Remote sensing ecological index calculation method and device, electronic equipment and medium

Abstract

The invention provides a remote sensing ecological index calculation method and device, electronic equipment and a medium, and the method comprises the steps: obtaining satellite remote sensing product data, ERA5 reanalysis data and an initial remote sensing ecological index of a target region, and dividing the satellite remote sensing product data and the ERA5 reanalysis data into evaluation factors of a plurality of classes according to a preset class; based on the initial remote sensing ecological index and the evaluation factor of each category, adopting a random forest machine learning model to determine a target evaluation factor of each category; performing principal component analysis on the target evaluation factor of each category to obtain a first principal component factor of each category; performing principal component analysis on the first principal component factor of each category to obtain a weight coefficient of each category; and performing weighted summation calculation on the first principal component factor and the weight coefficient of each category to obtain a remote sensing ecological index of the target area. According to the method, the regional inapplicability caused by fixed factors can be effectively avoided, and the accuracy and representativeness of a result are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of remote sensing ecological index calculation, and in particular to a remote sensing ecological index calculation method, device, electronic equipment and medium. Background Art

[0002] The remote sensing ecological index is an ecological environment evaluation index based on remote sensing data. It extracts information related to the ecological environment by integrating multiple ecological factors, and objectively evaluates and monitors the ecological status and environmental quality of a specific region. In the existing technology, the coupling scheme of four evaluation factors, such as dryness index, humidity index, normalized vegetation index and surface temperature, is the most commonly used. On the basis of this scheme, ecological factors representing the region are added and applied to specific research areas. However, this type of solution is too subjective in the selection of factors, resulting in the calculated remote sensing ecological index not being able to truly reflect the ecological status and environmental quality. Summary of the invention

[0003] In view of this, the purpose of the present invention is to provide a remote sensing ecological index calculation method, device, electronic equipment and medium, which can effectively avoid the regional inapplicability caused by fixed factors, while improving the accuracy and representativeness of the results.

[0004] In order to achieve the above object, the technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present invention provides a method for calculating a remote sensing ecological index, comprising: obtaining satellite remote sensing product data, ERA5 reanalysis data and an initial remote sensing ecological index of a target area, and dividing the satellite remote sensing product data and the ERA5 reanalysis data into multiple categories of evaluation factors according to preset categories; based on the initial remote sensing ecological index and the evaluation factor of each category, using a random forest machine learning model to determine the target evaluation factor of each category; performing principal component analysis on the target evaluation factor of each category to obtain the first principal component factor of each category; performing principal component analysis on the first principal component factor of each category to obtain the weight coefficient of the first principal component factor of each category; performing weighted sum calculation on the first principal component factor of each category and the weight coefficient to obtain the remote sensing ecological index of the target area.

[0006] In one embodiment, based on the initial remote sensing ecological index and the evaluation factor of each category, before using the random forest machine learning model to determine the target evaluation factor of each category, it also includes: resampling the evaluation factor of each category based on the spatial scale of the initial remote sensing ecological index to obtain the evaluation factor of each category with the same spatial scale as the initial remote sensing ecological index.

[0007] In one embodiment, based on the initial remote sensing ecological index and the evaluation factor of each category, a random forest machine learning model is used to determine the target evaluation factor of each category, including: constructing a sample set based on the initial remote sensing ecological index of the same spatial scale and the evaluation factor of each category; based on the sample set, a remote sensing ecological index estimation model is constructed using a random forest machine learning model, and the importance score of the evaluation factor of each category is calculated; and the target evaluation factor of each category is determined based on the importance score of the evaluation factor of each category.

[0008] In one embodiment, the target evaluation factor of each category is determined based on the importance score of the evaluation factor of each category, including: sorting the evaluation factors of each category in descending order based on the importance score of the evaluation factors of each category, and selecting a preset number of evaluation factors as the target evaluation factors of each category based on the sorting results.

[0009] In one embodiment, principal component analysis is performed on the target evaluation factors of each category, including: standardizing the target evaluation factors of each category, and performing principal component analysis based on the standardized target evaluation factors of each category.

[0010] In one embodiment, after performing principal component analysis on the target evaluation factor of each category to obtain the first principal component factor of each category, it also includes: obtaining the contribution rate of the first principal component factor; if the contribution rate of the first principal component factor is less than the contribution rate threshold, removing the category corresponding to the first principal component factor.

[0011] In a second aspect, the present invention provides a remote sensing ecological index calculation device, comprising: a category classification module, used to obtain satellite remote sensing product data, ERA5 reanalysis data and initial remote sensing ecological index of a target area, and divide the satellite remote sensing product data and ERA5 reanalysis data into multiple categories of evaluation factors according to preset categories; a target evaluation factor determination module, used to determine the target evaluation factor of each category based on the initial remote sensing ecological index and the evaluation factor of each category, using a random forest machine learning model; a first principal component analysis module, used to perform principal component analysis on the target evaluation factor of each category to obtain the first principal component factor of each category; a second principal component analysis module, used to perform principal component analysis on the first principal component factor of each category to obtain the weight coefficient of the first principal component factor of each category; a remote sensing ecological index calculation module, used to perform weighted sum calculation on the first principal component factor and the weight coefficient of each category to obtain the remote sensing ecological index of the target area.

[0012] In one embodiment, it also includes: a resampling module, which is used to resample the evaluation factor of each category based on the spatial scale of the initial remote sensing ecological index to obtain the evaluation factor of each category with the same spatial scale as the initial remote sensing ecological index.

[0013] In a third aspect, the present invention provides an electronic device comprising a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the steps of any one of the methods provided in the first aspect above.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of any one of the methods provided in the first aspect are executed.

[0015] The present invention brings the following beneficial effects:

[0016] The above-mentioned remote sensing ecological index calculation method, device, electronic device and medium provided by the present invention first obtain the satellite remote sensing product data, ERA5 reanalysis data and initial remote sensing ecological index of the target area, and divide the satellite remote sensing product data and ERA5 reanalysis data into multiple categories of evaluation factors according to preset categories; secondly, based on the initial remote sensing ecological index and the evaluation factor of each category, the random forest machine learning model is used to determine the target evaluation factor of each category; then the target evaluation factor of each category is subjected to principal component analysis to obtain the first principal component factor of each category; then the first principal component factor of each category is subjected to principal component analysis to obtain the weight coefficient of the first principal component factor of each category; finally, the first principal component factor of each category and the weight coefficient are weighted and calculated to obtain the remote sensing ecological index of the target area. The above method first divides the evaluation factors into multiple categories, then selects the evaluation factors using the random forest machine model, and determines the first principal component factor and weight coefficient of each category through principal component analysis, and finally obtains the final remote sensing ecological index through weighted calculation, thereby effectively avoiding the regional inapplicability caused by fixed factors, and at the same time improving the accuracy and representativeness of the results.

[0017] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0018] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0020] Figure 1 A flowchart of a remote sensing ecological index calculation method provided by an embodiment of the present invention;

[0021] Figure 2 A flowchart of another remote sensing ecological index calculation method provided by an embodiment of the present invention;

[0022] Figure 3 A schematic diagram of the structure of a remote sensing ecological index calculation device provided by an embodiment of the present invention;

[0023] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0025] At present, the most commonly used scheme for calculating remote sensing ecological index is the coupling scheme of four evaluation factors, namely dryness index, humidity index, normalized difference vegetation index and surface temperature. On this basis, ecological factors representing the region are added and applied to specific research areas. However, this type of solution is too subjective in factor selection, resulting in the calculated remote sensing ecological index not being able to truly reflect the ecological status and environmental quality.

[0026] Based on this, the remote sensing ecological index calculation method, device, electronic device and medium provided by the embodiments of the present invention can effectively avoid the regional inapplicability caused by fixed factors, while improving the accuracy and representativeness of the results.

[0027] To facilitate understanding of this embodiment, a remote sensing ecological index calculation method disclosed in an embodiment of the present invention is first introduced in detail. The method can be executed by an electronic device, such as a smart phone, a computer, a tablet computer, etc. Figure 1 The flowchart of a remote sensing ecological index calculation method shown in FIG. 1 illustrates that the method mainly includes the following steps S101 to S105:

[0028] Step S101: Obtain satellite remote sensing product data, ERA5 reanalysis data and initial remote sensing ecological index of the target area, and divide the satellite remote sensing product data and ERA5 reanalysis data into multiple categories of evaluation factors according to preset categories.

[0029] In one embodiment, satellite remote sensing product data, ERA5 reanalysis data and initial remote sensing ecological index (RSEI′) of the target area are first obtained, wherein the initial remote sensing ecological index can be an existing meteorological bureau product; then, the evaluation factors are divided into multiple categories according to preset categories, wherein the divided categories mainly include atmospheric environment element categories Meteorological and hydrological feature classes Vegetation feature class And other feature classes

[0030]

[0031] Wherein, N represents a non-zero natural number.

[0032] In specific implementation, the atmospheric environment element class may include:

[0033]

[0034] Vegetation feature classes can include:

[0035]

[0036] Meteorological and hydrological feature classes can include:

[0037]

[0038] Other feature classes can include:

[0039]

[0040] Among them, aerosol optical depth (AOD), normalized difference vegetation index (NDVI), net primary productivity (NPP), leaf area index (LAI), enhanced vegetation index (EVI), vegetation cover (VC), land surface temperature (LST), and land surface type data (LUCC) are MODIS sensor products; total ozone column (TO3), total nitrogen dioxide column (TNO2), and total sulfur dioxide column (TSO2) are OMI sensor products; temperature (T2M), precipitation (PRE), relative humidity (RH), surface downward solar radiation (SSRD), latent heat flux (LHF), evapotranspiration (EVA), and soil moisture (SM) are ERA5 reanalysis data products; DEM is SRTM (ShuttleRadar Topography Mission) data; POP is a product released by CIESIN (Center for International EarthScience Information Network).

[0041] Step S102: Based on the initial remote sensing ecological index and the evaluation factor of each category, a random forest machine learning model is used to determine the target evaluation factor of each category.

[0042] In one embodiment, a sample set can be constructed using the initial remote sensing ecological index and the evaluation factors of each category, and the sample set can be used to build a remote sensing ecological index estimation model using a random forest machine learning model to calculate the importance of all evaluation factors of each category, and then select the target evaluation factor of each category based on the importance.

[0043] Step S103: Perform principal component analysis on the target evaluation factor of each category to obtain the first principal component factor of each category.

[0044] In one embodiment, principal component analysis is a statistical method that aims to transform a set of variables that may be linearly correlated into a set of linearly uncorrelated variables through orthogonal transformation. These new variables are called principal components. The core idea of ​​principal component analysis is to transform multiple variables into a few comprehensive indicators through dimensionality reduction technology. These indicators can reflect most of the information of the original variables, while reducing the dimension of the data to facilitate further analysis and processing. Specifically, principal component analysis can include the following steps:

[0045] (1) Data standardization: First, the original data is standardized to ensure that each variable has a similar scale.

[0046] (2) Calculate the covariance matrix: Calculate the covariance matrix between variables.

[0047] (3) Eigendecomposition: Perform eigendecomposition on the covariance matrix to obtain eigenvalues ​​and eigenvectors.

[0048] (4) Select principal components: Select principal components based on the size of the eigenvalues. Usually, several eigenvectors with larger eigenvalues ​​are selected as principal components.

[0049] (5) Calculate the principal component scores: Use the selected eigenvectors to calculate the scores of each sample on each principal component.

[0050] (6) Interpret principal components: Interpret the meaning of each principal component based on the coefficients of the eigenvector to ensure that they can reasonably explain the main characteristics of the original data.

[0051] Based on this, in the embodiment of the present invention, the first principal component factor of each category can be obtained by performing principal component analysis on the target evaluation factor of each category.

[0052] Step S104: performing principal component analysis on the first principal component factor of each category to obtain the weight coefficient of the first principal component factor of each category.

[0053] In one embodiment, principal component analysis is performed again on the first principal component factor of each category calculated in step S103 to calculate the first principal component of the first principal component factor of each category, and the coefficient of the first principal component is determined as the weight coefficient of the first principal component factor of each category.

[0054] Step S105: Perform weighted sum calculation on the first principal component factor and the weight coefficient of each category to obtain the remote sensing ecological index of the target area.

[0055] In one implementation, the first principal component factor and the weight coefficient of each category are weighted and summed to obtain the remote sensing ecological index of the target area. Specifically, the remote sensing ecological index can be calculated according to the following formula:

[0056]

[0057] Among them, RSEI represents the remote sensing ecological index, r represents the traversal [E atm ,E wea ,E veg ,E oth ], PC1 r The first principal component factor corresponding to each category (Atmospheric environment elements), (meteorological and hydrological element classes), (vegetation feature class), (other feature classes), W r Represents the weight coefficient of the first principal component factor for each category.

[0058] The above-mentioned remote sensing ecological index calculation method provided by the embodiment of the present invention first divides the evaluation factors into multiple categories, then uses the random forest machine model to select the evaluation factors, and determines the first principal component factors and weight coefficients of each category through principal component analysis, and finally obtains the final remote sensing ecological index through weighted calculation, thereby effectively avoiding the regional inapplicability caused by fixed factors, while improving the accuracy and representativeness of the results.

[0059] In order to improve the accuracy of the results, in an embodiment of the present invention, before using a random forest machine learning model to determine the target evaluation factor of each category based on the initial remote sensing ecological index and the evaluation factor of each category, it also includes: resampling the evaluation factor of each category based on the spatial scale of the initial remote sensing ecological index to obtain an evaluation factor for each category with the same spatial scale as the initial remote sensing ecological index.

[0060] In the specific implementation, based on the initial remote sensing ecological index, the evaluation factors of all categories at the same time scale are resampled to the same spatial scale as the initial remote sensing ecological index to obtain the evaluation factors of each category with the same spatial scale as the initial remote sensing ecological index.

[0061] In one embodiment, for the aforementioned step S102, that is, when the target evaluation factor of each category is determined by using the random forest machine learning model based on the initial remote sensing ecological index and the evaluation factor of each category, the following methods may be used, including but not limited to:

[0062] First, a sample set was constructed based on the initial remote sensing ecological index of the same spatial scale and the evaluation factor for each category.

[0063] In the specific implementation, the sample set Q is constructed based on all the evaluation factors F and the initial remote sensing ecological index RSEI′ extracted at the same time scale and the same spatial scale. Q={RSEI′(t,m,n),F(t,m,n)∣t∈[t 1 ,t 2 ],m∈{1,2,…,M},n∈{1,2,…,N}.

[0064] Specifically, Q is the time t 1 to 2 The sample set of valid pixel points in the M×N grid, RSEI′(t,m,n) represents the initial remote sensing ecological index of the valid pixel point (m,n) at time t, F(t,m,n) represents the evaluation factor of the valid pixel point (m,n) at time t, and the M×N grid is the grid corresponding to the target area. For the specific division method, please refer to the existing grid division method.

[0065] Then, based on the sample set, the random forest machine learning model was used to construct a remote sensing ecological index estimation model, and the importance score of the evaluation factor for each category was calculated.

[0066] In the specific implementation, the above sample set Q is used to build a remote sensing ecological index estimation model using a random forest machine learning model to calculate the importance scores of all evaluation factors in each category. Specifically, the sample set can be used to train the random forest machine learning model, and the importance scores of the evaluation factors of each category can be calculated based on the trained model.

[0067] Finally, the target evaluation factor for each category is determined based on the importance score of the evaluation factor for each category.

[0068] In specific implementation, the evaluation factors of each category may be sorted in descending order based on the importance scores of the evaluation factors of each category, and a preset number of evaluation factors may be selected as target evaluation factors of each category based on the sorting results.

[0069] Specifically, the importance scores of the evaluation factors of each category are defined as: I(ATM(m)), I(WEA(n)), I(VEG(p)), I(OTH(q)), and the evaluation factors in each category are sorted in descending order according to the importance score I, and the top three evaluation factors in each category are selected as the target evaluation factors of each category. Assume that the index after sorting is Corresponding to m1, m2, ...; Corresponding to n1, n2, ...; Corresponding to p1, p2, ...; Corresponding to q1, q2, ..., the first three important evaluation factors are selected as:

[0070] Top3 atm ={ATM(m1)ATM(m2),ATM(m3)}

[0071] Top3 wea ={WEA(n1),WEA(n2),WEA(n3)}

[0072] Top3 veg ={VEG(p1),VEG(p2),VEG(p3)}

[0073] Top3 oth ={OTH(q1),OTH(q2),OTH(q3)}

[0074] Among them, Top 3 atm The target evaluation factor of the atmospheric environment element class, Top3 wea is the target evaluation factor of meteorological and hydrological element class, Top3veg is the target evaluation factor of the vegetation feature class, Top3 oth It is the target evaluation factor of other feature classes.

[0075] Since the dimensions of the evaluation factors of each category are different, directly performing principal component analysis will cause an imbalance in the indicator weights, so the data needs to be standardized to unify its dimensions to between [0, 1]. Specifically, for the above step S103, that is, when performing principal component analysis on the target evaluation factors of each category, the following methods can be adopted, including but not limited to: standardizing the target evaluation factors of each category, and performing principal component analysis based on the standardized target evaluation factors of each category.

[0076] In specific implementation, E = Top3 atm ∪Top3 wea ∪Top3 veg ∪Top3 oth It represents the standardized target evaluation factor of each category. The standardized formula of each target evaluation factor is:

[0077]

[0078] in, It represents the index value at pixel x after the evaluation factor y is standardized. It represents the index value of the evaluation factor y at pixel x. Indicates the minimum value of the evaluation factor y in the image, Indicates the maximum value of the evaluation factor y in the image.

[0079] In one embodiment, principal component analysis is used to calculate the first principal component factor of the first three important target evaluation factors of each category, namely:

[0080]

[0081] in, is the first principal component factor of the atmospheric environment element class, is the first principal component factor of the meteorological and hydrological element class, is the first principal component factor of the vegetation feature class, It is the first principal component factor of other feature classes.

[0082] Furthermore, after performing principal component analysis on the target evaluation factors of each category to obtain the first principal component factor of each category, it also includes: obtaining the contribution rate of the first principal component factor; if the contribution rate of the first principal component factor is less than the contribution rate threshold, removing the category corresponding to the first principal component factor.

[0083] In specific implementation, when performing principal component analysis on the target evaluation factors of each category, the contribution rate of the first principal component factor of each category can be obtained, and it is ensured that the contribution rate of the first principal component factor of each category is greater than the contribution rate threshold (85%), otherwise the evaluation factor of this category will be removed.

[0084] For ease of understanding, the present invention also provides another remote sensing ecological index calculation method, see Figure 2 As shown, it mainly includes the following steps S201 to S205:

[0085] Step S201: presetting multiple categories of evaluation factors.

[0086] Specifically, using satellite remote sensing product data and ERA5 reanalysis data, multiple categories of evaluation factors are preset, and the categories mainly include atmospheric environment elements, meteorological and hydrological elements, vegetation elements, and other elements. Then, based on the initial remote sensing ecological index, all the evaluation factors of the same time scale are resampled to the same spatial scale as the initial remote sensing ecological index.

[0087] Step S202: Random forest factor importance calculation.

[0088] Specifically, all the evaluation factors F and the initial remote sensing ecological index at the same time and space scales were extracted to construct a sample set. Using this sample set, a remote sensing ecological index estimation model was built using a random forest machine learning model, and the factor importance of all elements in each category was calculated, and the top three evaluation factors in each category were extracted.

[0089] Step S203: factor standardization processing.

[0090] Specifically, the top three evaluation factors extracted are standardized and their dimensions are unified to [0, 1].

[0091] Step S204: Calculate the first principal component of each category.

[0092] Specifically, principal component analysis is used in combination with the standardized evaluation factors in step S203 to calculate the first principal component of the first three important evaluation factors of each category, and principal component analysis is performed again on the first principal component of each category to calculate the first principal component, and the four coefficients of the first principal component are determined as the weight coefficients corresponding to each category.

[0093] Step S205: weighted sum to obtain the remote sensing ecological index.

[0094] Specifically, the weight coefficient obtained by principal component analysis is combined with the first principal component of the first three important evaluation factors of each category to perform weighted summation to obtain the final remote sensing ecological index.

[0095] The above method provided by the embodiment of the present invention can calculate the regional remote sensing ecological index under non-human intervention conditions by sorting the importance of factors of preset categories and performing two principal component analyses. Compared with the existing technology, it can not only effectively avoid the regional inapplicability caused by fixed factors, but also improve the accuracy and representativeness of the results.

[0096] For the remote sensing ecological index calculation method provided in the above embodiment, the present invention also provides a remote sensing ecological index calculation device, see Figure 3 The schematic diagram of the structure of a remote sensing ecological index calculation device is shown, indicating that the device mainly includes the following parts:

[0097] The classification module 301 is used to obtain the satellite remote sensing product data, ERA5 reanalysis data and initial remote sensing ecological index of the target area, and classify the satellite remote sensing product data and ERA5 reanalysis data into multiple categories of evaluation factors according to preset categories;

[0098] A target evaluation factor determination module 302 is used to determine the target evaluation factor of each category using a random forest machine learning model based on the initial remote sensing ecological index and the evaluation factor of each category;

[0099] The first principal component analysis module 303 is used to perform principal component analysis on the target evaluation factor of each category to obtain the first principal component factor of each category;

[0100] The second principal component analysis module 304 is used to perform principal component analysis on the first principal component factor of each category to obtain the weight coefficient of the first principal component factor of each category;

[0101] The remote sensing ecological index calculation module 305 is used to perform weighted sum calculation on the first principal component factor and the weight coefficient of each category to obtain the remote sensing ecological index of the target area.

[0102] The above-mentioned remote sensing ecological index calculation device provided by the embodiment of the present invention first divides the evaluation factors into multiple categories, then uses the random forest machine model to select the evaluation factors, and determines the first principal component factors and weight coefficients of each category through principal component analysis, and finally obtains the final remote sensing ecological index through weighted calculation, thereby effectively avoiding the regional inapplicability caused by fixed factors, while improving the accuracy and representativeness of the results.

[0103] In one embodiment, the above-mentioned device also includes: a resampling module, which is used to resample the evaluation factor of each category based on the spatial scale of the initial remote sensing ecological index to obtain the evaluation factor of each category with the same spatial scale as the initial remote sensing ecological index.

[0104] In one embodiment, the above-mentioned target evaluation factor determination module 302 is specifically used to: construct a sample set based on the initial remote sensing ecological index of the same spatial scale and the evaluation factor of each category; based on the sample set, a remote sensing ecological index estimation model is constructed using a random forest machine learning model, and the importance score of the evaluation factor of each category is calculated; based on the importance score of the evaluation factor of each category, the target evaluation factor of each category is determined.

[0105] In one embodiment, the target evaluation factor determination module 302 is specifically used to sort the evaluation factors of each category in descending order based on the importance scores of the evaluation factors of each category, and select a preset number of evaluation factors as the target evaluation factors of each category based on the sorting results.

[0106] In one implementation, the first principal component analysis module 303 is specifically used to: perform standardization processing on the target evaluation factors of each category, and perform principal component analysis based on the standardized target evaluation factors of each category.

[0107] In one implementation, the first principal component analysis module 303 is specifically used to: obtain the contribution rate of the first principal component factor; if the contribution rate of the first principal component factor is less than a contribution rate threshold, remove the category corresponding to the first principal component factor.

[0108] It should be noted that the implementation principle and technical effects of the device provided in the embodiment of the present invention are the same as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the device embodiment, reference may be made to the corresponding contents in the aforementioned method embodiment.

[0109] An embodiment of the present invention further provides an electronic device. Specifically, the electronic device includes a processor and a storage device. The storage device stores a computer program, and when the computer program is executed by the processor, it executes the method described in any one of the above implementation methods.

[0110] Figure 4 A structural diagram of an electronic device provided in an embodiment of the present invention, the electronic device 100 includes: a processor 40, a memory 41, a bus 42 and a communication interface 43, wherein the processor 40, the communication interface 43 and the memory 41 are connected via the bus 42; the processor 40 is used to execute an executable module stored in the memory 41, such as a computer program.

[0111] The memory 41 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk memory. The communication connection between the system network element and at least one other network element is realized through at least one communication interface 43 (which may be wired or wireless), and the Internet, wide area network, local area network, metropolitan area network, etc. may be used.

[0112] The bus 42 may be an ISA bus, a PCI bus, or an EISA bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0113] Among them, the memory 41 is used to store programs, and the processor 40 executes the program after receiving the execution instruction. The method executed by the device for flow process definition disclosed in any embodiment of the above-mentioned embodiment of the present invention can be applied to the processor 40 or implemented by the processor 40.

[0114] The processor 40 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 40. The above processor 40 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present invention can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in the embodiments of the present invention can be directly embodied as a hardware decoding processor to execute, or can be executed by a combination of hardware and software modules in the decoding processor. The software module can be located in a mature storage medium in the field such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 41, and the processor 40 reads the information in the memory 41 and completes the steps of the above method in combination with its hardware.

[0115] The computer program product of the readable storage medium provided in the embodiment of the present invention includes a computer-readable storage medium storing program code, and the instructions included in the program code can be used to execute the methods described in the previous method embodiments. The specific implementation can be referred to the previous method embodiments, which will not be repeated here.

[0116] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0117] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-mentioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-mentioned embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A remote sensing ecological index calculation method, characterized in that: include: Obtain satellite remote sensing product data, ERA5 reanalysis data and initial remote sensing ecological index of the target area, and classify the satellite remote sensing product data and the ERA5 reanalysis data into multiple categories of evaluation factors according to preset categories; Based on the initial remote sensing ecological index and the evaluation factor of each category, a random forest machine learning model is used to determine the target evaluation factor of each category; Performing principal component analysis on the target evaluation factor of each category to obtain the first principal component factor of each category; Perform principal component analysis on the first principal component factor of each category to obtain the weight coefficient of the first principal component factor of each category; The first principal component factor and the weight coefficient of each category are weighted and summed to obtain the remote sensing ecological index of the target area.

2. The method according to claim 1, characterized in that: Based on the initial remote sensing ecological index and the evaluation factor of each category, before using the random forest machine learning model to determine the target evaluation factor of each category, it also includes: The evaluation factor of each category is resampled based on the spatial scale of the initial remote sensing ecological index to obtain the evaluation factor of each category with the same spatial scale as the initial remote sensing ecological index.

3. The method according to claim 1, characterized in that: Based on the initial remote sensing ecological index and the evaluation factor of each category, a random forest machine learning model is used to determine the target evaluation factor of each category, including: Constructing a sample set based on the initial remote sensing ecological index of the same spatial scale and the evaluation factor of each category; Based on the sample set, a random forest machine learning model is used to construct a remote sensing ecological index estimation model, and the importance score of the evaluation factor of each category is calculated; A target evaluation factor for each category is determined based on the importance score of the evaluation factor for each category.

4. The method according to claim 3, characterized in that Determining a target evaluation factor for each category based on the importance score of the evaluation factor for each category includes: The evaluation factors of each category are sorted in descending order based on the importance scores of the evaluation factors of each category, and a preset number of evaluation factors are selected as target evaluation factors of each category based on the sorting results.

5. The method according to claim 1, characterized in that The principal component analysis is performed on the target evaluation factors of each category, including: The target evaluation factor of each category is standardized, and a principal component analysis is performed based on the standardized target evaluation factor of each category.

6. The method according to claim 1, characterized in that After performing principal component analysis on the target evaluation factor of each category to obtain the first principal component factor of each category, the method further includes: Obtaining the contribution rate of the first principal component factor; If the contribution rate of the first principal component factor is less than the contribution rate threshold, the category corresponding to the first principal component factor is removed.

7. A remote sensing ecological index calculation device, characterized in that: include: A classification module is used to obtain satellite remote sensing product data, ERA5 reanalysis data and initial remote sensing ecological index of the target area, and classify the satellite remote sensing product data and the ERA5 reanalysis data into multiple categories of evaluation factors according to preset categories; A target evaluation factor determination module, used to determine the target evaluation factor of each category using a random forest machine learning model based on the initial remote sensing ecological index and the evaluation factor of each category; A first principal component analysis module, used to perform principal component analysis on the target evaluation factor of each category to obtain a first principal component factor of each category; The second principal component analysis module is used to perform principal component analysis on the first principal component factor of each category to obtain the weight coefficient of the first principal component factor of each category; The remote sensing ecological index calculation module is used to perform weighted sum calculation on the first principal component factor and the weight coefficient of each category to obtain the remote sensing ecological index of the target area.

8. The device according to claim 7, characterized in that Also includes: The resampling module is used to resample the evaluation factor of each category based on the spatial scale of the initial remote sensing ecological index to obtain the evaluation factor of each category with the same spatial scale as the initial remote sensing ecological index.

9. An electronic device, characterized in that: The method comprises a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the steps of the method according to any one of claims 1 to 6.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are executed.

Images