Geographically weighted comprehensive index-based vegetation dynamic comprehensive index construction method

Abstract

The invention provides a vegetation dynamic comprehensive index construction method based on a geographically weighted comprehensive index, which comprises the following steps: acquiring monthly global land surface satellite remote sensing product data of a research area, and preprocessing the global land surface satellite remote sensing product data to obtain preprocessed remote sensing data; calculating land coverage partition weights for the preprocessed remote sensing data according to land coverage types; constructing a continuous weight surface of the land coverage partition weight by using spatial interpolation; and carrying out pixel-by-pixel dynamic weight fusion in the research area based on the continuous weight surface to obtain a vegetation dynamic comprehensive index. According to the method, the global trend and local adaptation are combined, the overall change trend of the vegetation growth condition in the macroscopic level is considered, the unique characteristics and spatial heterogeneity of vegetation growth in different areas can be fully reflected, the defect of one-step cutting in a traditional global model is avoided, and the method has the advantages of being high in robustness and high in robustness. The limitation of a traditional method in the aspects of spatial heterogeneity and ecological interpretation is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of agricultural remote sensing technology, and particularly relates to a vegetation dynamic comprehensive index construction method based on a geographical weighted comprehensive index. BACKGROUND

[0002] As the core component of the terrestrial ecosystem, vegetation drives the global carbon-water cycle through photosynthesis, and is the cornerstone of maintaining biodiversity, regulating climate and supporting the sustainable development of human society. Its growth status not only directly affects the primary productivity and carbon sink capacity of the ecosystem, but also participates in climate feedback by regulating surface albedo and evapotranspiration. With the continuous change of global climate, vegetation dynamic monitoring is crucial to understanding the structure and function of the ecosystem and formulating sustainable ecosystem development policies.

[0003] Traditional vegetation monitoring relies on ground observations, but is difficult to achieve continuous assessment on a large scale due to the sparsity of sites, high cost and spatial heterogeneity. With the continuous development of satellite remote sensing technology, significant progress has been made in the study of regional and even global vegetation growth conditions. Since the AVHRR sensor first provided global NDVI data in the 1980s, multispectral remote sensing has derived a series of indicators such as LAI (leaf area index), FVC (vegetation coverage), and NPP (net primary productivity), which respectively depict vegetation dynamics from the dimensions of greenness, canopy structure, coverage and carbon sink function, and each has its own advantages. However, due to the significant differences in physical meaning and sensitivity of different indices, it is difficult to comprehensively analyze the coordinated evolution of ecosystem structure and function by relying on a single index to evaluate vegetation dynamics: NDVI is widely used to represent vegetation greenness, but it is easily saturated in high biomass areas and insufficiently responsive to non-photosynthetic components (such as litter), which may underestimate the true biomass accumulation of forest canopies; EVI can alleviate this problem, but it is sensitive to atmospheric aerosol interference; LAI directly reflects the photosynthetic potential of the canopy, but it is disturbed by soil background in sparse vegetation areas; FVC can effectively monitor vegetation coverage in arid regions, but it cannot distinguish between functional types of vegetation. This multi-dimensional functional differentiation leads to the possibility of one-sided or even misleading representation of vegetation changes by a single index.

[0004] In recent years, multi-parameter collaborative analysis has gradually emerged and achieved remarkable results in vegetation growth detection. However, existing researches are mostly based on different data products, with significant differences in processing methods, spatial and temporal resolutions, and verification standards. Moreover, due to the influence of cloud pollution, some data are missing in space and time, increasing the error of different research results. The Global LAnd Surface Satellite (GLASS) product, independently developed by Beijing Normal University, is a long-time series, high-resolution, and high-precision remote sensing inversion product based on multi-source remote sensing data and measured site data through a unified algorithm framework. This product includes LAI, FVC, NPP, and other vegetation parameters, providing a new opportunity for comprehensive utilization of multi-vegetation indicators to understand vegetation dynamics. In addition, vegetation growth is strongly driven by local environment (such as climate, soil, and human activities), resulting in high spatial and temporal heterogeneity of vegetation parameters. To obtain a more comprehensive evaluation of vegetation growth, ecological mechanisms and spatial-temporal dynamics need to be combined.

[0005] Currently, there are three main methods for multi-parameter collaborative analysis of vegetation growth conditions: data fusion, statistical analysis, and model-based integration. The data fusion method combines different resolutions and types of data, such as high-resolution optical images and low-resolution microwave data, or remote sensing images combined with GIS data, ground monitoring data, and other data to obtain more comprehensive and accurate vegetation information. The statistical analysis method uses principal component analysis (PCA), cluster analysis, correlation analysis, and other statistical methods to comprehensively analyze multiple vegetation parameters and related environmental factors, extract main comprehensive indicators, or reveal the internal relationship between variables, thereby more deeply understanding vegetation growth conditions and their relationship with environmental factors. The model-based integration scheme constructs ecological models, biophysical models, and other models, taking multiple vegetation parameters and environmental factors as input variables to simulate vegetation growth processes and dynamic changes. Through the simulation results of the model, the vegetation growth conditions are evaluated, and the comprehensive influence of each factor on vegetation growth is analyzed.

[0006] The data obtained from different data sources may have resolution differences, including spatial resolution, temporal resolution and spectral resolution, which increases the difficulty of data fusion and collaborative analysis. In addition, due to the influence of cloud pollution, part of the data is missing in space and time, which affects the continuous monitoring of the long-term dynamic change of vegetation. For different vegetation parameters, the spatial distribution and change trend are obviously different, which makes it difficult to directly establish a unified spatial analysis framework and model when performing multi-parameter collaborative analysis, increasing the uncertainty of multi-parameter comprehensive analysis. Finally, some ecological and physical models often need to rely on data assumptions for pre-training, such as linear assumption, normal distribution assumption, etc., while the relationship between vegetation parameters in actual conditions may be more complex, which limits the applicability and accuracy of the model. SUMMARY

[0007] The application provides a vegetation dynamic comprehensive index construction method based on a geographical weighted comprehensive index, which is used to solve the defects of single vegetation index in evaluating the long-term growth of vegetation. Based on the GLASS long-time series vegetation parameter data set, the spatial heterogeneity modeling and multi-variable dynamic weighting are combined by introducing a geographical weighting mechanism to realize the combination of "global trend and local adaptation". The method not only considers the overall change trend of vegetation growth at the macro level, but also fully reflects the unique characteristics and spatial heterogeneity of vegetation growth in different regions, avoiding the disadvantages of "one-size-fits-all" in traditional global models and solving the limitations of traditional methods in spatial heterogeneity and ecological interpretation.

[0008] In a first aspect, the application provides a vegetation dynamic comprehensive index construction method based on a geographical weighted comprehensive index, comprising: Obtaining monthly global land satellite remote sensing product data of a study area, preprocessing the global land satellite remote sensing product data to obtain preprocessed remote sensing data; According to the land cover type, calculating the land cover partition weight of the preprocessed remote sensing data; Using spatial interpolation, constructing a continuous weight surface of the land cover partition weight; In the study area, based on the continuous weight surface, performing pixel-by-pixel dynamic weighted fusion to obtain a vegetation dynamic comprehensive index.

[0009] According to the vegetation dynamic comprehensive index construction method based on the geographical weighted comprehensive index provided by the application, the monthly global land satellite remote sensing product data of a study area is obtained, the global land satellite remote sensing product data is preprocessed to obtain preprocessed remote sensing data, which comprises: The HDF image of the study area is acquired, and the HDF image is converted, mosaicked, cropped and projected to obtain a TIF image with a preset temporal resolution and a preset spatial resolution. The TIF images were synthesized using monthly maximum data, and the true results of each vegetation parameter in the study area were obtained after image scaling and data quality control. The land cover data was determined using a global land cover classification dataset, which was integrated into multiple land type data.

[0010] According to the present invention, a method for constructing a vegetation dynamic comprehensive index based on a geographic weighted comprehensive index is provided, which calculates land cover zoning weights on the preprocessed remote sensing data according to land cover type, including: The preprocessed remote sensing data is normalized using maximum and minimum values. Multiple months are processed independently in the time dimension to capture phenological differences. In the spatial dimension, analysis units are constructed based on multiple types of annual land cover zoning to form a three-dimensional data matrix. Based on the land cover type, extract the vegetation parameter pixels corresponding to each category, and use the CRITIC method to calculate the weight of each vegetation parameter for all pixels within each land cover partition.

[0011] According to the present invention, a method for constructing a vegetation dynamic comprehensive index based on a geographic weighted comprehensive index is provided. Based on land cover type, vegetation parameter pixels corresponding to each category are extracted. For all pixels within each land cover partition, the weight of each vegetation parameter is calculated using the CRITIC method, including: Calculate the standard deviation of each parameter. The standard deviation represents the variation and fluctuation of values ​​within each indicator. The larger the standard deviation, the greater the numerical variation of the indicator. Calculate the correlation coefficient matrix between different parameters and take the absolute value to represent the correlation between indicators. The stronger the correlation with other indicators, the smaller the conflict between the indicator and other indicators. Multiply the standard deviation by the correlation coefficient to calculate the information content. The greater the information content of a certain parameter, the greater the role of that evaluation index in the entire evaluation index system. The objective weight is calculated based on the amount of information, and the formula is as follows:

[0012]

[0013] in Indicates the first The amount of information in each indicator The standard deviation of the indicator. The correlation coefficient between the indicators. For vegetation parameter weights, is the first vegetation parameter index.

[0014] According to the vegetation dynamic comprehensive index construction method based on a geographical weighted comprehensive index provided by the application, a continuous weight surface of land cover partition weight is constructed by using spatial interpolation, and the method comprises the following steps: For the land cover type data of each year, a preset number of points are randomly selected in the type area, if the number of type pixels is less than the preset number, all the points are selected, and the coordinates of all the points and the weight value of the type in the month are recorded; For each parameter, the weight value of the sampling point is used to construct an interpolation model, and a cubic spline sampling method is used for spatial continuous interpolation to obtain the weight surface of the parameter in the whole area.

[0015] According to the vegetation dynamic comprehensive index construction method based on a geographical weighted comprehensive index provided by the application, a continuous weight surface of land cover partition weight is constructed by using spatial interpolation, and the method comprises the following steps: The dynamic weighted fusion is performed pixel by pixel, and in a specific year y , month m , location lat , lon , the comprehensive index is calculated as:

[0016] Among them, from the continuous interpolation surface, is a global normalized vegetation parameter.

[0017] In the second aspect, the application further provides a vegetation dynamic comprehensive index construction system based on a geographical weighted comprehensive index, comprising: A preprocessing module is configured to obtain monthly global land surface satellite remote sensing product data of a research area, and to obtain preprocessed remote sensing data by preprocessing the global land surface satellite remote sensing product data; A calculation module is configured to calculate land cover partition weight of the preprocessed remote sensing data according to land cover types; A construction module is configured to construct a continuous weight surface of the land cover partition weight by using spatial interpolation; A processing module is configured to perform dynamic weighted fusion pixel by pixel in the research area based on the continuous weight surface to obtain a vegetation dynamic comprehensive index.

[0018] ​In a third aspect, the present application also provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for constructing a vegetation dynamic comprehensive index based on a geographical weighted comprehensive index according to any one of the above aspects when executing the program.

[0019] In a fourth aspect, the present application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program is executable on a processor to implement the method for constructing a vegetation dynamic comprehensive index based on a geographical weighted comprehensive index according to any one of the above aspects.

[0020] In a fifth aspect, the present application also provides a computer program product comprising a computer program, wherein the computer program is executable on a processor to implement the method for constructing a vegetation dynamic comprehensive index based on a geographical weighted comprehensive index according to any one of the above aspects.

[0021] The method for constructing a vegetation dynamic comprehensive index based on a geographical weighted comprehensive index provided by the present application integrates GLASS LAI, FVC, NDVI, and NPP and other multi-source remote sensing vegetation information by adopting a geographical weighted comprehensive index (GWS) method combining global trends and local adaptation, and finally constructs a spatiotemporally coordinated vegetation dynamic comprehensive index. The method solves the heterogeneity of the ecosystem through a land cover partition framework, captures the nonlinear relationship between parameters by using CRITIC objective weighting, and breaks the limitation of traditional comprehensive indexes using global uniform weights and thus ignoring the functional differences of different ecosystems. The method realizes weight continuity by means of spatial interpolation technology, which can eliminate the influence of boundary mutation effect of different land cover types. In addition, the data utilize a spatiotemporally seamless long-time series vegetation parameter dataset, which can support long-term dynamic observation and provide method and data support for revealing the long-time series evolution mechanism of vegetation. BRIEF DESCRIPTION OF DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings needed in the embodiments or prior art description. Obviously, the drawings in the following description are some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained without creative labor.

[0023] Figure 1 is a flowchart of the method for constructing a vegetation dynamic comprehensive index based on a geographical weighted comprehensive index provided by the present application; Figure 2 is a distribution result map of a comprehensive vegetation index (SVI) of the Yellow River Basin in different seasons from 2000 to 2021 provided by the present application; Figure 3The application provides a comparison chart of time series variation trends of a synthetical vegetation index (SVI) and other vegetation parameters. Figure 4 The application provides a structure schematic diagram of a vegetation dynamic synthetical index construction system based on a geographically weighted synthesis index. Figure 5 The application provides a structure schematic diagram of an electronic device. DETAILED DESCRIPTION

[0024] To make the objects, technical solutions and advantages of the application clearer, the technical solutions in the application will be described below in detail with reference to the drawings in the application. Obviously, the described embodiments are some but not all of the embodiments of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative work fall within the protection scope of the application.

[0025] In view of the problems in the prior art, the application develops a synthetical vegetation index (SVI) by comprehensively analyzing indexes closely related to vegetation health and dynamic change, such as greenness (NDVI), canopy structure (LAI), coverage (FVC) and productivity (NPP), so as to comprehensively reflect the growth condition of vegetation. In order to effectively capture spatial heterogeneity and improve the ecological interpretability of the model, a geographically weighted synthesis (GWS) method is further provided. The GWS combines the global trend and local adaptation by fusing spatial heterogeneity modeling and multi-variable dynamic weighting, and solves the limitations of traditional methods in spatial heterogeneity, ecological interpretability and management applicability.

[0026] Figure 1 The application provides a construction method of a vegetation dynamic synthetical index based on a geographically weighted synthesis index, as shown in Figure 1 The method comprises the following steps. Step 100: Global land surface satellite remote sensing product data of a research area is acquired, and the global land surface satellite remote sensing product data is preprocessed to obtain preprocessed remote sensing data. Step 200: Land cover partition weights are calculated from the preprocessed remote sensing data according to land cover types. Step 300: A continuous weight surface of the land cover partition weights is constructed by using spatial interpolation. Step 400: Vegetation dynamic synthetical indexes are obtained by performing pixel-by-pixel dynamic weighted fusion based on the continuous weight surface in the research area.

[0027] The technical scheme of the embodiment of the present application mainly includes four aspects of data preprocessing, land cover partition weight calculation, continuous weight construction and comprehensive index calculation.

[0028] Based on the above embodiment, step 100 comprises: Obtaining the HDF image of the study area, performing format conversion, image mosaicking, image cropping and projection on the HDF image to obtain a TIF image with a preset time resolution and a preset spatial resolution; Performing month maximum data synthesis on the TIF image, and obtaining the true results of each vegetation parameter in the study area through image scaling and data quality control; Determine that the land cover data adopts a global land cover classification data set, and integrate into a plurality of land type data.

[0029] Specifically, the input data of the embodiment of the present application is the monthly GLASS data of the study area, including greenness (NDVI), canopy structure (LAI), coverage (FVC) and productivity (NPP) and other indicators closely related to vegetation health and dynamic change. The HDF image of the study area is downloaded from the GLASS official website, and after image format conversion, image mosaicking, image cropping and projection, a TIF image with a time resolution of 8 days and a spatial resolution of 500m is obtained. Then, the obtained data is subjected to month maximum data synthesis, and the true results of each vegetation parameter in the study area are obtained through image scaling and data quality control. The land cover data adopts MCD12Q1 annual data, and for the sake of simplicity, similar land cover types are integrated and reclassified into forest, shrub, grassland, wetland, farmland, impervious surface, ice and snow and water body.

[0030] Based on the above embodiment, step 200 comprises: The maximum and minimum values are used to normalize the preprocessed remote sensing data, a plurality of months are independently processed in the time dimension to capture phenological differences, and an analysis unit is constructed according to the annual land cover partition types in the spatial dimension to form a three-dimensional data matrix. According to the land cover type, the vegetation parameter pixels corresponding to each category are extracted, and the CRITIC method is used to calculate the weight of each vegetation parameter for all pixels in each land cover partition.

[0031] Specifically, in the embodiment of the present application, since the dimensions of each parameter are different, the parameters need to be normalized before the comprehensive index is constructed. In this embodiment, min-max normalization is used for each parameter to the interval [0, 1].

[0032] (1) Wherein, represents the normalized value, It is the value of a specific grid cell. It is the minimum value of the grid cell. This represents the maximum value of the grid cell. In the temporal dimension, months 1-12 are processed independently to capture phenological differences; in the spatial dimension, analysis cells are constructed based on annual land cover zoning (8 major types). To reduce the impact of spatiotemporal heterogeneity on the calculation results, the globally normalized values ​​of the four parameters are first calculated, and then the zoning weights are calculated by extracting all globally normalized cells for the same month within the study period for each land cover type zoning. For example, for a specific month... m And land cover type c, extract all years within the study period m Monthly data, merge this type c The normalized parameters of all pixels are used to form a three-dimensional data matrix.

[0033] Subsequently, based on land cover type, vegetation parameter pixels corresponding to each category are extracted. For all pixels within each land cover zone, the CRITIC method is used to calculate the weight of each vegetation parameter. The CRITIC method is an objective weighting method that determines weights based on the contrast intensity and conflict between indicators. The specific calculation method is as follows: a. First, calculate the standard deviation of each parameter to represent the variation and fluctuation of values ​​within each indicator. The larger the standard deviation, the greater the numerical variation of the indicator, the more information it can reflect, and the stronger the evaluation strength of the indicator itself. Therefore, more weight should be assigned to the indicator. b. Then calculate the correlation coefficient matrix between different parameters and take the absolute value to represent the correlation between indicators. The stronger the correlation with other indicators, the less conflict there is between the indicator and other indicators, the more identical information it reflects, and the more repetitive the evaluation content it can reflect. To a certain extent, this weakens the evaluation strength of the indicator, and the weight assigned to the indicator should be reduced.

[0034] c. Multiply the standard deviation by the correlation coefficient to calculate the information content. The greater the information content of a parameter, the greater its role in the entire evaluation index system, and the more weight it should be assigned.

[0035] d. Finally, calculate the objective weight based on the amount of information. The formula is: (2) (3) in Indicates the first The amount of information in each indicator The standard deviation of the indicator. The correlation coefficient between the indicators. For vegetation parameter weights, is the first vegetation parameter index.

[0036] Based on the above embodiment, step 300 comprises: For the land cover type data of each year, a preset number of points are randomly selected in the type area, if the number of type pixels is less than the preset number, all are selected, the coordinates of all points and the weight value of the type in the month are recorded; For each parameter, the weight values of the sampling points are used to construct an interpolation model respectively, and a cubic spline sampling method is used for spatial continuous interpolation to obtain the weight surface of the parameter in the entire area.

[0037] Specifically, to solve the problem of boundary mutation of the partition, the embodiment of the application constructs a continuous weight surface by spatial interpolation. For the land cover type data of each year, 1000 points are randomly selected in the type area (if the number of type pixels is less than 1000, all are selected), the coordinates of these points and the weight value (4 weights) of the type in the month are recorded. Then, for each parameter, the weight values of the sampling points are used to construct an interpolation model respectively.

[0038] Here, the embodiment adopts a cubic spline sampling method (cubic) for spatial continuous interpolation to obtain the weight surface of the parameter in the entire area. Since interpolation is required for each month and each parameter, 4 times of interpolation (4 parameters) are required for each month.

[0039] Based on the above embodiment, step 400 comprises: The embodiment of the application performs dynamic weighted fusion pixel by pixel, and in a specific year y , month m , location lat , lon ), the comprehensive index is calculated as: (4) Wherein, derived from the continuous interpolation surface, is the global normalized vegetation parameter.

[0040] As shown in Figure 2 , the distribution result graph of the comprehensive vegetation index (SVI) of the Yellow River Basin in different seasons from 2000 to 2021 is shown, and as shown in Figure 3 , the time series change trend comparison graph of the comprehensive vegetation index (SVI) and other vegetation parameters is shown.

[0041] In summary, the beneficial effects of the application include: ​(1) The geographical weighted synthetic index (GWS) method combining "global trend and local adaptation" is adopted, the monthly independent processing (retaining the phenological characteristics) is adopted in time, and the land cover type partition is adopted in space, so that the overall change trend of the vegetation growth condition in the macro layer is considered, and the unique characteristics and spatial heterogeneity of the vegetation growth in different regions are fully reflected; (2) The CRITIC objective weighting method is adopted for different land cover partitions, the internal variability (standard deviation) and the correlation (conflict) between parameters are considered, and the balance between the information amount and the correlation of various parameters is realized; (3) The weight continuity is realized by means of the spatial interpolation technology, and the influence of the boundary mutation effect of different land cover types can be eliminated; (4) The dynamic weighted fusion of the comprehensive index is carried out on the pixel scale, the index value of each position in each month depends on the normalized value of the four parameters and the exclusive weight obtained by interpolation, the spatial continuity is retained, and the seasonal dynamics is captured through the monthly calculation.

[0042] The geographical weighted synthetic index-based vegetation dynamic comprehensive index construction system provided by the present application is described below, and the geographical weighted synthetic index-based vegetation dynamic comprehensive index construction system described below can be correspondingly referred to the geographical weighted synthetic index-based vegetation dynamic comprehensive index construction method described above.

[0043] Figure 4 The geographical weighted synthetic index-based vegetation dynamic comprehensive index construction system provided by the present application is described below, and the geographical weighted synthetic index-based vegetation dynamic comprehensive index construction system described below can be correspondingly referred to the geographical weighted synthetic index-based vegetation dynamic comprehensive index construction method described above. Figure 4 As shown in FIG. 1, the geographical weighted synthetic index-based vegetation dynamic comprehensive index construction system comprises a preprocessing module 41, a calculation module 42, a construction module 43 and a processing module 44, wherein: The preprocessing module 41 is used for acquiring the global land surface satellite remote sensing product data of a research region month by month, and pre-processing the global land surface satellite remote sensing product data to obtain pre-processed remote sensing data; the calculation module 42 is used for calculating the land cover partition weight of the pre-processed remote sensing data according to the land cover type; the construction module 43 is used for constructing the continuous weight surface of the land cover partition weight by means of spatial interpolation; and the processing module 44 is used for performing pixel-by-pixel dynamic weighted fusion based on the continuous weight surface in the research region to obtain the vegetation dynamic comprehensive index.

[0044] Figure 5 An example of an entity structure schematic diagram of an electronic device is shown in FIG. 2. Figure 5As shown, the electronic device can include a processor 510, a communications interface 520, a memory 530, and a communications bus 540, wherein the processor 510, the communications interface 520, and the memory 530 complete mutual communication through the communications bus 540. The processor 510 can invoke the logic instructions in the memory 530 to execute the vegetation dynamic comprehensive index construction method based on a geographical weighting comprehensive index, which includes: obtaining global land surface satellite remote sensing product data of a study area month by month, preprocessing the global land surface satellite remote sensing product data to obtain preprocessed remote sensing data; calculating land cover partition weights for the preprocessed remote sensing data according to land cover types; constructing a continuous weight surface of the land cover partition weights by using spatial interpolation; and performing pixel-by-pixel dynamic weighted fusion based on the continuous weight surface in the study area to obtain a vegetation dynamic comprehensive index.

[0045] In addition, the logic instructions in the memory 530 described above can be implemented in the form of a software function unit and sold or used as an independent product, and can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM, Read-Only Memory), a random access memory (RAM, Random Access Memory), a magnetic disk or an optical disk, and various media that can store program codes.

[0046] On the other hand, the present application also provides a computer program product, which includes a computer program, the computer program can be stored on a non-transitory computer readable storage medium, and the computer program is executed by a processor, so that the computer can execute the vegetation dynamic comprehensive index construction method based on a geographical weighting comprehensive index provided by the above-mentioned methods, which includes: obtaining global land surface satellite remote sensing product data of a study area month by month, preprocessing the global land surface satellite remote sensing product data to obtain preprocessed remote sensing data; calculating land cover partition weights for the preprocessed remote sensing data according to land cover types; constructing a continuous weight surface of the land cover partition weights by using spatial interpolation; and performing pixel-by-pixel dynamic weighted fusion based on the continuous weight surface in the study area to obtain a vegetation dynamic comprehensive index.

[0047] In yet another aspect, the present application also provides a non-transitory computer readable storage medium having stored thereon a computer program, which, when executed by a processor, implements the method for constructing a vegetation dynamic comprehensive index based on a geographical weighted comprehensive index as provided by the above methods, the method comprising: obtaining monthly global land surface satellite remote sensing product data of a study area, pre-processing the global land surface satellite remote sensing product data to obtain pre-processed remote sensing data; calculating land cover partition weights for the pre-processed remote sensing data according to land cover types; constructing a continuous weight surface of the land cover partition weights using spatial interpolation; and performing pixel-by-pixel dynamic weighted fusion based on the continuous weight surface in the study area to obtain a vegetation dynamic comprehensive index.

[0048] The device embodiments described above are merely illustrative, wherein the units described as separate components can or can not be physically separate, and the components shown as units can or can not be physical units, i.e., can be located in one place or distributed on multiple network units. Part or all of the modules can be selected to achieve the purposes of the embodiments according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0049] From the above description of the embodiments, those skilled in the art can clearly understand that the embodiments can be implemented by means of software and the necessary universal hardware platform, and of course, can also be implemented by hardware. Based on such understanding, the above technical solutions, essentially or in terms of contribution to the prior art, can be embodied in the form of a software product. The computer software product can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0050] Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement for some technical features; and these modifications 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 application.

Claims

1. A method for constructing a vegetation dynamic comprehensive index based on a geographical weighted comprehensive index, characterized in that, The method comprises the following steps: obtaining global land surface satellite remote sensing product data of a research area month by month, preprocessing the global land surface satellite remote sensing product data to obtain preprocessed remote sensing data; calculating land cover partition weights according to land cover types; constructing a continuous weight surface of the land cover partition weights by using spatial interpolation; performing pixel-by-pixel dynamic weighted fusion based on the continuous weight surface in the research area to obtain a vegetation dynamic comprehensive index.

2. The method according to claim 1, wherein, The method comprises the following steps: obtaining global land surface satellite remote sensing product data of a research area month by month, preprocessing the global land surface satellite remote sensing product data to obtain preprocessed remote sensing data, comprising: obtaining HDF images of the research area, performing format conversion, image mosaicking, image cropping and projection on the HDF images to obtain TIF images with a preset time resolution and a preset spatial resolution; performing month maximum data synthesis on the TIF images, and obtaining true results of each vegetation parameter in the research area through image scaling and data quality control; 3. The method according to claim 1, wherein, determining that the land cover data adopts a global land cover classification data set and is integrated into multiple land type data. The method comprises the following steps: normalizing the preprocessed remote sensing data using maximum and minimum values, independently processing multiple months in the time dimension to capture phenological differences, and constructing analysis units according to annual land cover partition types in the spatial dimension to form a three-dimensional data matrix; 4. The method according to claim 3, wherein, extracting vegetation parameter pixels corresponding to each category according to land cover types, and calculating the weight of each vegetation parameter in all pixels within each land cover partition using the CRITIC method. The method comprises the following steps: calculating the standard deviation of each parameter to represent the difference and fluctuation of the values of each index, and the greater the standard deviation, the greater the numerical difference of the index; calculating the correlation coefficient matrix between different parameters and taking the absolute value to represent the correlation between indexes, and the stronger the correlation with other indexes, the smaller the conflict between the index and other indexes; multiplying the standard deviation and the correlation coefficient to calculate the information amount, and the greater the information amount of a certain parameter, the greater the role of the evaluation index in the entire evaluation index system; wherein represents the information amount of the th index, is the index standard deviation, is the correlation coefficient between indexes, is the weight of the vegetation parameter, is the th vegetation parameter index.

5. The method according to claim 1, wherein, calculating the objective weight according to the information amount, and the calculation formula is: The method comprises the following steps: for the land cover type data of each year, randomly selecting a preset number of points in the type area, if the number of type pixels is less than the preset number, selecting all the points, recording the coordinates of all the points and the weight value of the type in the month; 6. The method according to claim 1, wherein, for each parameter, using the weight values of the sampling points to construct an interpolation model, using a cubic spline sampling method for spatial continuous interpolation to obtain the weight surface of the entire region for the parameter. The method comprises the following steps: performing pixel-by-pixel dynamic weighted fusion based on the continuous weight surface in the research area to obtain a vegetation dynamic comprehensive index, comprising: Dynamic weighted fusion is performed pixel by pixel, at specific years y , months m , locations lat , lon , the composite index is calculated as: wherein, from a continuous interpolation surface, is a global normalized vegetation parameter.

7. A vegetation dynamic comprehensive index construction system based on a geographical weighted comprehensive index, characterized in that, The method comprises the following steps: a preprocessing module is configured to obtain monthly global land surface satellite remote sensing product data of a study area, and preprocess the global land surface satellite remote sensing product data to obtain preprocessed remote sensing data; a calculation module is configured to calculate land cover partition weights of the preprocessed remote sensing data according to land cover types; a construction module is configured to construct a continuous weight surface of the land cover partition weights by using spatial interpolation; a processing module is configured to perform pixel-by-pixel dynamic weighted fusion based on the continuous weight surface in the study area to obtain a vegetation dynamic comprehensive index.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor implements the vegetation dynamic comprehensive index construction method based on the geographic weighted comprehensive index according to any one of claims 1 to 6 when executing the program. 9.A non-transitory computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program implements the vegetation dynamic comprehensive index construction method based on the geographic weighted comprehensive index according to any one of claims 1 to 6 when executed by the processor.

10. A computer program product comprising a computer program, characterized in that, The computer program implements the vegetation dynamic comprehensive index construction method based on the geographic weighted comprehensive index according to any one of claims 1 to 6 when executed by the processor.

Images