Pegmatite type lithium ore prospecting method based on aviation hyperspectral remote sensing

Abstract

The invention provides a pegmatite type lithium ore prospecting method based on aviation hyperspectral remote sensing. The method comprises the following steps: step 1, inputting hyperspectral reflectivity data; step 2, calculating continuum removal data of a characteristic spectrum section; step 3, calculating a characteristic absorption wavelength position DL of the characteristic spectrum section; step 4, calculating the matching degree of the characteristic spectrum sections; 5, recognizing mineral end members; 6, inputting pegmatite vein distribution information; 7, delineating an abnormal pegmatite vein; and 8, delineating the lithium-containing pegmatite vein. The invention provides a muscovite and spodumene end member identification and lithium-containing pegmatite vein identification method, the fine identification of seven subclasses of muscovite can be realized, the identification of the subclasses in the existing aviation hyperspectral data is the most and finest, and meanwhile, the muscovite and spodumene information can be finely distinguished. Based on spodumene and muscovite information, the lithium-containing pegmatite vein can be rapidly screened from a large number of pegmatite veins, and therefore the prospecting efficiency of pegmatite type lithium ore is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of hyperspectral mineral identification, in particular to a method for prospecting pegmatite-type lithium deposits based on aerial hyperspectral remote sensing, which belongs to the field of remote sensing geology. The method is suitable for prospecting and exploring pegmatite-type lithium deposits using remote sensing data with high spatial resolution and spectral resolution. Background Art

[0002] Lithium is an important mineral resource, especially in the field of new energy. With the rapid development of the new energy industry, the demand for related raw materials is also increasing rapidly. As an energy metal, lithium provides important raw materials for new energy. Among them, pegmatite-type lithium ore is one of the important sources of lithium resources and an important object of current lithium ore prospecting and exploration. The lithium-containing minerals of pegmatite-type lithium ore are mainly developed in pegmatite veins, while pegmatite veins are widely developed, and lithium-containing pegmatite veins are rare. How to quickly screen possible lithium-containing pegmatite veins is the key to pegmatite-type lithium ore prospecting. The main ore mineral in pegmatite-type lithium ore is spodumene, and the alteration mineral closely related to lithium mineralization is muscovite. Both have an important indicative role in pegmatite-type prospecting and exploration. It can be seen that the ability to obtain surface spodumene and muscovite information is of great significance to the prospecting and exploration of pegmatite-type lithium ore.

[0003] Hyperspectral remote sensing data has a nanometer-level spectral resolution, which enables it to accurately identify minerals with spectral absorption characteristics. Aerial hyperspectral remote sensing can simultaneously obtain high spatial resolution and high spectral resolution image data, further improving the accuracy of hyperspectral remote sensing in identifying surface mineral information. At the same time, aerial hyperspectral data acquisition uses fixed-wing aircraft, which has high acquisition efficiency and is suitable for large-scale survey operations. At the same time, ultra-high spectral resolution and spectral sampling interval can not only realize the identification of mineral types, but also finely distinguish mineral components. For example, muscovite minerals, which are closely related to pegmatite-type lithium mineralization, have a high possibility of containing lithium when the spectral characteristic absorption wavelength position near 2200nm drifts in the short-wave direction. When it drifts in the long-wave direction, its mineralization possibility is very high, or even zero. In addition, the spectral characteristics of spodumene and muscovite are relatively similar, and only a higher spectral sampling interval can distinguish the two.

[0004] AisaFENIX 1K airborne hyperspectral data has the highest spectral sampling interval in the current domestic airborne hyperspectral business applications, and it has technical advantages in prospecting for pegmatite-type lithium deposits. AisaFENIX 1K can not only distinguish spodumene from muscovite, but also subdivide muscovite into seven subcategories, namely muscovite (spectral characteristic absorption wavelength position at 2187nm), muscovite (2193nm), muscovite (2199nm, muscovite (2205nm), muscovite (2211nm), muscovite (2217nm), and muscovite (2223nm). This is also the airborne hyperspectral data with the most end members that can distinguish muscovite subcategories, which is very important for prospecting for pegmatite-type lithium deposits.

[0005] Traditionally, lithium ore prospecting in pegmatite veins is mostly done using satellite multi-spectral and hyperspectral data and ground field surveys. Satellite multispectral data cannot identify spodumene due to its poor spectral resolution, and cannot accurately identify muscovite, let alone distinguish its subclasses. Satellite hyperspectral data has a spatial resolution of 30m, and the identification of spodumene and muscovite is more suitable for particularly large pegmatite veins. However, most of the current mineral-bearing pegmatite veins are meter-scale, so it cannot meet the needs. Faced with a large number of developed pegmatite veins, ground field surveys are difficult to complete vein-by-vein verification, and the efficiency is very low. Summary of the invention

[0006] The purpose of the present invention is to provide a pegmatite lithium ore prospecting method based on aerial hyperspectral remote sensing, so as to realize the rapid delineation of lithium-bearing pegmatite veins and improve the prospecting efficiency. The advantages of high spatial resolution and high spectral resolution of aerial hyperspectral data are utilized, and the characteristics of AisaFENIX 1K aerial hyperspectral data are targeted at the prospecting needs of pegmatite lithium ore. A spodumene, muscovite and lithium-bearing pegmatite vein identification method is developed, so as to realize the rapid identification of the distribution of lithium-bearing pegmatite veins by using AisaFENIX 1K aerial hyperspectral data, innovate the prospecting method of pegmatite lithium ore and improve the prospecting efficiency.

[0007] In order to achieve the above-mentioned purpose, the present invention provides a method for prospecting pegmatite-type lithium deposits based on aerial hyperspectral remote sensing. The overall technical scheme is shown in the attached Figure 1 The specific steps of this method are as follows:

[0008] Step 1: Input hyperspectral reflectance data

[0009] The hyperspectral reflectance data include AisaFENIX 1K aerial hyperspectral data and spodumene and muscovite spectral end metadata. The AisaFENIX 1K aerial hyperspectral data reflectance data is short-wave infrared range data obtained after pre-processing such as radiation calibration and atmospheric correction. The spodumene end member spectrum is obtained by measuring with an ASD spectrometer, and the muscovite spectral end metadata is obtained from the USGS spectral library and the TSG spectral library.

[0010] In the present invention, according to the band setting of AisaFENIX 1K aviation hyperspectral data, muscovite is divided into 7 subclasses, namely, subclass 1 with characteristic absorption wavelength position at 2187nm, subclass 2 at 2193nm, subclass 3 at 2199nm, subclass 4 at 2205nm, subclass 5 at 2211nm, subclass 6 at 2217nm, and subclass 7 at 2223nm. The end member spectra of the 7 muscovite subclasses are end member 1, end member 2, end member 3, end member 4, end member 5, end member 6, and end member 7. Currently, there is no spectrum of end member 1 in the existing spectral database, but analysis shows that the characteristic spectral shapes of end members 2, 3, 4, and 5 are almost the same, except for the characteristic absorption wavelength positions, that is, the characteristic spectral shapes of the muscovite subclass end member spectra with characteristic absorption wavelength positions less than or equal to 2211nm tend to be consistent, so the spectrum of end member 1 adopts end member 2 with the closest characteristic absorption wavelength position. All spectral end members are shown in Figure 2a , Figure 2b , Figure 2c , Figure 2d , Figure 2e , Figure 2f , Figure 2g .

[0011] Step 2: Calculate the characteristic spectrum segment to remove the continuum data

[0012] Calculate the decontinuum data of AisaFENIX 1K aerial hyperspectral data in the spectral range of 2150nm-2285nm. In the present invention, the continuum is defined as the line between the reflection peaks in the reflection spectrum curve between the characteristic spectral segments. The decontinuum is calculated by dividing the reflection spectrum of the characteristic spectral segment by its continuum spectrum. The calculation method of the decontinuum is: Rc = Rr / Rq

[0013] Among them, Rc is the spectral data after removing the continuum of the characteristic spectral segment, Rr is the original spectral data of the characteristic spectral segment, and Rq is the continuum data of the original spectrum of the characteristic spectral segment.

[0014] Step 3: Calculate the characteristic absorption wavelength position of the characteristic spectral segment

[0015] That is, according to the characteristic spectrum segment calculated in step 2, the continuum data Rc is removed, and the characteristic absorption wavelength position is obtained by calculating the minimum reflectivity value of Rc:

[0016] DL = where (Rc = Rmin);

[0017] Among them, Rmin is the minimum reflectivity value.

[0018] Step 4: Calculate the matching degree of characteristic spectrum segments

[0019] That is, according to the characteristic absorption wavelength position DL calculated in step 3, the matching degree MD between each spectral end member and the AisaFENIX 1K aviation hyperspectral data is determined and calculated respectively;

[0020] 1) If DL = 2187nm, the calculated spectrum of end member 2 in the 2144nm-2260nm spectral range is matched with the AisaFENIX 1K aerial hyperspectral data in the 2138nm-2254nm range. The matching algorithm uses the spectral angle method. The closer the calculation result is to 0, the higher the similarity. The calculation formula is as follows:

[0021]

[0022] Wherein, Sa is the spectral angle; Rri is the reflectance value of the hyperspectral data; Sri is the end member spectral reflectance value;

[0023] 2) If DL = 2193 nm, calculate the matching degree MD between the spectrum of end member 2 and the AisaFENIX 1K aerial hyperspectral data in the range of 2144 nm-2260 nm. The matching degree algorithm also uses the formula in 1).

[0024] 3) If WL = 2199nm, calculate the matching degree MD between the end member 3 spectrum and the AisaFENIX 1K aerial hyperspectral data in the range of 2144nm-2278nm. The matching degree algorithm also uses the formula in 1).

[0025] 4) If WL = 2205 nm, calculate the matching degree MD between the end member 4 spectrum and the spodumene spectrum and the AisaFENIX 1K aerial hyperspectral data in the range of 2144 nm-2278 nm. The matching degree algorithm also uses the formula in 1).

[0026] 5) If WL = 2211 nm, calculate the matching degree MD between the spectrum of end member 5 and the AisaFENIX 1K aerial hyperspectral data in the range of 2144 nm-2285 nm. The matching degree algorithm also uses the formula in 1).

[0027] 6) If WL = 2217nm, calculate the matching degree MD between the spectrum of end member 6 and the AisaFENIX 1K aerial hyperspectral data in the range of 2150nm-2285nm. The matching degree algorithm also uses the formula in 1).

[0028] 7) If WL = 2223 nm, calculate the matching degree MD between the spectrum of end member 7 and the AisaFENIX 1K aerial hyperspectral data in the range of 2150 nm-2285 nm. The matching degree algorithm also uses the formula in 1).

[0029] 8) If DL is not equal to the above value, the matching degree is 1.

[0030] Step 5: Mineral end member identification

[0031] A threshold is set for each matching degree MD calculated in the step 4, and the result that satisfies the threshold condition is the mineral end member identification result. In order to further distinguish spodumene from muscovite, before setting the threshold for spodumene and end member 4, it is necessary to compare the spectral angles Sa of the two. When the matching degree MD of spodumene is less than the matching degree of end member 4, the matching result can only be spodumene. On the contrary, when the matching degree MD of spodumene is greater than or equal to the matching degree of end member 4, it can only be muscovite. The threshold setting is based on the matching degree MD result. First, check the change of the spectral spectrum type of the AisaFENIX1K aerial hyperspectral image corresponding to the matching degree MD from low to high. When the spectral spectrum type of the AisaFENIX 1K aerial hyperspectral image is changed to the point where it is basically impossible to determine whether it contains mineral end members, the value corresponding to the matching degree MD is the threshold of the matching degree MD. This threshold range has good universal applicability for AisaFENIX 1K aerial hyperspectral data.

[0032] Step 6: Input pegmatite vein distribution information

[0033] The distribution information of pegmatite veins can be delineated through existing survey results or visual interpretation of high spatial resolution remote sensing data. Visual interpretation of high spatial resolution remote sensing data is mainly based on the morphology and color characteristics of pegmatite veins.

[0034] Step 7: Delineation of abnormal pegmatite veins

[0035] The end members of the muscovite subtype in the lithium-bearing pegmatite vein must be 2199nm and below. That is, when the sum of end members 1, 2 and 3 identified on the pegmatite vein exceeds the other end members, the pegmatite vein is judged to be an abnormal pegmatite vein.

[0036] Step 8: Delineation of lithium-bearing pegmatite veins

[0037] Based on the abnormal pegmatite veins obtained in step seven, combined with the identified spodumene information, the abnormal pegmatite veins with identified spodumene mineral information are circled as lithium-bearing pegmatite veins.

[0038] The advantages and beneficial effects of the present invention are as follows: the present invention gives full play to the advantages of high spatial resolution and high spectral resolution of aerial hyperspectral remote sensing data, proposes a method for identifying muscovite and spodumene end members and lithium-bearing pegmatite veins, and can achieve fine identification of seven subclasses of muscovite. It is the most and most detailed subclass identification in the existing aerial hyperspectral data, and can also finely distinguish the information of muscovite and spodumene. Based on the information of spodumene and muscovite, lithium-bearing pegmatite veins can be quickly screened out from a large number of pegmatite veins, thereby improving the efficiency of prospecting for pegmatite-type lithium deposits. The present invention is not only applicable to AisaFENIX 1K aerial hyperspectral remote sensing data, but also to other hyperspectral remote sensing data with similar technical parameters. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 Flow chart for implementing the present invention.

[0040] Figure 2a , Figure 2b , Figure 2c , Figure 2d , Figure 2e , Figure 2f , Figure 2g This is a characteristic diagram of the mineral end-member spectral curve for AisaFENIX 1K aerial hyperspectral remote sensing data.

[0041] Figure 3 This is a diagram showing the identification results of muscovite subclass 1 according to an embodiment of the present invention.

[0042] Figure 4 This is a diagram showing the identification results of muscovite subclass 2 according to an embodiment of the present invention.

[0043] Figure 5 This is a diagram showing the identification results of muscovite subclass 3 in an embodiment of the present invention.

[0044] Figure 6 This is a diagram showing the identification results of muscovite subclass 4 according to an embodiment of the present invention.

[0045] Figure 7 This is a diagram showing the identification results of muscovite subclass 5 according to an embodiment of the present invention.

[0046] Figure 8 This is a diagram showing the identification results of muscovite subclass 6 according to an embodiment of the present invention.

[0047] Fig. 9 This is a diagram showing the identification results of muscovite subclass 7 according to an embodiment of the present invention.

[0048] Fig.10 This is a diagram of spodumene identification results according to an embodiment of the present invention.

[0049] Fig.11 This is a distribution map of pegmatite veins according to an embodiment of the present invention.

[0050] Fig.12 This is a diagram showing the results of identifying abnormal pegmatite veins according to an embodiment of the present invention.

[0051] Fig.13 This is a diagram showing the delineation results of lithium-bearing pegmatite veins in an embodiment of the present invention. DETAILED DESCRIPTION

[0052] Figure 1 This is the implementation process of the method of the present invention. In order to better illustrate the implementation and effect of the method, an application experiment of prospecting for pegmatite-type lithium deposits was carried out using AisaFENIX 1K aerial hyperspectral remote sensing data as an example. The specific steps are as follows:

[0053] Step 1: Input hyperspectral reflectance data

[0054] The data for this example is the AisaFENIX 1K aerial hyperspectral data obtained in Qinghe County, Altay Prefecture on June 6, 2024, with a spatial resolution of 1.0m, a spectral coverage range of 400-2500nm, and a short-wave infrared spectrum sampling interval of 6nm. Aerial hyperspectral data is the emissivity data obtained after preprocessing such as radiation correction and atmospheric correction; the mineral end-member spectral data comes from the USGS spectral library, TSG spectral library and measured spectra. Aerial hyperspectral reflectance data and end-member spectral data are both 0-1 floating point data; the end-member spectra have been resampled to aerial hyperspectral data. Both are input into the ENVI IDL program.

[0055] Step 2: Calculate the characteristic spectrum segment to remove the continuum data

[0056] Calculate the decontinuum data of AisaFENIX 1K aerial hyperspectral data in the spectral range of 2150nm-2285nm. Use IDL language to write the calculation function of spectrum decontinuum. The calculation method of decontinuum is: Rc=Rr / Rq

[0057] Among them, Rc is the spectral data after removing the continuum of the characteristic spectral segment, Rr is the original spectral data of the characteristic spectral segment, and Rq is the continuum data of the original spectrum of the characteristic spectral segment.

[0058] Step 3: Calculate the characteristic absorption wavelength position of the characteristic spectral segment

[0059] According to the characteristic spectrum segment decontinuum data Rc calculated in step 2, the characteristic absorption wavelength position is obtained by calculating the minimum reflectivity value of Rc. Using IDL language, a function for calculating the characteristic absorption wavelength position is written. The calculation method for decontinuum is:

[0060] DL = where (Rc = Rmin);

[0061] Among them, Rmin is the minimum reflectivity value.

[0062] Step 4: Calculate the matching degree of characteristic spectrum segments

[0063] According to the characteristic absorption wavelength position DL calculated in step 3, the matching degree MD between each spectral end member and the AisaFENIX 1K aviation hyperspectral data is determined and calculated respectively;

[0064] 1) If DL = 2187nm, the calculated end member 2 spectrum in the 2144nm-2260nm spectral range is matched with the AisaFENIX 1K aerial hyperspectral data in the 2138nm-2254nm range. The matching algorithm uses the spectral angle method. The closer the calculation result is to 0, the higher the similarity. The matching calculation function is written in IDL language. The calculation formula is as follows:

[0065]

[0066] Wherein, Sa is the spectral angle; Rri is the reflectance value of the hyperspectral data; Sri is the end member spectral reflectance value;

[0067] 2) If DL = 2193 nm, calculate the matching degree MD between the spectrum of end member 2 and the AisaFENIX 1K aerial hyperspectral data in the range of 2144 nm-2260 nm. The matching degree algorithm also uses the formula in 1).

[0068] 3) If WL = 2199nm, calculate the matching degree MD between the end member 3 spectrum and the AisaFENIX 1K aerial hyperspectral data in the range of 2144nm-2278nm. The matching degree algorithm also uses the formula in 1).

[0069] 4) If WL = 2205 nm, calculate the matching degree MD between end member 4 and spodumene spectrum and AisaFENIX 1K aerial hyperspectral data in the range of 2144 nm-2278 nm. The matching degree algorithm also uses the formula in 1).

[0070] 5) If WL = 2211 nm, calculate the matching degree MD between the spectrum of end member 5 and the AisaFENIX 1K aerial hyperspectral data in the range of 2144 nm-2285 nm. The matching degree algorithm also uses the formula in 1).

[0071] 6) If WL = 2217nm, calculate the matching degree MD between the spectrum of end member 6 and the AisaFENIX 1K aerial hyperspectral data in the range of 2150nm-2285nm. The matching degree algorithm also uses the formula in 1).

[0072] 7) If WL = 2223 nm, calculate the matching degree MD between the spectrum of end member 7 and the AisaFENIX 1K aerial hyperspectral data in the range of 2150 nm-2285 nm. The matching degree algorithm also uses the formula in 1).

[0073] 8) If DL is not equal to the above value, the matching degree is 1.

[0074] This step calculates that the matching degree values ​​of muscovite subclass 1, muscovite subclass 2, muscovite subclass 3, muscovite subclass 4, muscovite subclass 5, muscovite subclass 6, muscovite subclass 7, and spodumene in the above characteristic spectral bands are in the range of 0.088-0.789, 0.067-0.801, 0.023-0.966, 0.031-0.969, 0.053-0.964, 0.038-0.938, 0.053-0.946, and 0.026-0.964, respectively.

[0075] Step 5: Mineral end member identification

[0076] According to the fourth step, the matching degree MD result is calculated and the threshold is set. In this embodiment, the matching degree MD threshold ranges of muscovite subclass 1, muscovite subclass 2, muscovite subclass 3, muscovite subclass 4, muscovite subclass 5, muscovite subclass 6, muscovite subclass 7, and spodumene are 0≤S1≤0.106, 0≤S2≤0.110, 0≤S3≤0.085, 0≤S4≤0.086, 0≤S5≤0.110, 0≤S6≤0.075, 0≤S7≤0.098, and 0≤S8≤0.054 respectively; among them, the size of muscovite subclass 4 and spodumene is compared with each other before the threshold is set. Mineral identification results are shown in Figure 3-10 This threshold range is also widely applicable to other AisaFENIX 1K aerial hyperspectral data.

[0077] Step 6: Input pegmatite vein distribution information

[0078] The distribution information of pegmatite veins in this example was determined by visual interpretation of remote sensing data from the Gaofen-7 satellite with a spatial resolution of 0.65 m. Fig.11 .

[0079] Step 7: Delineation of abnormal pegmatite veins

[0080] The pegmatite veins mainly developed in the muscovite subtypes 1, 2 and 3 are abnormal pegmatite veins, see the attached Fig.12 .

[0081] Step 8: Delineation of lithium-bearing pegmatite veins

[0082] Based on the abnormal pegmatite veins obtained in step 7, combined with the identified spodumene information, the abnormal pegmatite veins with spodumene identified are delineated as lithium-bearing pegmatite veins, see attached Fig.13 .

Claims

1. A method for prospecting pegmatite-type lithium deposits based on aerial hyperspectral remote sensing, characterized in that: Here are the steps: Step 1: Input hyperspectral reflectance data: The hyperspectral reflectance data include AisaFENIX 1K aerial hyperspectral data and spodumene and muscovite spectral end metadata. The AisaFENIX 1K aerial hyperspectral data reflectance data is short-wave infrared range data obtained after radiometric calibration and atmospheric correction preprocessing. The spodumene end member spectrum is obtained by ASD spectrometer measurement, and the muscovite spectral end metadata is obtained from USGS spectral library and TSG spectral library. Step 2: Calculate the characteristic spectrum segment to remove the continuum data: Calculate the decontinuum data of AisaFENIX 1K aerial hyperspectral data in the spectral range of 2150nm-2285nm. The continuum is defined as the connecting line between the reflection peaks in the reflection spectrum curve between the characteristic spectral bands. Step 3: Calculate the characteristic absorption wavelength position DL of the characteristic spectrum segment: Step 4: Calculate the matching degree of characteristic spectrum segment: According to the calculated characteristic absorption wavelength position DL, the matching degree MD between each spectral end member and the AisaFENIX 1K aviation hyperspectral data is determined and calculated respectively; Step 5: Identification of mineral end members: A threshold is set for each matching degree MD, and the mineral end member identification result is the one that meets the threshold condition; Step 6. Input pegmatite vein distribution information: The distribution information of pegmatite veins is delineated through existing survey results or visual interpretation of high spatial resolution remote sensing data; Visual interpretation of high spatial resolution remote sensing data is based on the morphology and color characteristics of pegmatite veins; Step 7: Delineation of abnormal pegmatite veins: The end member of the muscovite subtype in the lithium-bearing pegmatite vein must be 2199nm or below, that is, when the sum of end members 1, 2 and 3 identified on the pegmatite vein exceeds the other end members, the pegmatite vein is judged to be an abnormal pegmatite vein; Step 8: Delineation of lithium-bearing pegmatite veins: Based on the abnormal pegmatite veins obtained in step seven, combined with the identified spodumene information, the abnormal pegmatite veins with identified spodumene mineral information are circled as lithium-bearing pegmatite veins.

2. The method for prospecting pegmatite-type lithium deposits based on aerial hyperspectral remote sensing according to claim 1, characterized in that: In step 1, according to the band settings of AisaFENIX 1K aerial hyperspectral data, muscovite is divided into 7 subclasses, namely, subclass 1 with characteristic absorption wavelength at 2187nm, subclass 2 at 2193nm, subclass 3 at 2199nm, subclass 4 at 2205nm, subclass 5 at 2211nm, subclass 6 at 2217nm, and subclass 7 at 2223nm; the end member spectra of the 7 muscovite subclasses are end member 1, end member 2, end member 3, end member 4, end member 5, end member 6, and end member 7.

3. The method for prospecting pegmatite-type lithium deposits based on aerial hyperspectral remote sensing according to claim 2, characterized in that: Currently, there is no end member 1 spectrum in the existing spectral database, and the end member 1 spectrum uses the end member 2 with the closest characteristic absorption wavelength position.

4. The method for prospecting pegmatite-type lithium deposits based on aerial hyperspectral remote sensing according to claim 1, characterized in that: In step 2, continuum removal is to divide the reflected spectrum of the characteristic spectral segment by its continuum spectrum, and the calculation method for removing the continuum is: Rc=Rr / Rq; wherein Rc is the spectral data after removing the continuum of the characteristic spectral segment, Rr is the original spectral data of the characteristic spectral segment, and Rq is the continuum data of the original spectrum of the characteristic spectral segment.

5. A method for prospecting pegmatite-type lithium deposits based on aerial hyperspectral remote sensing according to claim 1 or 4, characterized in that: In step three, the characteristic absorption wavelength position is obtained by calculating the minimum reflectivity value of Rc based on the characteristic spectrum segment calculated in step two to remove the continuum data Rc: DL = where (Rc = Rmin); Among them, Rmin is the minimum reflectivity value.

6. The method for prospecting pegmatite-type lithium deposits based on aerial hyperspectral remote sensing according to claim 1, characterized in that: In step four, 1) If DL = 2187nm, the calculated spectrum of end member 2 in the 2144nm-2260nm spectral range is matched with the AisaFENIX 1K aerial hyperspectral data in the 2138nm-2254nm range. The matching algorithm uses the spectral angle method. The closer the calculation result is to 0, the higher the similarity. The calculation formula is as follows: Wherein, Sa is the spectral angle; Rri is the reflectance value of the hyperspectral data; Sri is the end member spectral reflectance value; 2) If DL = 2193nm, calculate the matching degree MD between the spectrum of end member 2 and the AisaFENIX 1K aerial hyperspectral data in the range of 2144nm-2260nm; the matching degree algorithm also uses the formula in 1); 3) If WL = 2199nm, calculate the matching degree MD between the spectrum of end member 3 and the AisaFENIX 1K aerial hyperspectral data in the range of 2144nm-2278nm; the matching degree algorithm also uses the formula in 1); 4) If WL = 2205 nm, calculate the matching degree MD between the end member 4 spectrum and the spodumene spectrum and the AisaFENIX 1K aerial hyperspectral data in the range of 2144 nm-2278 nm respectively; the matching degree algorithm also uses the formula in 1); 5) If WL = 2211 nm, calculate the matching degree MD between the spectrum of end member 5 and the AisaFENIX 1K aerial hyperspectral data in the range of 2144 nm-2285 nm; the matching degree algorithm also uses the formula in 1); 6) If WL = 2217nm, calculate the matching degree MD between the spectrum of end member 6 and the AisaFENIX 1K aerial hyperspectral data in the range of 2150nm-2285nm; the matching degree algorithm also uses the formula in 1); 7) If WL = 2223nm, calculate the matching degree MD between the spectrum of end member 7 and the AisaFENIX 1K aerial hyperspectral data in the range of 2150nm-2285nm; the matching degree algorithm also uses the formula in 1); 8) If DL is not equal to the above value, the matching degree is 1.

7. The method for prospecting pegmatite-type lithium deposits based on aerial hyperspectral remote sensing according to claim 1, characterized in that: In step five, in order to further distinguish spodumene from muscovite, before setting the threshold for spodumene and end member 4, it is necessary to compare the spectral angles Sa of the two. When the matching degree MD of spodumene is less than the matching degree of end member 4, the matching result can only be spodumene. On the contrary, when the matching degree MD of spodumene is greater than or equal to the matching degree of end member 4, it can only be muscovite. The threshold setting is based on the matching degree MD result.

8. The method for prospecting pegmatite-type lithium deposits based on aerial hyperspectral remote sensing according to claim 7, characterized in that: Check the changes in the spectral type of the AisaFENIX 1K aerial hyperspectral image when the matching degree MD changes from low to high. When the spectral type of the AisaFENIX 1K aerial hyperspectral image can no longer determine whether it contains mineral end members, the value corresponding to the matching degree MD is the threshold of the matching degree MD.

Images