Rapid estimation method for lithium content in clay

Abstract

The invention relates to the field of estimation of soil lithium content through hyperspectral remote sensing, in particular to a rapid estimation method for the lithium content in clay. The method comprises the following steps: acquiring spectral reflectivity data and corresponding lithium element content of a clay sample; according to the spectral reflectivity data, spectral data of multiple spectral resolutions are obtained, and the optimal spectral resolution is obtained by combining the spectral data and the lithium element content; obtaining a plurality of groups of fractional order derivative spectrums based on the optimal spectral resolution, and obtaining a characteristic spectrum data set through the correlation between the fractional order derivative spectrums and the lithium element content; and constructing a lithium content rapid estimation model, training and verifying the lithium content rapid estimation model by using the characteristic spectrum data set, and completing rapid estimation of the lithium content in the clay through the trained lithium content rapid estimation model. According to the method, the characteristic spectrum data set under multiple fractional derivative spectrums is obtained under the optimal resolution, and then the problem of quickly, efficiently and accurately estimating the lithium content of the clay is solved by utilizing the quick estimation model.

Description

Technical Field

[0001] The invention relates to the field of estimating soil lithium content by hyperspectral remote sensing, and in particular to a method for quickly estimating lithium content in clay. Background Art

[0002] Lithium is an "emerging" environmental pollutant, especially lithium pollution in soil, which can endanger human health through the food chain and has attracted widespread attention. Traditional survey methods mainly rely on field sampling and indoor large-scale equipment chemical analysis. Although they are highly accurate, they are costly and inefficient. Moreover, monitoring of large areas is mainly based on spatial interpolation, which is very dependent on the accuracy of data and the uniformity of spatial distribution.

[0003] Although some studies have used hyperspectral remote sensing to estimate the content of heavy metals in soil, most of them mainly use raw reflectance spectral data or perform various transformations such as inverse, logarithm and envelope removal to build models. There is no systematic analysis of the impact of spectral resolution on the accuracy of the estimation model. The transformation spectra are all based on integer orders, and the impact of non-zero integer derivatives on the accuracy of the estimation model is not analyzed. In terms of feature band selection, feature bands are only extracted for a certain spectrum, and there is no comprehensive feature extraction of multiple forms of spectra. In particular, there are very few studies on the estimation of lithium content in clay.

[0004] Therefore, there is an urgent need for a method that is highly applicable and can quickly, efficiently and accurately estimate the lithium content in clay. Summary of the invention

[0005] In view of the shortcomings of existing methods and the needs of practical applications, in order to comprehensively extract features of various forms of spectra, improve the accuracy of non-zero integer derivatives for estimation models, and solve the problem of quickly, efficiently and accurately estimating the lithium content of clay. The present invention provides a method for quickly estimating the lithium content in clay, comprising the following steps: obtaining spectral reflectance data and corresponding lithium content of clay samples; obtaining spectral data of multiple spectral resolutions according to the spectral reflectance data, combining the spectral data and the lithium content to obtain the optimal spectral resolution; obtaining multiple groups of fractional derivative spectra based on the optimal spectral resolution, and obtaining a characteristic spectral data set through the correlation between the fractional derivative spectra and the lithium content; constructing a rapid estimation model for lithium content, using the characteristic spectral data set to train and verify the rapid estimation model for lithium content, and completing the rapid estimation of lithium content in clay through the trained rapid estimation model for lithium content. The present invention determines the optimal spectral resolution of spectral data by window sampling, and then divides the characteristic spectral data sets under multiple fractional derivative spectra, and then extracts the characteristic spectral features and then uses the rapid estimation model for lithium content to complete the rapid estimation of lithium content in clay, thereby improving the applicability of the present invention and solving the problem of quickly, efficiently and accurately estimating the lithium content of clay.

[0006] Optionally, obtaining spectral data of multiple spectral resolutions according to the spectral reflectance data comprises the following steps: The size of the sliding window is set; based on the sliding window, the spectral reflectance data is divided to obtain spectral data of multiple spectral resolutions. By setting different sliding window sizes, spectral reflectance data of different spectral resolutions are obtained, which is conducive to improving the estimation accuracy of the present invention.

[0007] Optionally, combining the spectral data and the lithium content to obtain an optimal spectral resolution comprises the following steps: According to the spectral data and the lithium content, a spectral data training set and a spectral data verification set are established; a machine classification model is constructed, the machine classification model is trained using the spectral data training set, and the verification result is obtained by combining the spectral data verification set and the trained machine classification model; the verification result is comprehensively analyzed using the determination coefficient, relative root mean square error and relative analysis error to obtain the optimal spectral resolution. The present invention uses a variety of error analysis methods, which is conducive to quickly and accurately obtaining the optimal spectral resolution.

[0008] Optionally, the multiple groups of fractional-order derivative spectra obtained based on the optimal spectral resolution include the following steps: The step interval is set; and according to the step interval, fractional order derivative spectra of order 0 to 2 are obtained. The present invention further subdivides the integer order derivative spectrum, thereby reducing the influence of non-zero integer order derivatives on the accuracy of the estimation model.

[0009] Optionally, the fractional derivative spectra of orders 0 to 2 are obtained according to the step intervals, satisfying the following formula: in, Represents the function The a-order fractional derivative of , represents the order of the fractional order, Indicates the band is The spectral reflectance at , Indicates the band position, represents the Gamma function, It represents the spectral reflectance at the wavelength t. The present invention obtains multiple groups of fractional-order derivative spectra through a model formula, which is further conducive to feature extraction, thereby improving the accuracy of the present invention.

[0010] Optionally, obtaining a characteristic spectrum data set through the correlation between the fractional derivative spectrum and the lithium content comprises the following steps: The correlation between the fractional derivative spectrum and the lithium content is calculated by a first correlation model; the fractional derivative spectrum is screened based on the correlation to obtain a significantly correlated spectrum data set; significant characteristic spectrum bands are extracted from the significantly correlated spectrum data set to construct a characteristic spectrum data set. The present invention screens useful spectrum data sets by a first correlation model to extract effective characteristic spectra, which is further conducive to improving the estimation speed of the present invention.

[0011] It can be seen that the first correlation model satisfies the following formula: in, represents the first correlation, represents the number of samples, express The reflectance of the samples, expressed as Indicates The lithium content of each sample is represents the band mean of all samples, The present invention uses a model to quantify the first correlation, which is objective and accurate, and further ensures the accuracy of the data.

[0012] Optionally, the fractional derivative spectra are screened based on the correlation to obtain a significantly correlated spectrum data set, satisfying the following formula: in, The present invention removes the first correlation less than 0.6, which reduces the calculation data and is beneficial to improving the efficiency of the present invention.

[0013] Optionally, extracting significant characteristic spectral bands from the significant correlation spectral data set comprises the following steps: Extract band features from the significantly correlated spectral data set; construct a band feature importance assessment model, and use the band feature importance assessment model to obtain the importance of the band features; and obtain significant feature spectral bands based on the importance. The present invention obtains significant feature spectral bands through an assessment model, integrates feature extraction of multiple forms of spectra, and further improves the accuracy of the present invention.

[0014] Optionally, the band feature importance evaluation model satisfies the following formula: in, Indicates The band characteristics are Importance on the dataset, Indicates sample data that has not been used to train a specific decision branch. Indicates Branches The number of samples of data, Indicates the number of branches, represents the indicator function, Indicates The actual result of samples, Indicates The sample The prediction result of each branch, After random permutation Branch pairs Data The prediction results of samples. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A flow chart of a method for quickly estimating lithium content in clay provided by an embodiment of the present invention; Figure 2 A framework diagram of a system for quickly estimating lithium content in clay provided by an embodiment of the present invention; Figure 3 A schematic diagram of the structure of a device for quickly estimating lithium content in clay provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0016] The specific embodiments of the present invention will be described in detail below. It should be noted that the embodiments described herein are only for illustration and are not intended to limit the present invention. In the following description, a large number of specific details are set forth in order to provide a thorough understanding of the present invention. However, it is obvious to those of ordinary skill in the art that these specific details do not need to be adopted to implement the present invention. In other examples, in order to avoid confusing the present invention, known circuits, software or methods are not specifically described.

[0017] Throughout the specification, references to "one embodiment," "an embodiment," "an example," or "an example" mean that a particular feature, structure, or characteristic described in conjunction with the embodiment or example is included in at least one embodiment of the present invention. Therefore, the phrases "in one embodiment," "in an embodiment," "an example," or "an example" appearing in various places throughout the specification do not necessarily all refer to the same embodiment or example. In addition, particular features, structures, or characteristics may be combined in one or more embodiments or examples in any suitable combination and / or subcombination. In addition, it should be understood by those of ordinary skill in the art that the figures provided herein are for illustrative purposes and that the figures are not necessarily drawn to scale.

[0018] See also Figure 1In order to integrate the feature extraction of various forms of spectra, improve the accuracy of the estimation model of non-zero integer derivatives, and solve the problem of quickly, efficiently and accurately estimating the lithium content of clay. The present invention provides a method for quickly estimating the lithium content in clay, such as Figure 1 As shown, in one embodiment, the method includes the following steps: S1. Obtain the spectral reflectance data and corresponding lithium content of the clay sample.

[0019] In the embodiment, soil samples of the study area collected in the field are tested in the laboratory to obtain spectral reflectance data and corresponding lithium content of the samples.

[0020] Specifically, a FieldSpec4 portable spectrometer was used to measure the sample at a vertical distance of 20 cm from the sample to obtain the spectral reflectance data of the sample; an iCAP7400 inductively coupled plasma emission spectrometer was used to detect the sample at a temperature of 21 degrees Celsius and a relative humidity of 40% to obtain the lithium content of the sample.

[0021] S2. Obtain spectral data of multiple spectral resolutions according to the spectral reflectance data, and obtain an optimal spectral resolution by combining the spectral data and the lithium content.

[0022] Specifically, step S2, obtaining spectral data of multiple spectral resolutions according to the spectral reflectance data, includes the following steps: S21. Set the size of the sliding window.

[0023] The spectral resolution of the spectral data obtained in step S1 is 1 nm. In order to obtain spectral data with different spectral resolutions, a window sliding method is used to perform spectral resampling.

[0024] S22. Based on the sliding window, divide the spectral reflectance data to obtain spectral data of multiple spectral resolutions.

[0025] According to the size of the sliding window, spectral data of different sliding windows, i.e., different spectral resolutions, are obtained segment by segment. In an embodiment, the sizes of the sliding windows are 5, 10, 20, 25, and 40, respectively, so as to obtain spectral data of 5nm, 10nm, 20nm, 25nm, and 40nm spectral resolutions.

[0026] Specifically, the spectral data with a spectral resolution of 5nm is obtained from the spectral data with a spectral resolution of 1nm, using a smoothing method with a window of 5, that is, starting from the 1st band and ending at the 5th band, the spectral average of these 5 bands is calculated as the first band with a spectral resolution of 5nm, and then the window slides forward 5 times, that is, the spectral average of the 5 bands from the 6th band to the 10th band is calculated as the second band with a spectral resolution of 5nm, and so on. The remaining bands are calculated, and finally the sample spectral data with a spectral resolution of 5nm is obtained.

[0027] Furthermore, the step S2 of combining the spectral data and the lithium content to obtain an optimal spectral resolution comprises the following steps: According to the spectral data and the lithium content, a spectral data training set and a spectral data verification set are established; Constructing a machine classification model, training the machine classification model using the spectral data training set, and combining the spectral data validation set and the trained machine classification model to obtain a validation result; The verification results were comprehensively analyzed using the determination coefficient, relative root mean square error and relative analytical error to obtain the optimal spectral resolution.

[0028] Specifically, based on spectral data with spectral resolutions of 1nm, 5nm, 10nm, 20nm, 25nm and 40nm, the data set is divided into training set and test set in a certain proportion, and a machine classification model is constructed. The machine classification model includes support vector machines and neural networks. It can also be constructed according to the partial least squares method. The spectral resolution with the best model effect is comprehensively determined based on the performance in the validation set, combined with the determination coefficient, relative root mean square error and relative analytical error.

[0029] Furthermore, the coefficient of determination satisfies the following formula: The relative root mean square error satisfies the following formula: The relative analysis error satisfies the following formula: in, , and The samples are The true value, predicted value and average value of .

[0030] S3. Obtain multiple groups of fractional derivative spectra based on the optimal spectral resolution, and obtain a characteristic spectrum data set through the correlation between the fractional derivative spectra and the lithium content.

[0031] In an embodiment, the step of obtaining multiple groups of fractional-order derivative spectra based on the optimal spectral resolution in S3 comprises the following steps: S31. Set the step interval.

[0032] Specifically, the size of the step interval affects the number of fractional derivative spectra. In this embodiment, the step interval is 0.05.

[0033] S32. Obtain fractional-order derivative spectra of order 0 to 2 according to the step intervals.

[0034] 40 groups of fractional derivative spectra of order 0-2 and interval 0.05 are calculated at the best spectral resolution. In other embodiments, more fractional derivative spectra can be obtained according to actual computing power and other conditions.

[0035] Specifically, the fractional derivative spectra of order 0 to 2 are obtained according to the step interval, satisfying the following formula: in, Represents the function The a-order fractional derivative of , represents the order of the fractional order, Indicates the band is The spectral reflectance at , Indicates the band position, represents the Gamma function, Represents the spectral reflectance at wavelength t.

[0036] Further, obtaining a characteristic spectrum data set through the correlation between the fractional derivative spectrum and the lithium content comprises the following steps: S33. Calculate the correlation between the fractional derivative spectrum and the lithium content through a first correlation model.

[0037] Specifically, the first correlation model satisfies the following formula: in, represents the first correlation, represents the number of samples, express The reflectance of the samples, expressed as Indicates The lithium content of each sample is represents the band mean of all samples, Represents the mean lithium content of all samples. The correlation between all fractional derivative spectra and lithium content is calculated using the first correlation model.

[0038] S34, screening the fractional derivative spectra based on the correlation to obtain a significantly correlated spectrum data set.

[0039] Specifically, the fractional derivative spectra are screened based on the correlation to obtain a significantly correlated spectral data set that satisfies the following formula: in, Denotes the first correlation. All spectral data with a correlation of each fractional derivative spectrum with lithium content greater than 0.6 were extracted and combined to obtain a significantly correlated spectral data set.

[0040] S35, extracting significant characteristic spectral bands from the significantly correlated spectral data set, and then constructing a characteristic spectral data set.

[0041] In an embodiment, the extracting of significant characteristic spectral bands from the significant correlation spectral data set comprises the following steps: S351, extracting band features from the significantly correlated spectral data set.

[0042] Band features are extracted from the significantly correlated spectral data set using a machine learning feature extraction model, wherein the machine learning feature extraction model includes a gradient boosting tree, a random forest, and a decision tree stump.

[0043] S352: construct a band feature importance assessment model, and use the band feature importance assessment model to obtain the importance of the band feature.

[0044] In the embodiment, the band features are extracted by a machine learning model such as a random forest, and the band feature importance evaluation model satisfies the following formula: in, Indicates The band characteristics are Importance on the dataset, Indicates sample data that has not been used to train a specific decision branch. Indicates Branches The number of samples of data, Indicates the number of branches, represents the indicator function, Indicates The actual result of samples, Indicates The sample The prediction result of each branch, After random permutation Branch pairs Data The prediction results of samples.

[0045] S353. Obtain significant characteristic spectral bands according to the importance.

[0046] All band features were used to train the random forest model, and then the importance of each band feature was calculated. The band features were then sorted according to their importance, and the top k band features with the highest importance were selected. The model effect of the selected band features was evaluated by cross-validation method based on the determination coefficient, relative root mean square error, and relative analytical error. Significant characteristic spectral bands were obtained, and then a characteristic spectral data set was constructed based on the corresponding lithium content.

[0047] S4. Constructing a rapid estimation model for lithium content, using the characteristic spectrum data set to train and verify the rapid estimation model for lithium content, and completing the rapid estimation of lithium content in clay through the trained rapid estimation model for lithium content.

[0048] Based on the characteristic spectral data set of step S3, a training set and a validation set are divided into a certain proportion; then, a lithium content rapid estimation model is constructed according to the partial least squares method or other machine learning algorithms, and the lithium content rapid estimation model is trained and verified using the characteristic spectral data set, and the spectral data of the clay to be detected area and the trained lithium content rapid estimation model are combined to complete the rapid estimation of the lithium content in the clay.

[0049] See also Figure 2 In an embodiment, in order to efficiently execute the method for rapidly estimating the lithium content in clay provided by the present invention, the present invention also provides a system for rapidly estimating the lithium content in clay, comprising: an input device, an output device, a processor, and a memory, wherein the input device, the output device, the processor, and the memory are interconnected, and the memory contains program instructions, and the program instructions are used for the steps of the method for rapidly estimating the lithium content in clay. The system for rapidly estimating the lithium content in clay of the present invention has a compact structure and stable performance, and can stably execute the method for rapidly estimating the lithium content in clay of the present invention, further improving the overall applicability and practical application capability of the present invention.

[0050] In an embodiment, the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc. An input device may be used to obtain data information. An output device may be used to output the results obtained by storing program instructions contained in a computer program in a memory provided by the present invention. The memory may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include a non-volatile random access memory.

[0051] In yet another alternative embodiment, see Figure 3 In order to efficiently execute the method for quickly estimating the lithium content in clay provided by the present invention, this embodiment also provides a device for quickly estimating the lithium content in clay, such as Figure 3 As shown, including: The memory 10 is used to store computer programs; the processor 20 is used to execute the computer programs to implement the above-mentioned method for quickly estimating the lithium content in clay. The memory 10, the processor 20, the communication interface 31 and the communication bus 32. The memory 10, the processor 20 and the communication interface 31 all communicate with each other through the communication bus 32.

[0052] In an embodiment, the memory 10 is used to store one or more program instructions. The memory 10 may store program instructions for implementing the following functions: Obtain spectral reflectance data and corresponding lithium content of clay samples; obtain spectral data of multiple spectral resolutions based on the spectral reflectance data, and obtain the optimal spectral resolution by combining the spectral data and the lithium content; obtain multiple sets of fractional derivative spectra based on the optimal spectral resolution, and obtain a characteristic spectral data set through the correlation between the fractional derivative spectra and the lithium content; construct a rapid estimation model for lithium content, use the characteristic spectral data set to train and verify the rapid estimation model for lithium content, and complete the rapid estimation of lithium content in clay through the trained rapid estimation model for lithium content.

[0053] In one possible implementation, the memory 10 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, and at least one application required for a function, etc.; the data storage area may store data created during use. In addition, the memory 10 may include a read-only memory and a random access memory, and provide instructions and data to the processor. A portion of the memory may also include NVRAM. The memory stores an operating system and operating instructions, executable modules or data structures, or a subset thereof, or an extended set thereof, wherein the operating instructions may include various operating instructions for implementing various operations. The operating system may include various system programs for implementing various basic tasks and processing hardware-based tasks.

[0054] The processor 20 may be a central processing unit (CPU), an application specific integrated circuit, a digital signal processor, a field programmable gate array or other programmable logic device, and the processor 20 may be a microprocessor or any conventional processor, etc. The processor 20 may call a program stored in the memory 10. The communication interface 31 may be an interface of a communication module, and is used to connect with other devices or systems.

[0055] Of course, it should be noted that Figure 3 The structure shown does not constitute a limitation on the device for rapidly estimating the lithium content in clay in this embodiment. In practical applications, the device for rapidly estimating the lithium content in clay may include: Figure 3 More or fewer components than shown, or combinations of certain components.

[0056] The embodiment also provides a storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method for rapidly estimating the lithium content in clay are implemented.

[0057] The storage medium may include: a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

[0058] In summary, the present invention determines the optimal spectral resolution of spectral data through window sampling, and then divides the characteristic spectral data set under multiple fractional derivative spectra, and then extracts the characteristic spectral features and uses the lithium content rapid estimation model to complete the rapid estimation of the lithium content in clay, which improves the applicability of the present invention and solves the problem of quickly, efficiently and accurately estimating the lithium content in clay.

[0059] Therefore, the present invention effectively overcomes various shortcomings in the prior art and has high industrial utilization value.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope recorded in the present invention.

Claims

1. A method for quickly estimating lithium content in clay, characterized in that: The method for quickly estimating the lithium content in the clay comprises the following steps: Obtain spectral reflectance data and corresponding lithium content of clay samples; According to the spectral reflectance data, spectral data of multiple spectral resolutions are obtained, and the optimal spectral resolution is obtained by combining the spectral data and the lithium content; Obtaining multiple groups of fractional derivative spectra based on the optimal spectral resolution, and obtaining a characteristic spectral data set through the correlation between the fractional derivative spectra and the lithium content; A lithium content rapid estimation model is constructed, the characteristic spectral data set is used to train and verify the lithium content rapid estimation model, and the lithium content in the clay is rapidly estimated by the trained lithium content rapid estimation model.

2. The method for rapidly estimating lithium content in clay according to claim 1, characterized in that: The step of obtaining spectral data of multiple spectral resolutions according to the spectral reflectance data comprises the following steps: Set the size of the sliding window; Based on the sliding window, the spectral reflectance data is divided to obtain spectral data of multiple spectral resolutions.

3. The method for rapidly estimating lithium content in clay according to claim 1, characterized in that: The step of combining the spectral data and the lithium content to obtain an optimal spectral resolution comprises the following steps: According to the spectral data and the lithium content, a spectral data training set and a spectral data verification set are established; Constructing a machine classification model, training the machine classification model using the spectral data training set, and combining the spectral data validation set and the trained machine classification model to obtain a validation result; The verification results were comprehensively analyzed using the determination coefficient, relative root mean square error and relative analytical error to obtain the optimal spectral resolution.

4. The method for rapidly estimating lithium content in clay according to claim 1, characterized in that: The method of obtaining multiple groups of fractional-order derivative spectra based on the optimal spectral resolution comprises the following steps: Set the step interval; According to the step interval, fractional derivative spectra of order 0 to 2 are obtained.

5. The method for rapidly estimating lithium content in clay according to claim 4, characterized in that: According to the step interval, the fractional derivative spectrum of order 0 to 2 is obtained, which satisfies the following formula: in, Represents a function The a-order fractional derivative of , represents the order of the fractional order, Indicates the band is The spectral reflectance at , Indicates the band position, represents the Gamma function, Represents the spectral reflectance at wavelength t.

6. The method for rapidly estimating lithium content in clay according to claim 1, characterized in that: The step of obtaining a characteristic spectrum data set by the correlation between the fractional derivative spectrum and the lithium content comprises the following steps: Calculating the correlation between the fractional derivative spectrum and the lithium content by a first correlation model; Screening the fractional derivative spectra based on the correlation to obtain a significantly correlated spectral data set; A significant characteristic spectral band is extracted from the significantly correlated spectral data set, and then a characteristic spectral data set is constructed.

7. The method for rapidly estimating lithium content in clay according to claim 6, characterized in that: The first correlation model satisfies the following formula: in, represents the first correlation, represents the number of samples, express The reflectance of the samples, expressed as Indicates The lithium content of each sample is represents the band mean of all samples, Represents the mean lithium content of all samples.

8. The method for rapidly estimating lithium content in clay according to claim 6, characterized in that: The fractional derivative spectra are screened based on the correlation to obtain a significantly correlated spectral data set that satisfies the following formula: in, Indicates the first correlation.

9. The method for rapidly estimating lithium content in clay according to claim 6, characterized in that: The step of extracting significant characteristic spectral bands from the significant correlation spectral data set comprises the following steps: Extracting band features from the significantly correlated spectral data set; Constructing a band feature importance assessment model, and using the band feature importance assessment model to obtain the importance of the band feature; According to the importance, significant characteristic spectral bands are obtained.

10. The method for rapidly estimating lithium content in clay according to claim 9, characterized in that: The band feature importance evaluation model satisfies the following formula: in, Indicates The band characteristics are Importance on the dataset, Indicates sample data that has not been used to train a specific decision branch. Indicates Branches The number of samples of data, Indicates the number of branches, represents the indicator function, Indicates The actual result of samples, Indicates The sample The prediction result of each branch, After random permutation Branch pairs Data The prediction results of samples.

Images