Alterated mineral information extraction method and device based on remote sensing data and electronic equipment

Abstract

The invention provides an altered mineral information extraction method and device based on remote sensing data and electronic equipment, and relates to the technical field of remote sensing, and the method comprises the steps: obtaining multispectral remote sensing image data, and preprocessing the multispectral remote sensing image data to obtain preprocessed remote sensing data; inputting the preprocessed remote sensing data into a preset spectral index model to obtain an altered mineral spectral index; screening each pixel of the altered mineral spectral index according to an altered mineral threshold value to obtain an altered mineral pure pixel; obtaining an altered mineral end member set by identifying the spectral characteristics of the pure pixels of the altered mineral; and performing linear mixing and unmixing on the preprocessed remote sensing data and the altered mineral end member set through mixed tuning matched filtering to obtain altered mineral distribution information. According to the method, the altered mineral information extraction precision is improved.

Description

Technical Field

[0001] The present invention relates to the field of remote sensing technology, and in particular, to a method, device and electronic device for extracting altered mineral information based on remote sensing data. Background Art

[0002] Satellite images are widely used for mapping geological and environmental features at different scales. Currently, remote sensing data obtained by satellites is usually used for mineral exploration, saving time and effort compared with traditional manual ground surveys. The technology for extracting altered mineral information is of great significance in mineral exploration and environmental monitoring. Alteration refers to the changes in mineralogical, chemical and physical properties of rocks under the action of external environments. However, due to the similar reflection characteristics of the spectra of altered minerals in remote sensing images, the relatively low spectral resolution may lead to spectral overlap between the two, resulting in blurred features or misidentification during the process of extracting altered mineral information from remote sensing data. Summary of the Invention

[0003] The problem solved by the present invention is how to improve the accuracy of extracting altered mineral information.

[0004] To solve the above problems, the present invention provides a method, device and electronic device for extracting altered mineral information based on remote sensing data.

[0005] In a first aspect, the present invention provides a method for extracting altered mineral information based on remote sensing data, including: Obtaining multi-spectral remote sensing image data, and preprocessing the multi-spectral remote sensing image data to obtain preprocessed remote sensing data; Inputting the preprocessed remote sensing data into a preset spectral index model to obtain an altered mineral spectral index; Screening each pixel of the altered mineral spectral index according to an altered mineral threshold to obtain altered mineral pure pixels; Obtaining an altered mineral endmember set by identifying the spectral characteristics of the altered mineral pure pixels; Performing linear unmixing on the preprocessed remote sensing data and the altered mineral endmember set through hybrid tuned matching filtering to obtain altered mineral distribution information.

[0006] Optionally, the altered mineral spectral index includes a ferrous ferric spectral index, an iron oxide spectral index, an aluminum hydroxyl spectral index, a magnesium hydroxyl spectral index and an iron hydroxyl spectral index. The step of inputting the preprocessed remote sensing data into a preset spectral index model to obtain an altered mineral spectral index includes: Obtaining data of each band through the preprocessed remote sensing data; Input the data of each band into a preset spectral index model, and obtain the ferrous-ferric iron spectral index, the iron oxide spectral index, the aluminum hydroxyl spectral index, the magnesium hydroxyl spectral index, and the iron hydroxyl spectral index through band combination; Among them, the ferrous-ferric iron spectral index is: , where, is the ferrous-ferric iron spectral index, B21 is the band data in the wavelength range of 1.98 - 2.02μm, B13 is the band data in the wavelength range of 0.80 - 0.83μm, B14 is the band data in the wavelength range of 0.84 - 0.86μm, and B15 is the band data in the wavelength range of 0.94 - 0.98μm; Among them, the iron oxide spectral index is: , where, is the iron oxide spectral index, B7 is the band data in the wavelength range of 0.62 - 0.65μm, and B3 is the band data in the wavelength range of 0.48 - 0.5μm; Among them, the aluminum hydroxyl spectral index is: , where, is the aluminum hydroxyl spectral index, B23 is the band data in the wavelength range of 1.8–1.9μm, and B18 is the band data in the wavelength range of 1.2–1.3μm; Among them, the magnesium hydroxyl spectral index is: , where, is the magnesium hydroxyl spectral index, B25 is the band data in the wavelength range of 2.1–2.2μm, and B26 is the band data in the wavelength range of 2.3–2.4μm; Among them, the iron hydroxyl spectral index is: , where, is the iron hydroxyl spectral index, and B19 is the band data in the wavelength range of 1.58–1.65μm.

[0007] Optionally, screening each pixel of the altered mineral spectral index according to the altered mineral threshold to obtain the pure pixels of the altered mineral includes: Performing adaptive threshold segmentation on the altered mineral spectral index by the maximum inter-class variance algorithm to obtain the altered mineral threshold; An altered mineral mask model is obtained according to the altered mineral threshold; Among them, the altered mineral mask model is: , where MASK is the altered mineral mask model, is the altered mineral spectral index, is the altered mineral threshold, Otherswise is the situation not covered by the above conditions, and i = 1, 2, 3, 4, 5; Each pixel of the altered mineral spectral index is input into the altered mineral mask model for screening to obtain the pure pixels of the altered minerals.

[0008] Optionally, obtaining the altered mineral endmember set by identifying the spectral characteristics of the pure pixels of the altered minerals includes: Performing noise separation processing on the pure pixels of the altered minerals to obtain a pure pixel noise separation result; Among them, the pure pixel noise separation result is: , where R is the pure pixel noise separation result, S is the pure pixel of the altered minerals, and P is the principal component analysis projection matrix; Calculating the included angle between each pixel according to the pure pixel noise separation result to obtain spectral similarity; Among them, the spectral similarity is: , where is the spectral similarity, and are the spectral reflectance vectors of two pixels respectively; The altered mineral endmember set is obtained through the spectral similarity.

[0009] Optionally, performing linear unmixing on the preprocessed remote sensing data and the altered mineral endmember set through hybrid tuned matching filtering to obtain altered mineral distribution information includes: Performing linear unmixing on the preprocessed remote sensing data and the altered mineral endmember set through hybrid tuned matching filtering to obtain endmember abundances; Among them, the endmember abundances are: , where Y is the preprocessed remote sensing data, is the spectrum of the k-th endmember in the altered mineral endmember set, is the endmember abundance corresponding to the k-th endmember, and p is the number of endmembers in the altered mineral endmember set; Obtain the altered mineral distribution information through the terminal abundance.

[0010] Optionally, the preprocessing of the multi-spectral remote sensing image data to obtain preprocessed remote sensing data includes: Convert the digital value of the multi-spectral remote sensing image data into radiance through radiometric calibration to obtain radiance image data; Among them, the radiance image data is: , Among them, is the radiance in the radiance image data, is the calibration gain coefficient, is the calibration offset, and DN is the digital value of the multi-spectral remote sensing image data; Perform stacking processing on the radiance image data to obtain the preprocessed remote sensing data.

[0011] Optionally, the stacking processing of the radiance image data to obtain the preprocessed remote sensing data includes: Perform stacking processing on the radiance image data and convert the stacked radiance image data into BIL format remote sensing data; Obtain the preprocessed remote sensing data by normalizing the BIL format remote sensing data.

[0012] In a second aspect, the present invention provides an altered mineral information extraction device based on remote sensing data, including: a preprocessed remote sensing data acquisition module for acquiring multi-spectral remote sensing image data and preprocessing the multi-spectral remote sensing image data to obtain preprocessed remote sensing data; An altered mineral spectral index acquisition module for inputting the preprocessed remote sensing data into a preset spectral index model to obtain an altered mineral spectral index; An altered mineral pure pixel extraction module for screening each pixel of the altered mineral spectral index according to an altered mineral threshold to obtain an altered mineral pure pixel; An altered mineral endmember set acquisition module for obtaining an altered mineral endmember set by identifying the spectral characteristics of the altered mineral pure pixels; An altered mineral distribution information acquisition module for performing linear unmixing on the preprocessed remote sensing data and the altered mineral endmember set through hybrid tuned matching filtering to obtain altered mineral distribution information.

[0013] In a third aspect, the present invention provides an electronic device, including a memory and a processor; The memory is used to store a computer program; The processor is configured to implement the method for extracting altered mineral information based on remote sensing data as described in the first aspect when executing the computer program.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method for extracting altered mineral information based on remote sensing data as described in the first aspect is implemented.

[0015] The beneficial effects of the method, apparatus, and electronic device for extracting altered mineral information based on remote sensing data of the present invention are as follows: By acquiring multispectral remote sensing image data, more bands can be obtained, effectively improving the spectral resolution. Preprocessing the multispectral remote sensing image data facilitates subsequent data processing. By preprocessing the remote sensing data to obtain the altered mineral spectral index, the spatial distribution of altered minerals in the region can be effectively interpreted. Each pixel of the altered mineral spectral index is screened according to the altered mineral threshold to obtain the pure pixels of the altered minerals, screening out the pixels with the most prominent spectral characteristics and reducing the influence of mixed pixels. By further identifying and extracting the spectral characteristics that best represent different mineral categories, an endmember set of the altered minerals is obtained. Linear unmixing of the mineral components is performed through hybrid tuned matching filtering, which is highly sensitive to trace mineral components and avoids information omission. Finally, the distribution information of the altered minerals is obtained, ensuring clear feature recognition and improving the accuracy of extracting altered mineral information. Description of the Drawings

[0016] Figure 1 It is a schematic flowchart of a method for extracting altered mineral information based on remote sensing data according to an embodiment of the present invention; Figure 2 It is a schematic structural diagram of an apparatus for extracting altered mineral information based on remote sensing data according to an embodiment of the present invention; Figure 3 It is a schematic structural diagram of an electronic device according to an embodiment of the present invention. Detailed Embodiments

[0017] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following detailed description of the specific embodiments of the present invention is provided in conjunction with the accompanying drawings. Although some embodiments of the present invention are shown in the drawings, it should be understood that the present invention can be implemented in various forms and should not be construed as limited to the embodiments described herein. On the contrary, these embodiments are provided to more thoroughly and completely understand the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not used to limit the protection scope of the present invention.

[0018] It should be understood that the various steps recited in the method embodiments of the present invention may be executed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this regard.

[0019] As used herein, the term "comprising" and its variations are open-ended, i.e., "including but not limited to"; the term "based on" means "at least partially based on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiment". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts such as "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order of the functions executed by these devices, modules or units or their interdependent relationships.

[0020] It should be noted that the modification of "one" and "plural" mentioned in the present invention is illustrative rather than restrictive. Those skilled in the art should understand that unless otherwise clearly specified in the context, it should be understood as "one or more".

[0021] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only for illustrative purposes and are not used to limit the scope of these messages or information.

[0022] In the related art, remote sensing image data covers the visible light (VIS), visible and near-infrared radiometer (VNIR), and short-wave infrared radiometer (SWIR) regions in terms of spectrum. There are a total of 26 spectral bands, which can create high-resolution (15m) multispectral image data and describe the information of the Earth's surface in detail. Among them, altered minerals in geology refer to secondary minerals formed by the change of original minerals due to the change of physical and chemical conditions. Altered minerals include trivalent iron oxide minerals, iron oxide minerals, aluminum hydroxyl minerals, magnesium hydroxyl minerals, and iron hydroxyl minerals. Trivalent iron oxide minerals are mainly composed of trivalent iron (Fe³⁺), and common ones include: hematite (Hematite, Fe2O3), goethite (Goethite, FeO(OH)), and lepidocrocite (γ-FeO(OH)). Iron oxide minerals are a type of minerals containing iron elements and existing in the form of oxides, mainly including some other iron oxide minerals, including magnetite. Aluminum hydroxyl minerals contain aluminum and hydroxyl (-OH), and common ones are: kaolinite (Kaolinite, Al2Si2O5(OH)4), gibbsite (Gibbsite, Al(OH)3), and boehmite (γ-AlO(OH)). Magnesium hydroxyl minerals contain magnesium and hydroxyl (-OH), including: talc (Talc, H2Mg3(SiO3)4 or Mg3Si4O 10 (OH)2), and brucite (Brucite, Mg(OH)2). Iron hydroxyl minerals are those minerals containing iron and hydroxyl (-OH), such as ferrihydrite (Ferrihydrite, Fe5HO8·4H2O).

[0023] In view of the problems existing in the above related art, this embodiment provides a method, device, and electronic device for extracting altered mineral information based on remote sensing data.

[0024] As Figure 1 shown, a method for extracting altered mineral information based on remote sensing data provided by an embodiment of the present invention includes: Step 110, obtaining multispectral remote sensing image data and preprocessing the multispectral remote sensing image data to obtain preprocessed remote sensing data.

[0025] Specifically, multispectral remote sensing image data is obtained by using a satellite, including high-resolution visible light, near-infrared, and short-wave infrared band data. This data source has a higher spectral resolution, a wider spectral coverage range, and a more detailed spectral interval, which can maximize the coverage of spectral absorption peak characteristics of altered minerals in the near-infrared, short-wave infrared, etc., so as to effectively reduce the problem of spectral overlap and improve the extraction accuracy of altered minerals.

[0026] Step 120: Input the preprocessed remote sensing data into a preset spectral index model to obtain the spectral index of altered minerals.

[0027] Specifically, the spectral index of altered minerals is a method that can utilize the reflectance characteristics of minerals in specific bands and enhance the information of target minerals by calculating the ratios or combinations of specific bands. The method based on spectral index utilizes the absorption characteristics of minerals in specific bands, calculates the ratios or indices between bands, and enhances the spectral information of target minerals.

[0028] Step 130: Screen each pixel of the spectral index of altered minerals according to the threshold of altered minerals to obtain pure pixels of altered minerals.

[0029] Specifically, by setting a reasonable threshold and applying the set threshold to screen the entire image, retaining those pixels that exceed the threshold as pure pixels of altered minerals, the pixels with the most prominent spectral characteristics can be screened out, reducing the influence of mixed pixels, and thus extracting representative pure pixels.

[0030] Step 140: Obtain the endmember set of altered minerals by identifying the spectral characteristics of the pure pixels of altered minerals.

[0031] Specifically, further extract representative endmember spectra from the pure pixels of altered minerals screened based on spectral index for subsequent spectral unmixing analysis. The core objective of endmember extraction is to identify the spectral characteristics in the data that can best represent different mineral categories and ensure that these endmembers can accurately describe the composition of mixed pixels in the image.

[0032] Step 150: Perform linear unmixing on the preprocessed remote sensing data and the endmember set of altered minerals through mixture tuned matched filtering to obtain the distribution information of altered minerals.

[0033] Specifically, mixture tuned matched filtering (MTMF) is a spectral analysis technique used for remote sensing image processing, mainly used to improve the accuracy of target recognition and reduce false positive results. Based on the previously extracted endmember spectra, perform linear unmixing. Then estimate the abundance of mineral components in each pixel and generate a mineral abundance distribution map.

[0034] In this embodiment, obtaining multi-spectral remote sensing image data can acquire more bands and effectively improve spectral resolution. Preprocessing the multi-spectral remote sensing image data facilitates subsequent data processing. By preprocessing the remote sensing data, the spectral indices of altered minerals are obtained, effectively interpreting the spatial distribution of altered minerals in the area. Each pixel of the spectral indices of altered minerals is screened according to the altered mineral threshold to obtain pure pixels of altered minerals, screening out the pixels with the most prominent spectral features and reducing the influence of mixed pixels. By further identifying and extracting the spectral features that best represent different mineral categories, an endmember set of altered minerals is obtained. Linear unmixing of mineral components is performed through mixed tuned matching filtering, which is highly sensitive to trace mineral components and avoids information omission. Finally, the distribution information of altered minerals is obtained, ensuring clear feature recognition and improving the accuracy of extracting altered mineral information.

[0035] Optionally, the spectral indices of altered minerals include the ferrous to ferric spectral index, the iron oxide spectral index, the aluminum hydroxyl spectral index, the magnesium hydroxyl spectral index, and the iron hydroxyl spectral index. The step of inputting the preprocessed remote sensing data into a preset spectral index model to obtain the spectral indices of altered minerals includes: Obtaining data of each band through the preprocessed remote sensing data; Inputting the data of each band into a preset spectral index model to obtain the ferrous to ferric spectral index, the iron oxide spectral index, the aluminum hydroxyl spectral index, the magnesium hydroxyl spectral index, and the iron hydroxyl spectral index through band combination; Among them, the ferrous to ferric spectral index is: , Among them, is the ferrous to ferric spectral index, B21 is the band data with a wavelength in the range of 1.98 - 2.02 μm, B13 is the band data with a wavelength in the range of 0.80 - 0.83 μm, B14 is the band data with a wavelength in the range of 0.84 - 0.86 μm, and B15 is the band data with a wavelength in the range of 0.94 - 0.98 μm; Among them, the iron oxide spectral index is: , Among them, is the iron oxide spectral index, B7 is the band data with a wavelength in the range of 0.62 - 0.65 μm, and B3 is the band data with a wavelength in the range of 0.48 - 0.5 μm; Among them, the aluminum hydroxyl spectral index is: , Among them, is the aluminum hydroxyl spectral index, B23 is the band data in the wavelength range of 1.8 - 1.9 μm, and B18 is the band data in the wavelength range of 1.2 - 1.3 μm; wherein, the magnesium hydroxyl spectral index is: , wherein, is the magnesium hydroxyl spectral index, B25 is the band data in the wavelength range of 2.1 - 2.2 μm, and B26 is the band data in the wavelength range of 2.3 - 2.4 μm; wherein, the iron hydroxyl spectral index is: , wherein, is the iron hydroxyl spectral index, and B19 is the band data in the wavelength range of 1.58 - 1.65 μm.

[0036] Specifically, the pre - processed remote sensing data includes short - wave infrared bands, near - infrared bands, and visible light bands. B21, B13, B14, B15 (short - wave infrared bands and near - infrared bands): The short - wave infrared band B21 is in the range of 1.98 - 2.02 μm and can capture the strong absorption characteristics of iron minerals (especially trivalent iron oxide minerals) in this band range. The reflectivity of this band to trivalent iron is relatively low, while the reflectivity to ferrous iron is relatively high. The near - infrared bands can effectively reflect the electronic transition characteristics and spectral characteristics of the crystal structure of iron minerals. The three selected near - infrared bands (B13, B14, B15) can comprehensively reflect the spectral differences between ferrous iron and trivalent iron by covering different characteristic peak regions of iron minerals.

[0037] B3, B7 (visible light bands): Iron oxide minerals (such as hematite, pyrite) usually have strong absorption characteristics in the blue light band. The blue light band can capture the spectral characteristics of these minerals. Especially for iron minerals with more exposed surfaces or higher weathering degrees, the blue light band can provide more significant reflection differences. Iron oxide minerals usually show certain reflection characteristics in the red light band, especially in minerals with a higher content of trivalent iron (Fe³⁺), where the reflectivity is higher.

[0038] B23, B18 (short - wave infrared bands and near - infrared bands): The aluminum hydroxyl index product extracts the spectral characteristics of aluminum hydroxyl minerals through the ratio of the B23 band to the B18 band. The B23 band (short - wave infrared band) is in the range of 1.8 - 1.9 μm and has significant absorption characteristics of aluminum hydroxyl minerals, while the B18 band (near - infrared band) covers the reflection characteristics related to aluminum hydroxyl minerals.

[0039] B18, B25, B26 (near-infrared and short-wave infrared bands): The magnesium hydroxyl index product extracts the spectral characteristics of magnesium hydroxyl minerals through the difference ratio between the B18 band and the B25 and B26 bands. The B18 band (near-infrared band) is in the range of 1.2–1.3 μm and has obvious absorption characteristics of magnesium hydroxyl minerals, while the B25 and B26 bands are in the ranges of 2.1–2.3 μm and 2.3–2.4 μm respectively, covering the reflection characteristics related to magnesium hydroxyl minerals.

[0040] B19, B23 (short-wave infrared bands): The iron hydroxyl index product extracts the spectral characteristics of iron hydroxyl minerals through the ratio between the B19 band and the B23 band. The B19 band (short-wave infrared band) is in the range of 1.58–1.65 μm and has significant absorption characteristics of iron hydroxyl minerals, while the B23 band (short-wave infrared band) is in the range of 1.8–1.9 μm, covering another important reflection characteristic related to iron hydroxyl minerals. The ferrous to ferric spectral index, iron oxide spectral index, aluminum hydroxyl spectral index, magnesium hydroxyl spectral index, and iron hydroxyl spectral index correspond to ferric oxide minerals, iron oxide minerals, aluminum hydroxyl minerals, magnesium hydroxyl minerals, and iron hydroxyl minerals respectively.

[0041] In this optional embodiment, the spectral index is calculated by band ratio, which can effectively reduce the interference of other surface factors (such as vegetation, soil, and rocks) on spectral characteristics. By combining the short-wave infrared and near-infrared bands, the spectral signal of altered minerals can be highlighted, thereby reducing the impact of background complexity on the inversion results. The combination of the blue light band and the red light band can provide more comprehensive spectral information of iron oxide. Through the combination of bands, altered minerals can be more effectively distinguished from other ground objects.

[0042] Optionally, screening each pixel of the altered mineral spectral index according to the altered mineral threshold to obtain the pure pixels of the altered minerals includes: Performing adaptive threshold segmentation on the altered mineral spectral index by the maximum inter-class variance algorithm to obtain the altered mineral threshold; Obtaining an altered mineral mask model according to the altered mineral threshold; Among them, the altered mineral mask model is: , where MASK is the altered mineral mask model, is the altered mineral spectral index, is the altered mineral threshold, Otherswise is the situation not covered by the above conditions, and i = 1, 2, 3, 4, 5; Inputting each pixel of the altered mineral spectral index into the altered mineral mask model for screening to obtain the pure pixels of the altered minerals.

[0043] Specifically, in remote sensing alteration mineral identification, the selection of pure pixels is crucial for spectral unmixing results. Since the reflectance differences of different alteration minerals are obvious in specific bands, the corresponding spectral indices can reflect the spatial distribution of minerals. By setting reasonable thresholds, pixels with the most prominent spectral features can be screened out, reducing the influence of mixed pixels, and thus representative pure pixels can be extracted. The alteration mineral threshold is obtained through the Otsu algorithm, which is an adaptive threshold segmentation method based on the maximum inter-class variance criterion. Five different spectral indices are input respectively, the inter-class variance between the foreground and the background is calculated, and the threshold that maximizes the inter-class variance is selected as the optimal segmentation point to obtain the corresponding threshold. The alteration mineral threshold includes ferrous-ferric spectral threshold, iron oxide spectral threshold, aluminum hydroxyl spectral threshold, magnesium hydroxyl spectral threshold, and iron hydroxyl spectral threshold. The spectral indices of each pixel are input into the alteration mineral mask model for screening operations. Only pixels that satisfy the condition that the spectral index of the alteration mineral is greater than or equal to the alteration mineral threshold will be retained as the final pure pixels of the alteration mineral. For example, when the ferrous-ferric spectral index corresponding to a certain pixel is greater than the ferrous-ferric spectral threshold, this pixel is retained as a pure pixel of the alteration mineral.

[0044] In this alternative embodiment, the optimal threshold is automatically determined by the maximum inter-class variance, ensuring that the segmentation result can separate pure pixels as much as possible, improving the recognition accuracy. It simplifies the operation process, reduces the errors caused by human factors, and can also significantly improve the accuracy and efficiency of pure pixel recognition.

[0045] Optionally, obtaining the alteration mineral endmember set by identifying the spectral features of the pure pixels of the alteration mineral includes: Performing noise separation processing on the pure pixels of the alteration mineral to obtain the pure pixel noise separation result; Among them, the pure pixel noise separation result is: , where R is the pure pixel noise separation result, S is the pure pixel of the alteration mineral, and P is the principal component analysis projection matrix; Calculating the angle between each pixel according to the pure pixel noise separation result to obtain the spectral similarity; Among them, the spectral similarity is: , where is the spectral similarity, and are the spectral reflectance vectors of two pixels respectively; Obtaining the alteration mineral endmember set through the spectral similarity.

[0046] Specifically, representative endmember spectra are further extracted from the pure pixels of altered minerals selected based on spectral indices for subsequent spectral unmixing analysis. The core objective of endmember extraction is to identify the spectral features in the data that best represent different mineral classes and ensure that these endmembers can accurately describe the composition of mixed pixels in the image. First, the Minimum Noise Fraction Transform (MNF) can be used to reduce the dimensionality of the data to enhance the signal-to-noise ratio and highlight the main spectral features. MNF separates the noise from the useful information in the pure pixels of altered minerals through Principal Component Analysis (PCA). The principal component analysis projection matrix is obtained through principal component analysis, so that the most important features in the data are retained. MNF first performs a preliminary principal component transformation based on the covariance matrix of the pure pixel data of altered minerals, removes the noise components to obtain the pure pixel noise separation result. Then, further principal component transformation is performed on the data after noise removal to determine the intrinsic dimensionality of the data and reduce redundant information. The Spectral Angle Mapper (SAM) or other methods are applied to identify endmembers for candidate pixels. SAM measures spectral similarity by calculating the angle between spectral vectors. The spectral similarity between the spectral reflectance vectors of every two pixels with the same spectral index in the pure pixel noise separation result is calculated respectively. The spectral similarity is the angle between the spectral vectors. When the angle is small, it indicates that the spectral characteristics of the two are similar, so it is not retained; when the angle is large, it indicates that the spectral characteristics of the two are not similar, so it is retained. The pixels corresponding to each spectral index in the pure pixel noise separation result are screened in turn, and several endmembers with the most significant spectral features and representing different mineral classes are selected according to the similarity measure to form an endmember set , where , , , , represent the endmembers corresponding to ferric iron oxide minerals, iron oxide minerals, aluminum hydroxyl minerals, magnesium hydroxyl minerals, and iron hydroxyl minerals respectively.

[0047] In this optional embodiment, by removing noise interference, the true mineral features can be identified more accurately, and misjudgments caused by noise can be reduced. The method based on spectral similarity to determine the mineral endmember set can enhance the credibility of the results and reduce the repetitive workload.

[0048] Optionally, the linear unmixing of the preprocessed remote sensing data and the altered mineral endmember set through hybrid tuned matching filtering to obtain the altered mineral distribution information includes:[[]] Performing linear unmixing of the preprocessed remote sensing data and the altered mineral endmember set through hybrid tuned matching filtering to obtain the endmember abundances; where the endmember abundances are:[[]] , where Y is the preprocessed remote sensing data, is the spectrum of the k-th endmember in the altered mineral endmember set, is the endmember abundance corresponding to the k-th endmember, and p is the number of endmembers in the altered mineral endmember set; The altered mineral distribution information is obtained through the endmember abundance.

[0049] Specifically, mixed tuned matched filtering (MTMF) is used to unmix image data and estimate the relative abundances of mineral components. MTMF combines a linear mixing model with matched filtering techniques, can unmix mineral components, and is highly sensitive to trace mineral components. First, linear mixing unmixing is performed based on the previously extracted endmember spectra.

[0050] In some more specific embodiments, by inputting the endmember abundances into MATLAB, a mineral abundance distribution map is automatically generated. MATLAB is a commercial mathematical software, an advanced technical computing language and interactive environment for algorithm development, data visualization, data analysis, and numerical computation. The endmember abundances include ferric iron oxide mineral abundance data, iron oxide mineral abundance data, aluminum hydroxyl mineral abundance data, magnesium hydroxyl mineral abundance data, and iron hydroxyl mineral abundance data. The endmember abundances usually exist in the form of a matrix or a raster, where each element represents the abundance value of a pixel. The abundance data of each mineral is stored as a separate matrix variable. For each mineral abundance data, the imagesc function can be used to generate the abundance distribution map. imagesc is a function in MATLAB for displaying data in the form of a matrix or an array, where the value of each element determines the color intensity of the corresponding image pixel. In addition, a color bar can be added to facilitate understanding the relationship between the abundance values and the colors.

[0051] Optionally, the preprocessing of the multispectral remote sensing image data to obtain the preprocessed remote sensing data includes: Converting the digital values of the multispectral remote sensing image data into radiance through radiometric calibration to obtain radiance image data; where the radiance image data is: , where is the radiance in the radiance image data, is the calibration gain coefficient, is the calibration offset, and DN is the digital value of the multispectral remote sensing image data; Performing stacking processing on the radiance image data to obtain the preprocessed remote sensing data.

[0052] Optionally, the step of obtaining the preprocessed remote sensing data by stacking the radiance image data includes: Stacking the radiance image data, and converting the stacked radiance image data into BIL format remote sensing data; Obtaining the preprocessed remote sensing data by normalizing the BIL format remote sensing data.

[0053] Specifically, the DN value (digital number) of the image is converted into radiance through radiometric correction, and the sensor characteristics and atmospheric conditions are considered during the correction process to ensure that the radiance of the image can accurately reflect the surface reflectance. Subsequently, the image is stacked, and the data of different bands are integrated into a multi-band image to form a complete remote sensing data set. Next, the image data is converted into BIL format for subsequent processing and analysis. Among them, BIL (Band Interleaved by Line) is a file format for storing multi-band image data. It is commonly used in the fields of remote sensing and geographic information system (GIS), especially when processing satellite or aerial images. The BIL format is a standard format supported by ENVI software, but it can also be read and written by many other image processing software and programming environments.

[0054] In some more specific embodiments, various gases in the Earth's atmosphere, such as water vapor, carbon dioxide, and aerosols, will selectively absorb, scatter, and reflect electromagnetic waves in the remote sensing image, thus affecting the true radiation information of ground objects. Therefore, in order to accurately reflect the spectral characteristics of ground objects, it is necessary to perform atmospheric correction on the multi-spectral remote sensing image data obtained by satellites. The FLAASH (Fast Line-of-sight Atmospheric Analysis of Spectral Hypercubes) atmospheric correction algorithm in the MODTRAN5 radiative transfer model is adopted. The FLAASH algorithm was jointly developed by Spectral Sciences, Inc. and the U.S. Air Force Research Laboratory, and has high precision and is widely used in the atmospheric correction of hyperspectral data. This algorithm can effectively remove the influence of components such as water vapor and aerosols in the atmosphere, and through radiometric correction of remote sensing image data, provide accurate radiance values for subsequent calculation of surface reflectance.

[0055] In some more specific embodiments, in remote sensing image analysis, the main interferents in the study area include vegetation, mountain shadows, clouds, etc. These interferents may cause errors in mineral information extraction. Therefore, different bands of remote sensing data are used to remove the interferents. First, false color synthesis is performed using bands 1, 3, and 6 of the remote sensing data to construct an RGB image, which is convenient for distinguishing different types of interferents, such as vegetation, mountain shadows, and clouds. Through the false color composite image, the positions of these interferents can be visually identified, and further processing is performed using an algorithm based on pixel classification. Next, a classification method based on thresholds is used to perform masking processing on various interferents. The specific operation is to accurately distinguish the vegetation, cloud, and water body areas by setting different spectral reflectance thresholds and remove them from the image. Finally, the image data after removing the interferents can more accurately reflect the geological characteristics. The reflectance range of the image is normalized, and the reflectance value is scaled between 0 and 1. To extract a specific study area, the image is also subjected to regional subsetting, and only the data within the study area is retained for subsequent analysis. Finally, spectral indices of altered minerals are constructed based on specific bands to identify and distinguish the spatial distributions of these iron oxide and hydroxyl altered minerals.

[0056] As Figure 2 shown, an altered mineral information extraction device based on remote sensing data provided by an embodiment of the present invention includes: A preprocessing remote sensing data acquisition module 10, configured to acquire multispectral remote sensing image data and perform preprocessing on the multispectral remote sensing image data to obtain preprocessed remote sensing data; An altered mineral spectral index acquisition module 20, configured to input the preprocessed remote sensing data into a preset spectral index model to obtain an altered mineral spectral index; An altered mineral pure pixel extraction module 30, configured to screen each pixel of the altered mineral spectral index according to an altered mineral threshold to obtain an altered mineral pure pixel; An altered mineral endmember set acquisition module 40, configured to obtain an altered mineral endmember set by identifying the spectral characteristics of the altered mineral pure pixels; An altered mineral distribution information acquisition module 50, configured to perform linear unmixing on the preprocessed remote sensing data and the altered mineral endmember set through hybrid tuned matching filtering to obtain altered mineral distribution information.

[0057] Optionally, the altered mineral spectral index acquisition module 20 is specifically configured to: obtain data of each band through the preprocessed remote sensing data; Input the data of each band into a preset spectral index model and obtain the ferrous ferric spectral index, the iron oxide spectral index, the aluminum hydroxyl spectral index, the magnesium hydroxyl spectral index, and the iron hydroxyl spectral index through band combination; Among them, the ferrous-ferric spectral index is: , where is the ferrous-ferric spectral index, B21 is the band data in the wavelength range of 1.98 - 2.02 μm, B13 is the band data in the wavelength range of 0.80 - 0.83 μm, B14 is the band data in the wavelength range of 0.84 - 0.86 μm, and B15 is the band data in the wavelength range of 0.94 - 0.98 μm; Among them, the iron oxide spectral index is: , where is the iron oxide spectral index, B7 is the band data in the wavelength range of 0.62 - 0.65 μm, and B3 is the band data in the wavelength range of 0.48 - 0.5 μm; Among them, the aluminum hydroxyl spectral index is: , where is the aluminum hydroxyl spectral index, B23 is the band data in the wavelength range of 1.8–1.9 μm, and B18 is the band data in the wavelength range of 1.2–1.3 μm; Among them, the magnesium hydroxyl spectral index is: , where is the magnesium hydroxyl spectral index, B25 is the band data in the wavelength range of 2.1–2.2 μm, and B26 is the band data in the wavelength range of 2.3–2.4 μm; Among them, the iron hydroxyl spectral index is: , where is the iron hydroxyl spectral index, and B19 is the band data in the wavelength range of 1.58–1.65 μm.

[0058] Optionally, the altered mineral pure pixel extraction module 30 is specifically configured to: perform adaptive threshold segmentation on the altered mineral spectral index through the maximum inter-class variance algorithm to obtain the altered mineral threshold; Obtain an altered mineral mask model according to the altered mineral threshold; Among them, the altered mineral mask model is: , where MASK is the altered mineral mask model, is the altered mineral spectral index, is the threshold of the altered mineral, and Otherswise is the case not covered by the above conditions, where i = 1, 2, 3, 4, 5; Each pixel of the altered mineral spectral index is input into the altered mineral mask model for screening to obtain the pure pixels of the altered mineral.

[0059] Optionally, the altered mineral endmember set acquisition module 40 is specifically configured to: perform noise separation processing on the pure pixels of the altered mineral to obtain a pure pixel noise separation result; Among them, the pure pixel noise separation result is: , Among them, R is the pure pixel noise separation result, S is the pure pixel of the altered mineral, and P is the principal component analysis projection matrix; Calculate the angle between each pixel according to the pure pixel noise separation result to obtain spectral similarity; Among them, the spectral similarity is: , Among them, is the spectral similarity, and are the spectral reflectance vectors of two pixels respectively; Obtain the altered mineral endmember set through the spectral similarity.

[0060] Optionally, the altered mineral distribution information acquisition module 50 is specifically configured to: perform linear unmixing on the preprocessed remote sensing data and the altered mineral endmember set through hybrid tuned matching filtering to obtain endmember abundances; Among them, the endmember abundances are: , Among them, Y is the preprocessed remote sensing data, is the spectrum of the k-th endmember in the altered mineral endmember set, is the endmember abundance corresponding to the k-th endmember, and p is the number of endmembers in the altered mineral endmember set; Obtain the altered mineral distribution information through the endmember abundances.

[0061] Optionally, the preprocessed remote sensing data acquisition module 10 is specifically configured to: convert the digital value of the multispectral remote sensing image data into radiance through radiometric correction to obtain radiance image data; Among them, the radiance image data is: , Among them, is the radiance in the radiance image data, is the calibration gain coefficient, is the calibration offset, and DN is the digital value of the multi-spectral remote sensing image data; Performing stacking processing on the radiance image data to obtain the preprocessed remote sensing data.

[0062] Optionally, the preprocessed remote sensing data acquisition module 10 is specifically configured to: perform stacking processing on the radiance image data, and convert the stacked radiance image data into BIL format remote sensing data; Obtaining the preprocessed remote sensing data by normalizing the BIL format remote sensing data.

[0063] The altered mineral information extraction device based on remote sensing data in this embodiment is used to implement the altered mineral information extraction method based on remote sensing data as described above. Its advantages compared with the prior art are the same as those of the altered mineral information extraction method based on remote sensing data compared with the prior art, and will not be elaborated here.

[0064] As Figure 3 shown, an electronic device 300 provided by an embodiment of the present invention includes a memory 310 and a processor 320; the memory 310 is used to store a computer program; the processor 320 is used to implement the altered mineral information extraction method based on remote sensing data as described above when executing the computer program.

[0065] Or, an electronic device 300 includes a memory 310 and a processor 320 coupled to the memory 310; the memory 310 is configured to store a computer program; the processor 320 is configured to perform the following operations when executing the computer program: Obtain multi-spectral remote sensing image data, and preprocess the multi-spectral remote sensing image data to obtain preprocessed remote sensing data; Input the preprocessed remote sensing data into a preset spectral index model to obtain an altered mineral spectral index; Screen each pixel of the altered mineral spectral index according to the altered mineral threshold to obtain altered mineral pure pixels; Obtain an altered mineral endmember set by identifying the spectral characteristics of the altered mineral pure pixels; Perform linear unmixing on the preprocessed remote sensing data and the altered mineral endmember set through hybrid tuned matching filtering to obtain altered mineral distribution information.

[0066] A computer-readable storage medium provided by an embodiment of the present invention stores a computer program thereon. When the computer program is executed by a processor, the altered mineral information extraction method based on remote sensing data as described above is implemented.

[0067] That is to say, a non-volatile computer-readable storage medium stores a computer program thereon. When the computer program is executed by a processor, the processor performs the following operations: Obtain multi-spectral remote sensing image data, and perform preprocessing on the multi-spectral remote sensing image data to obtain preprocessed remote sensing data; Input the preprocessed remote sensing data into a preset spectral index model to obtain an altered mineral spectral index; Screen each pixel of the altered mineral spectral index according to an altered mineral threshold to obtain altered mineral pure pixels; Obtain an altered mineral endmember set by identifying the spectral characteristics of the altered mineral pure pixels; Perform linear unmixing on the preprocessed remote sensing data and the altered mineral endmember set through hybrid tuned matching filtering to obtain altered mineral distribution information.

[0068] Now, an electronic device 300 that can be a server or a client of the present invention will be described. It is an example of a hardware device that can be applied to various aspects of the present invention. The electronic device 300 is intended to represent various forms of digital electronic computer devices, such as, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device 300 can also represent various forms of mobile devices, such as, a personal digital processor, a cellular phone, a smart phone, a wearable device, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0069] The electronic device 300 includes a computing unit that can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) or a computer program loaded from a storage unit into a random access memory (RAM). In the RAM, various programs and data required for device operation can also be stored. The computing unit, the ROM, and the RAM are connected to each other through a bus. An input / output (I / O) interface is also connected to the bus.

[0070] Those of ordinary skill in the art can understand that all or part of the processes of implementing the methods in the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0071] Although the present invention has been disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will all fall within the protection scope of the present invention.

Claims

1. An altered mineral information extraction method based on remote sensing data, characterized in that, Including: Obtain multispectral remote sensing image data, and preprocess the multispectral remote sensing image data to obtain preprocessed remote sensing data; Input the preprocessed remote sensing data into a preset spectral index model to obtain an altered mineral spectral index; Screen each pixel of the altered mineral spectral index according to an altered mineral threshold to obtain altered mineral pure pixels; Obtain an altered mineral endmember set by identifying the spectral characteristics of the altered mineral pure pixels; Perform linear unmixing on the preprocessed remote sensing data and the altered mineral endmember set through mixed tuned matching filtering to obtain altered mineral distribution information.

2. The method for extracting altered mineral information based on remote sensing data according to claim 1, characterized in that The altered mineral spectral index includes a ferrous ferric spectral index, an iron oxide spectral index, an aluminum hydroxyl spectral index, a magnesium hydroxyl spectral index, and an iron hydroxyl spectral index. The step of inputting the preprocessed remote sensing data into a preset spectral index model to obtain an altered mineral spectral index includes: Obtain data for each band through the preprocessed remote sensing data; Input the data for each band into a preset spectral index model and obtain the ferrous ferric spectral index, the iron oxide spectral index, the aluminum hydroxyl spectral index, the magnesium hydroxyl spectral index, and the iron hydroxyl spectral index through band combination; Among them, the ferrous ferric spectral index is: ， Among them, is the ferrous-ferric spectral index, B21 is the band data in the wavelength range of 1.98 - 2.02 μm, B13 is the band data in the wavelength range of 0.80 - 0.83 μm, B14 is the band data in the wavelength range of 0.84 - 0.86 μm, and B15 is the band data in the wavelength range of 0.94 - 0.98 μm; Among them, the iron oxide spectral index is: ， Among them, is the iron oxide spectral index, B7 is the band data in the wavelength range of 0.62 - 0.65 μm, and B3 is the band data in the wavelength range of 0.48 - 0.5 μm; Among them, the aluminum hydroxyl spectral index is: ， Among them, is the aluminum hydroxyl spectral index, B23 is the band data in the wavelength range of 1.8–1.9 μm, and B18 is the band data in the wavelength range of 1.2–1.3 μm; Among them, the magnesium hydroxyl spectral index is: ， Among them, is the magnesium hydroxyl spectral index, B25 is the band data in the wavelength range of 2.1 - 2.2 μm, and B26 is the band data in the wavelength range of 2.3 - 2.4 μm; Among them, the iron hydroxyl spectral index is: ， Among them, is the iron hydroxyl spectral index, and B19 is the band data in the wavelength range of 1.58 - 1.65 μm.

3. The altered mineral information extraction method based on remote sensing data according to claim 2, wherein The step of screening each pixel of the altered mineral spectral index according to an altered mineral threshold to obtain altered mineral pure pixels includes: Perform adaptive threshold segmentation on the altered mineral spectral index through the maximum inter-class variance algorithm to obtain the altered mineral threshold; Obtain an altered mineral mask model according to the altered mineral threshold; Among them, the altered mineral mask model is: , where MASK is the altered mineral mask model, is the altered mineral spectral index, is the altered mineral threshold, and Otherwise is the situation not covered by the above conditions, where i = 1, 2, 3, 4, 5; Input each pixel of the altered mineral spectral index into the altered mineral mask model for screening to obtain the altered mineral pure pixels.

4. The altered mineral information extraction method based on remote sensing data according to claim 1, wherein The step of obtaining an altered mineral endmember set by identifying the spectral characteristics of the altered mineral pure pixels includes: Perform noise separation processing on the altered mineral pure pixels to obtain a pure pixel noise separation result; Among them, the pure pixel noise separation result is: , Where R is the pure pixel noise separation result, S is the altered mineral pure pixel, and P is the principal component analysis projection matrix; Calculate the angle between each pixel according to the pure pixel noise separation result to obtain spectral similarity; Among them, the spectral similarity is: , Among them, is the spectral similarity, and are the spectral reflectance vectors of two pixels, respectively; Obtain the altered mineral endmember set through the spectral similarity.

5. The method for extracting altered mineral information based on remote sensing data according to claim 1, wherein The step of performing linear unmixing on the preprocessed remote sensing data and the altered mineral endmember set through mixed tuned matching filtering to obtain altered mineral distribution information includes: Perform linear unmixing on the preprocessed remote sensing data and the altered mineral endmember set through mixed tuned matching filtering to obtain endmember abundances; Among them, the endmember abundances are: , where Y is the preprocessed remote sensing data, is the spectrum of the k-th endmember in the altered mineral endmember set, is the endmember abundance corresponding to the k-th endmember, and p is the number of endmembers in the altered mineral endmember set; Obtain altered mineral distribution information through the endmember abundances.

6. The method for extracting altered mineral information based on remote sensing data according to claim 1, wherein The step of preprocessing the multispectral remote sensing image data to obtain preprocessed remote sensing data includes: The digital values of the multi-spectral remote sensing image data are converted into radiance through radiometric calibration to obtain radiance image data; wherein, the radiance image data is: , Wherein, is the radiance in the radiance image data, is the calibration gain coefficient, is the calibration offset, and DN is the digital value of the multispectral remote sensing image data; The preprocessed remote sensing data is obtained by performing stacking processing on the radiance image data.

7. The method for extracting altered mineral information based on remote sensing data according to claim 6, wherein The step of obtaining the preprocessed remote sensing data by performing stacking processing on the radiance image data includes: Performing stacking processing on the radiance image data, and converting the radiance image data after stacking processing into BIL format remote sensing data; The preprocessed remote sensing data is obtained by performing normalization processing on the BIL format remote sensing data.

8. An altered mineral information extraction device based on remote sensing data, characterized in that, It includes: A preprocessed remote sensing data acquisition module, configured to acquire multi-spectral remote sensing image data and perform preprocessing on the multi-spectral remote sensing image data to obtain preprocessed remote sensing data; An altered mineral spectral index acquisition module, configured to input the preprocessed remote sensing data into a preset spectral index model to obtain an altered mineral spectral index; An altered mineral pure pixel extraction module, configured to screen each pixel of the altered mineral spectral index according to an altered mineral threshold to obtain altered mineral pure pixels; An altered mineral endmember set acquisition module, configured to obtain an altered mineral endmember set by identifying the spectral characteristics of the altered mineral pure pixels; An altered mineral distribution information acquisition module, configured to perform linear unmixing on the preprocessed remote sensing data and the altered mineral endmember set through hybrid tuned matching filtering to obtain altered mineral distribution information.

9. An electronic device, characterized in that, It includes a memory and a processor; The memory is used to store a computer program; The processor is configured to, when executing the computer program, implement the method for extracting altered mineral information based on remote sensing data according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, A computer program is stored on the storage medium, and when the computer program is executed by the processor, the method for extracting altered mineral information based on remote sensing data according to any one of claims 1 to 7 is implemented.

Images