXX, including the total airborne gamma radioactivity count and the data of three natural radioactive elements (U, Th, K). The grid data is converted into raster data based on ArcGIS software.

[0077] Step 2.3: Data source spatial registration According to the results of steps 2.1 and 2.2 above, the aerial gamma spectral data are spatially interpolated based on the spatial resolution (30m) and projection parameters of the Gf-5 hyperspectral data to generate raster data with the same spatial size (30m). ArcGIS software is used to convert the coordinate system of the spatially interpolated gamma spectral data to the projection coordinate system UTM_Zone_48N, and the geographic coordinate system is GCS_WGS_1984.

[0078] Step 3: Feature fusion of hyperspectral and aerial gamma spectroscopy data Step 3.1 Dimensionality reduction and spectral feature extraction of Gf-5 hyperspectral data According to the result of step 2.1 above, the Gf-5 hyperspectral data has 283 effective bands after preprocessing. There are many bands and a large amount of data, which has the "dimensionality curse", resulting in a decrease in interpretation accuracy. In this disclosure, the minimum noise separation method (MNF) and the principal component analysis method (PCA) feature extraction algorithm can be used to reduce the noise and dimension of the spectral features, and generate noise reduction feature information Fm (15 bands) and spectral band dimension reduction feature information Fp (3 bands) respectively.

[0079] Step 3.2 Feature extraction of airborne gamma spectroscopy data The airborne gamma energy spectrum data collects the radioactive values of uranium (U), thorium (Th), and potassium (K). By analyzing the geochemical characteristics of these three elements and summarizing the laws of migration and enrichment of elements during geological processes, uranium (U) and potassium (K) are prone to migration during geological processes, while the migration of thorium (Th) is relatively low. The three elements and their ratios each have advantages in rock masses, structures, and alterations. In this disclosure, six parameters, namely uranium (U), thorium (Th), potassium (K), total channel (Tc), potassium / thorium (K / Th), and uranium / thorium (U / Th), are selected as the favorable energy spectrum parameters for uranium mineralization in this area.

[0080] Step 3.3 Multi-feature combination The hyperspectral noise reduction features Fm (15 bands), spectral band dimensionality reduction features Fp (3 bands), and airborne gamma energy spectrum data features (U, Th, K, Tc, K / Th, U / Th) obtained in the above steps 3.1 and 3.2 are combined in a series (stack) form for normalization operations (Table 2) to obtain seven multi-feature combinations (Table 3).

[0081] Table 2 Multi-feature extraction of hyperspectral remote sensing and airborne gamma energy spectrum information

[0082] Table 3 Multi-feature combinations

[0083] Step 3.4 Multi-feature fusion Based on the principal component analysis method (PCA), the seven multi-feature combinations obtained in the above step 3.3 are subjected to feature fusion to obtain seven most representative fusion feature characterizations (F1, F2, F3, F4, F5, F6, F7), eliminating redundant information between features and greatly reducing the data dimensionality.

[0084] Step 4 Lithology classification Step 4.1 Collect training samples and validation samples; Taking the preprocessed Gf-5 hyperspectral image in step 2.1 as the base map, referring to the 1:100,000 uranium ore geological map of the study area, training samples and validation samples of various geological bodies in the study area are collected based on the ROI Tool dialog box of ENVI5.3 software. The samples mainly include Quaternary (Q), Neogene Middle Miocene red bed deposits (N1), Upper Group limestone and phyllite of the Late Proterozoic Hanmushan Group (Pt3hm 2 ), Lower Group phyllite, limestone, and volcanic rocks of the Late Proterozoic Hanmushan Group (Pt3hm 1 ), Middle Proterozoic Dunzigou Group Middle Subgroup siliceous banded limestone (Pt2dn 2 ), Lower Subgroup metamorphic conglomerate of the Middle Proterozoic Dunzigou Group (Pt2dn 1), the two-mica quartz schist, biotite quartz schist, and marble of the Tamazigou Formation in the Longshoushan Group of the Lower Proterozoic (Pt1 2 t), the banded migmatite, serpentine marble, schist, and gneiss of the Baijiazuizi Formation in the Longshoushan Group of the Lower Proterozoic (Pt1 1 b), the late Caledonian alkaline complex (E3 3- 1 b), the late Caledonian granite (J l dw_hgy), the Caledonian diorite (J l d_scy). A total of 11,302 training samples and 7,335 validation samples (unit: pixel) were selected from 11 lithological categories in the study area. The specific number of samples for each category is shown in Table 4.

[0085] Table 4 Selection of Lithological Classification Samples

[0086] Step 4.2 Lithological Classification by Support Vector Machine (SVM) For the training sample feature fusion image obtained in Step 4.1 above, based on the support vector machine model (SVM), the preferred kernel function is the radial basis function (RBF). The penalty coefficient C is set to 100, and Gamma is set to 0.026. The seven feature combination images (F1 (MNF + PCA), F2 (MNF + PCA + U), F3 (MNF + PCA + Th), F4 (MNF + PCA + K), F5 (MNF + PCA + Tc), F6 (MNF + PCA + U / Th), F7 (MNF + PCA + K / Th)) are jointly used for classification to obtain the classification results.

[0087] Step 4.3 Comparison of Lithological Classification Accuracy Using the validation samples obtained in Step 4.1, based on the 1:100,000 uranium ore geological map of the study area, the overall accuracy and Kappa coefficient of the classification are statistically analyzed through the error confusion matrix (Table 5) to further quantitatively compare the classification accuracies of different feature combinations. It is concluded that the MNF + PCA + K / Th optimal combination has the best classification effect; compared with the original regional geological map, the fusion result more clearly shows the boundaries of geological bodies, with a more distinct hierarchy; it enriches the information of the original geological map, effectively revises the boundaries of rock masses, and generally realizes a more refined interpretation of geological bodies.

[0088] Table 5 Comparison of Lithological Classification Accuracies of Different Feature Combinations

[0089] Determine the optimal combination and conduct a comprehensive analysis of the lithological mapping corresponding to the optimal combination.

[0090] Step 5 Identification of Alteration Rock Zones Step 5.1 Mineral Identification of Gf-5 Hyperspectral Data Using the preprocessed GF-5 hyperspectral data in Step 2.2, select the mineral spectra in the USGS spectral library as the reference endmember spectra based on the ENVI 5.3 software. Perform spectral analysis on the reference endmember and the GF-5 image spectra using the spectral feature matching algorithm, and finally identify 10 mineral species.

[0091] Step 5.2 Uranium high-field extraction from airborne gamma-ray spectrometry data Using the uranium content characteristics of the airborne gamma-ray spectrometry data extracted in the above Step 3.2, adopt the density slicing method to extract the uranium high-field.

[0092] Step 5.3 Identification of hematitized alkali metasomatized alteration rock zone Using the lithologic mapping, mineral identification results, and uranium high-field obtained in the above Steps 4.3, 5.1, and 5.2 for comprehensive analysis, it can be found that the area to be measured may include a hematitized alkali metasomatized alteration rock zone related to uranium ore prospecting. The hematitization in this zone is strongly developed, and at the same time, the radioactive value is high. According to geological data and field geological surveys, etc., the field spectral measurement curve of the hematitized alkali metasomatized rock in this zone also obviously has two spectral absorption bands similar to hematitization.

[0093] It can be seen that through the above method, the alteration rock zone can be identified more accurately.

[0094] Based on the same inventive concept, the present disclosure also provides an alteration rock zone identification device. Please refer to Figure 3 The alteration rock zone identification device 300 may include an acquisition unit 310, a preprocessing unit 320, a feature extraction unit 330, a fusion unit 340, and a determination unit 350.

[0095] The acquisition unit 310 is configured to acquire hyperspectral data and gamma-ray spectrometry data of the area to be detected; The preprocessing unit 320 is configured to preprocess the hyperspectral data to obtain the ground reflection data of the ground objects, and preprocess the above gamma-ray spectrometry data to obtain grid data; The feature extraction unit 330 is configured to perform feature extraction processing on the above ground reflection data according to the first feature extraction method to obtain the first type of feature data; and perform feature extraction processing on the above grid data according to the second feature extraction method to obtain the second type of feature data; wherein, the first type of feature extraction method includes dimensionality reduction and / or noise reduction; The fusion unit 340 is configured to fuse the first type of feature data and the second type of feature data according to a predefined method to obtain at least one set of fusion data; The determination unit 350 is configured to determine the target fusion data from at least one set of fusion data, and use the above target fusion data to determine the alteration rock zone information of the area to be detected.

[0096] In some embodiments, the determining unit 350 is further specifically configured to: Determine the lithology classification accuracy corresponding to each group of fusion data in at least one group of fusion data; Determine target fusion data from the at least one group of fusion data according to the lithology classification accuracy corresponding to each group of fusion data.

[0097] In some embodiments, the determining unit 350 is further specifically configured to: Obtain training rock type samples and verification rock type samples from a pre-built database; Perform classification learning on a pre-selected support vector machine using the training rock type samples, and perform lithology classification on the at least one group of fusion data using the support vector machine after learning; Verify the lithology classification results corresponding to each group of fusion data using the verification rock type samples, and determine the lithology classification accuracy corresponding to each group of fusion data.

[0098] In some embodiments, the alteration rock zone identification device 300 is further specifically configured to: after preprocessing the hyperspectral data to obtain the ground reflection data of the ground object, and preprocessing the gamma energy spectrum data to obtain grid data, Perform interpolation registration processing on the ground reflection data and the grid data so that the spatial sizes of the ground reflection data and the grid data are the same.

[0099] In some embodiments, the feature extraction unit 330 is further specifically configured to: Perform minimum noise separation processing on the ground reflection data to obtain noise reduction feature data; And perform principal component analysis processing on the ground reflection data to obtain spectral band dimensionality reduction feature data; Wherein, the first type of feature data includes noise reduction feature data and spectral band dimensionality reduction feature data.

[0100] In some embodiments, the feature extraction unit 330 is further specifically configured to: Determine key elements; Determine key element data, ratio data of key elements, and total channel data of the grid data from the grid data, and determine the total channel data of the grid data; Wherein, the second type of feature data includes key element data, ratio data of key elements, and total channel data.

[0101] In some embodiments, the determining unit 350 is further specifically configured to: Perform spectral analysis on the ground reflection data to obtain a mineral species identification result; Perform density segmentation on the above-mentioned second type of feature data to obtain uranium high-field data; Based on the above-mentioned target fusion data, mineral species identification results, and the above-mentioned uranium high-field data, determine the altered rock zone information of the area to be detected.

[0102] Regarding the device in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment related to the method, and will not be elaborated here.

[0103] Figure 4 is a block diagram of an electronic device 400 shown according to an exemplary embodiment. As Figure 4 shown, the electronic device 400 may include: a processor 401, a memory 402. The electronic device 400 may further include one or more of a multimedia component 403, an input / output (I / O) interface 404, and a communication component 405.

[0104] Among them, the processor 401 is used to control the overall operation of the electronic device 400 to complete all or part of the steps in the above-mentioned altered rock zone identification method. The memory 402 is used to store various types of data to support the operation of the electronic device 400. These data may include, for example, instructions for any application or method operating on the electronic device 400, as well as application-related data, such as contact data, received and sent messages, pictures, audio, video, and so on. The memory 402 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The multimedia component 403 may include a screen and an audio component. Among them, the screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 402 or sent through the communication component 405. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 404 provides an interface between the processor 401 and other interface modules, and the above-mentioned other interface modules may be a keyboard, a mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 405 is used for wired or wireless communication between the electronic device 400 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, near field communication (NFC), 2G, 3G, 4G, NB-IoT, eMTC, or other 5G, etc., or a combination of one or more of them, is not limited here. Therefore, the corresponding communication component 405 may include: a Wi-Fi module, a Bluetooth module, an NFC module, and so on.

[0105] In one exemplary embodiment, the electronic device 400 may be implemented by one or more application specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components, and is used to execute the above-mentioned altered rock zone identification method.

[0106] In another exemplary embodiment, a computer-readable storage medium including program instructions is further provided. When the program instructions are executed by a processor, the steps of the above-mentioned altered rock zone identification method are implemented. For example, the computer-readable storage medium may be the above-mentioned memory 402 including program instructions, and the above-mentioned program instructions may be executed by the processor 401 of the electronic device 400 to complete the above-mentioned altered rock zone identification method.

[0107] Figure 5 is a block diagram of an electronic device 500 shown according to an exemplary embodiment. For example, the electronic device 500 may be provided as a server. Referring to Figure 5 , the electronic device 500 includes a processor 522, the number of which may be one or more, and a memory 532 for storing computer programs executable by the processor 522. The computer programs stored in the memory 532 may include one or more modules each corresponding to a set of instructions. In addition, the processor 522 may be configured to execute the computer program to execute the above-mentioned altered rock zone identification method.

[0108] In addition, the electronic device 500 may further include a power supply component 526 and a communication component 550. The power supply component 526 may be configured to perform power management of the electronic device 500, and the communication component 550 may be configured to implement communication of the electronic device 500, for example, wired or wireless communication. In addition, the electronic device 500 may further include an input / output (I / O) interface 558. The electronic device 500 may operate based on an operating system stored in the memory 532.

[0109] In another exemplary embodiment, there is also provided a computer-readable storage medium including program instructions which, when executed by a processor, implement the steps of the above-described altered rock zone identification method. For example, the non-transitory computer-readable storage medium may be the above-described memory 532 including program instructions, and the above program instructions may be executed by the processor 522 of the electronic device 500 to complete the above-described altered rock zone identification method.

[0110] In another exemplary embodiment, there is also provided a computer program product, which includes a computer program executable by a programmable device, and the computer program has a code portion for executing the above-described altered rock zone identification method when executed by the programmable device.

[0111] The preferred embodiments of the present disclosure have been described in detail above in conjunction with the accompanying drawings. However, the present disclosure is not limited to the specific details in the above embodiments. Within the scope of the technical concept of the present disclosure, various simple modifications can be made to the technical solutions of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.

[0112] In addition, it should be noted that, in the above specific embodiments, the various specific technical features described can be combined in any suitable manner without conflict. The present disclosure will not separately describe various possible combination manners.

[0113] Furthermore, any combination can be made between various different embodiments of the present disclosure as long as it does not violate the idea of the present disclosure, and it should also be regarded as the content disclosed by the present disclosure.

Claims

1. A method for identifying an altered rock zone, characterized in that, Including: Obtaining hyperspectral data and gamma energy spectrum data of the area to be detected; Preprocessing the hyperspectral data to obtain ground reflection data of the ground objects, and preprocessing the gamma energy spectrum data to obtain grid data; Performing feature extraction processing on the ground reflection data according to the first feature extraction method to obtain the first type of feature data; and performing feature extraction processing on the grid data according to the second feature extraction method to obtain the second type of feature data; wherein, the first type of feature extraction method includes dimensionality reduction and / or noise reduction; Fusing the first type of feature data and the second type of feature data according to a predefined method to obtain at least one set of fused data; Determining target fused data from at least one set of fused data, and using the target fused data to determine the altered rock zone information of the area to be detected.

2. The method according to claim 1, characterized in that, The determining target fused data from at least one set of fused data includes: Determining the lithology classification accuracy corresponding to each set of fused data in at least one set of fused data; Determining target fused data from the at least one set of fused data according to the lithology classification accuracy corresponding to each set of fused data.

3. The method according to claim 2, characterized in that, The method further includes: Obtaining training rock type samples and verification rock type samples from a pre-built database; Performing classification learning on a pre-selected support vector machine using the training rock type samples, and performing lithology classification on the at least one set of fused data using the learned support vector machine; Verifying the lithology classification results corresponding to each set of fused data using the verification rock type samples to determine the lithology classification accuracy corresponding to each set of fused data.

4. The method according to claim 1, characterized in that, After preprocessing the hyperspectral data to obtain ground reflection data of the ground objects and preprocessing the gamma energy spectrum data to obtain grid data, the method further includes: Performing interpolation registration processing on the ground reflection data and the grid data so that the spatial sizes of the ground reflection data and the grid data are the same.

5. The method according to claim 1, characterized in that, The performing feature extraction processing on the ground reflection data according to the first feature extraction method to obtain the first type of feature data includes: Performing minimum noise separation processing on the ground reflection data to obtain noise reduction feature data; And performing principal component analysis processing on the ground reflection data to obtain spectral band dimensionality reduction feature data; Wherein, the first type of feature data includes noise reduction feature data and spectral band dimensionality reduction feature data.

6. The method according to claim 1, characterized in that, Performing feature extraction processing on the grid data according to the second feature extraction method to obtain the second type of feature data includes: Determining key elements; Determining key element data and ratio data of the key elements from the grid data, and determining the total channel data of the grid data; Wherein, the second type of feature data includes key element data, ratio data of the key elements and total channel data.

7. The method according to claim 1, characterized in that, The using the target fused data to determine the altered rock zone information of the area to be detected includes: Performing spectral analysis on the ground reflection data to obtain a mineral species identification result; Performing density segmentation on the second type of feature data to obtain uranium high field data; Based on the target fused data, the mineral species identification result and the uranium high field data, determining the altered rock zone information of the area to be detected.

8. An apparatus for identifying an altered rock zone, characterized in that, Including: An acquisition unit for acquiring hyperspectral data and gamma energy spectrum data of a region to be detected; A preprocessing unit for preprocessing the hyperspectral data to obtain ground reflection data of ground objects, and preprocessing the gamma energy spectrum data to obtain grid data; A feature extraction unit for performing feature extraction processing on the ground reflection data according to a first feature extraction method to obtain first-class feature data; and performing feature extraction processing on the grid data according to a second feature extraction method to obtain second-class feature data; wherein the first feature extraction method includes dimensionality reduction and / or noise reduction; A fusion unit for fusing the first-class feature data and the second-class feature data according to a predefined method to obtain at least one set of fusion data; A determination unit for determining target fusion data from at least one set of fusion data, and using the target fusion data to determine the altered rock zone information of the region to be detected.

9. A non - transitory computer - readable storage medium, on which a computer program is stored, characterized in that, When the program is executed by a processor, it implements the steps of the method according to any one of claims 1-7.

10. An electronic device, characterized in that, Comprising: A memory having a computer program stored thereon; A processor for executing the computer program in the memory to implement the steps of the method according to any one of claims 1-7.

Images