CN111259963B - Driving factor analysis method and device for regional vegetation index and storage medium - Google Patents
Driving factor analysis method and device for regional vegetation index and storage medium Download PDFInfo
- Publication number
- CN111259963B CN111259963B CN202010051514.5A CN202010051514A CN111259963B CN 111259963 B CN111259963 B CN 111259963B CN 202010051514 A CN202010051514 A CN 202010051514A CN 111259963 B CN111259963 B CN 111259963B
- Authority
- CN
- China
- Prior art keywords
- data
- cluster
- combined
- value
- vegetation
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a driving factor analysis method, a device and a storage medium for regional vegetation indexes, wherein the method comprises the following steps: acquiring vegetation index data and at least two driving factor data of vegetation indexes of a research area; carrying out local space autocorrelation analysis on vegetation index data to obtain analysis results containing multiple clustering types; combining different cluster types according to a preset combination rule to obtain at least one combined cluster type; and carrying out importance analysis on driving factors of vegetation indexes belonging to each combined cluster type in at least one combined cluster type by utilizing a random forest model. According to the embodiment of the application, through the technical scheme, the analysis precision of the driving factors can be improved by the recombined partial combined cluster types, so that the precision requirement is met, and the driving factor analysis result of specific and important cluster types is extracted from the whole body.
Description
Technical Field
The application relates to the technical field of data processing, in particular to a driving factor analysis method and device for regional vegetation indexes and a storage medium.
Background
In the vegetation ecological index, the vegetation net primary productivity (Net Primary Productivity, NPP) is taken as an example, is the basis of the structure and the function of a plant ecological system, is important data for researching the carbon sequestration capability of the vegetation, is also a key element for analyzing and judging the carbon sink/source process of the ecological system, is a main parameter for representing global change response, and can be also used as an important reference index for assisting government decision and analyzing economic development degree.
In view of the characteristics of the vegetation index data of different types of areas, the driving factors thereof and the complexity of the relationship between the different driving factors, a proper algorithm is one of the keys influencing the analysis precision effect of the driving factors.
Currently, there are various methods in terms of driving factor analysis of regional vegetation indexes, and linear regression analysis is a more common method. However, the method often has the problems of low result precision and difficulty in meeting the expected target in practical application. One of the reasons is that regression analysis methods have strict assumption conditions, i.e., linearity (dependent variable and each independent variable are linear), independence (for all observations, the error terms are independent of each other), normalization (error terms obey normal distribution), isovariabilities (random error terms in regression functions have the same variance), independent of each other. However, the actual data hardly meets the conditions, and when the method is used, the applicability of the method and the actual characteristics of the data are ignored, the actual relationship between the data is difficult to objectively and accurately reflect, large errors can be caused, and the reliability of the result is reduced.
In addition, the existing analysis method generally analyzes the whole research area, and due to the problems of spatial heterogeneity of environmental factors, complex interaction among different environmental factors, different influences on regional vegetation indexes and the like, the regional vegetation indexes always show higher variation degree and complex variation trend, and the error of the analysis result of driving factors is always larger, so that main driving factors causing the phenomena are not easy to find, and particularly, the method is more prominent in mountain research areas with larger topography fluctuation and serious vegetation distribution fragmentation. How to solve this problem is also in need of intensive research.
Disclosure of Invention
The embodiment of the application aims to provide a driving factor analysis method, a driving factor analysis device and a storage medium for regional vegetation indexes, so as to solve the problems that the accuracy of results is low, the accuracy requirement is not easy to meet and the special application scene is available in the prior art.
In view of the foregoing, in a first aspect, the present application provides a driving factor analysis method for a regional vegetation index, the method including: acquiring vegetation index data and at least two driving factor data of vegetation indexes of a research area; carrying out local space autocorrelation analysis on vegetation index data to obtain analysis results containing multiple clustering types; combining different cluster types according to a preset combination rule to obtain at least one combined cluster type; and carrying out importance analysis on driving factors of vegetation indexes belonging to each combined cluster type in at least one combined cluster type by utilizing a random forest model.
Therefore, in the embodiment of the application, the divided cluster types are recombined, and the output result of the random forest model can be used for determining, so that the analysis precision of the driving factors can be improved by the recombined partial combined cluster types, thereby meeting the precision requirement.
In addition, the scheme in the embodiment of the application is also suitable for the driving factor analysis of the geographic data of other vegetation indexes and non-vegetation indexes, and the combined clustering type is also suitable for the spatial distribution pattern analysis of various geographic data.
In one possible embodiment, the vegetation index data comprises net primary productivity data; the driving factor data for net primary productivity includes at least two of normalized vegetation index data, population density data, domestic total production data, night light brightness data, digital elevation model data, annual average rainfall data, annual average air temperature data, and grade data.
Therefore, the embodiment of the application can more comprehensively, comprehensively and objectively analyze the driving factors in aspects of biology, climate, topography, people and the like.
In one possible embodiment, the local spatial autocorrelation analysis is performed on the vegetation index data to obtain an analysis result including a plurality of cluster types, including: importing vegetation index data into a geographic information system; and obtaining analysis results containing various cluster types through a geographic information system, wherein the cluster types comprise a high-value cluster HH, a low-value cluster LL, an abnormal value HL with a high value mainly surrounded by a low value, an abnormal value LH with a low value mainly surrounded by a high value and an insignificant point.
Therefore, according to the embodiment of the application, through the analysis result of the cluster type, the actual distribution condition of the cluster type of the vegetation index of the research area is determined, and the actual distribution condition can reflect the aggregation and heterogeneity of the spatial distribution of the vegetation index.
In one possible embodiment, the at least one combined cluster type comprises at least one of the following types: the first, second, third, fourth and fifth combined cluster types are combined according to a preset combination rule to obtain at least one combined cluster type, including: combining the high-value cluster HH and the low-value cluster LL to obtain a first combined cluster type; and/or combining the high-value cluster HH and the abnormal value HL with the high value mainly surrounded by the low value to obtain a second combined cluster type; and/or combining the low-value cluster LL and the abnormal value LH with the low value mainly surrounded by the high value to obtain a third combined cluster type; and/or combining the abnormal value HL with the high value mainly surrounded by the low value and the abnormal value LH with the low value mainly surrounded by the high value to obtain a fourth combined cluster type; and/or combining the high-value cluster HH, the low-value cluster LL, the outlier HL with the high value mainly surrounded by the low value and the outlier LH with the low value mainly surrounded by the high value to obtain a fifth combined cluster type.
Therefore, the embodiment of the application determines the combination mode among different cluster types based on the analysis result, analyzes the driving factors of the different cluster types by at least one combined cluster type, and compares the analysis results based on the driving factor analysis, so that the analysis results of the driving factors are more accurate and comprehensive.
In one possible embodiment, the local spatial autocorrelation analysis is performed on the vegetation index data to obtain an analysis result including a plurality of cluster types, including: the combined data combined by all cluster types is acquired.
In one possible embodiment, before the importance analysis of the driving factors of the vegetation indexes belonging to each of the at least one combined cluster type by using the random forest model, the driving factor analysis method further includes: acquiring a plurality of data sets according to the combination data, the plurality of cluster types and at least one combination cluster type, wherein the plurality of data sets comprise vegetation index data corresponding to the combination data and driving factor data of vegetation indexes belonging to the combination data, vegetation index data corresponding to each cluster type in the plurality of cluster types and driving factor data of vegetation indexes belonging to each cluster type in the plurality of cluster types, vegetation index data corresponding to each combination cluster type in the at least one combination cluster type and driving factor data of vegetation indexes belonging to each combination cluster type in the at least one combination cluster type.
In one possible embodiment, using a random forest model, the importance analysis of the driving factors of the vegetation index belonging to each of the at least one combined cluster type includes: and respectively acquiring an importance ranking result of the driving factors of the vegetation indexes belonging to the combined data, an importance ranking result of the driving factors of the vegetation indexes of each cluster type in the plurality of cluster types and an importance ranking result of the driving factors of the vegetation indexes of each cluster type in the at least one combined cluster type by utilizing the random forest model and the plurality of data sets.
Therefore, the application provides scientific basis for vegetation monitoring and conservation in the research area, ecological engineering achievement evaluation, forest management scheme establishment, management policy establishment and the like by acquiring the importance ranking result of the driving factors of the vegetation indexes belonging to the combined data, the importance ranking result of the driving factors of the vegetation indexes of each type of the cluster type and the importance ranking result of the driving factors of the vegetation indexes of each type of the combined cluster type.
In one possible embodiment, the driving factor analysis method further includes: and respectively acquiring a combined precision evaluation index corresponding to the combined data, a first precision evaluation index corresponding to each of a plurality of cluster types and a second precision evaluation index corresponding to each of at least one combined cluster type by utilizing the random forest model and the plurality of data sets.
Therefore, the method verifies that the recombined partial combined cluster type improves the precision compared with the cluster type and the combined data by combining the precision evaluation index, the first precision evaluation index and the second precision evaluation index.
In a second aspect, the present application further provides a driving factor analysis device for a regional vegetation index, including: the first acquisition module is used for acquiring vegetation index data and at least two driving factor data of vegetation indexes of the research area; the first analysis module is used for carrying out local spatial autocorrelation analysis on the vegetation index data so as to obtain analysis results containing multiple clustering types; the second acquisition module is used for combining different cluster types according to a preset combination rule to acquire at least one combined cluster type; and the second analysis module is used for carrying out importance analysis on driving factors of vegetation indexes belonging to each combined cluster type in the at least one combined cluster type by utilizing the random forest model.
In a third aspect, the present application provides an electronic device comprising: a processor, a memory and a bus, the memory storing machine-readable instructions executable by the processor, the processor and the memory in communication over the bus when the electronic device is running, the machine-readable instructions when executed by the processor perform the method of any of the first aspect and optional implementation of the first aspect.
In a fourth aspect, the present application provides a storage medium having stored thereon a computer program which, when executed by a processor, performs the method of the first aspect and any optional implementation of the first aspect.
In a fifth aspect, the application provides a computer program product which, when run on a computer, causes the computer to perform the method of the first aspect and any optional implementation of the first aspect.
Additional features and advantages of the application will be set forth in the description which follows, and in part will be apparent from the description, or may be learned by practice of the embodiments of the application. The objectives and other advantages of the application will be realized and attained by the structure particularly pointed out in the written description and claims thereof as well as the appended drawings.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the embodiments of the present application will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present application and should not be considered as limiting the scope, and other related drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flow chart of a driving factor analysis method for regional vegetation indexes according to an embodiment of the present application;
FIG. 2 is a flowchart of step 102 in a driving factor analysis method for regional vegetation index according to an embodiment of the present application;
fig. 3 is a block diagram of a driving factor analysis device for a regional vegetation index according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be described below with reference to the accompanying drawings in the embodiments of the present application.
It should be noted that: like reference numerals and letters denote like items in the following figures, and thus once an item is defined in one figure, no further definition or explanation thereof is necessary in the following figures. Meanwhile, in the description of the present application, the terms "first", "second", and the like are used only to distinguish the description, and are not to be construed as indicating or implying relative importance.
At present, the existing scheme usually analyzes driving factors of the whole body (or called as combined data) which is not clustered, but the problems that the accuracy of the result is low, the accuracy requirement is not easily met, and the whole body is not clustered and recombined appropriately so as to realize more accurate and comprehensive analysis are frequently encountered through the mode.
Or, although the existing schemes can analyze the driving factors of the NPP in a certain research area, some schemes often have the problems that the result accuracy is not high and the expected target is not easily met; in addition, some technical schemes have special applicable scenes, and when the technical schemes are used in inapplicable scenes, larger deviation can be caused, and the reliability of results is reduced. Similarly, driver analysis of other vegetation indices often suffers from the above-described problems.
In addition, spatial data of indexes to be analyzed are subjected to proper spatial clustering, and different combination types of results based on the spatial clustering and the results thereof are combined with driving factor analysis to refine analysis results, so that different spatial clustering types and the combination thereof are compared with the dissimilarity of driving factors of overall data before clustering, which is not disclosed in the prior art.
As shown in fig. 1, the embodiment of the application provides a driving factor analysis method for a regional vegetation index. It should be understood that the method shown in fig. 1 may be performed by a driving factor analysis device of a regional vegetation index, which may correspond to the driving factor analysis device 300 of a regional vegetation index shown in fig. 3 hereinafter, which may be various apparatuses capable of performing the method, for example, a personal computer, a server, a network device, or the like, to which embodiments of the present application are not limited. The method specifically comprises the following steps:
Step S101, vegetation index data and at least two driving factor data of vegetation indexes of a research area are obtained.
It should be appreciated that the driving factors for the vegetation index may include the vegetation index or driving factors pertaining to the vegetation aspect. For example, the driving factors corresponding to vegetation may include normalized vegetation index, etc., as the application is not limited in this regard. In addition, the driving factors of the vegetation index may also include driving factors of non-vegetation aspect, for example, the driving factors of the vegetation index may include population density, etc., which is not limited in the present application.
In this step S101, the vegetation index data and/or driving factor data of the research area may be obtained by direct download from a website displaying its related content. The partial index data can also be obtained by acquiring related satellite remote sensing image data, and then carrying out corresponding processing and calculation; wherein, part of index data can also be obtained by calculating by using related data and a model (such as data of available light energy utilization rate and the like and CASE model to calculate NPP and the like); the partial index data can also be obtained by relevant interpolation calculation after field measurement or investigation, and the application is not limited to this.
Further, the vegetation index data may include NPP data, forest biomass data, vegetation coverage data, and the like. Among these, vegetation is a generic term for a plant community covering the earth's surface.
It should be understood that, although three data types of vegetation index data are illustrated in the above embodiments, those skilled in the art may set the data types of the vegetation index data according to actual needs, which is not limited in the present application.
In addition, in the case of downloading vegetation index data from a website that presents its relevant content, the above-described research area may be a part of the area covered by the downloaded data.
In this case, the downloaded data may be preprocessed, and then the preprocessed data may be trimmed, i.e., the subsequent analysis may analyze only the data to be analyzed of the investigation region.
For example, in the case where the vegetation index data is NPP data, the user may download NPP raster data, research area administrative boundary vector data, and driving factor raster data of NPP in 2000 to 2010 of china from the resource environment data cloud platform. Among the driving factor raster data of NPP include NDVI (Normalized Difference Vegetation Index, normalized vegetation index) raster data, population density raster data, GDP (Gross Domestic Product, home produced total) raster data, night light brightness raster data, digital elevation model raster data (grade data can be derived from digital elevation model data in ArcGIS software), annual average rainfall raster data, and annual average air temperature raster data. After the downloaded data are acquired, if the geospatial properties of the data are different, the resolution of the coordinate system and the raster data need to be unified, that is, the projection, the coordinates and the resolution of various data are unified after the projection and the coordinate conversion and the raster resampling. Finally, NPP raster data of the investigation region and driving factor raster data of the NPP are cut out by using the investigation region administrative boundary vector data.
In addition, although in step S101, only the driving factor data of the NPP is exemplified, it will be understood by those skilled in the art that the driving factor data of the vegetation index may be set according to actual requirements on the premise that the driving factor data of the vegetation index is not less than two.
For example, in the case where the vegetation index data is forest biomass data, the driving factors of the forest biomass may also include photosynthesis and respiration, which is not limited by the present application.
Although the driving factor data is described in step S101, the driving factor data may be driving factor data that is determined in advance to be able to affect the vegetation index data, or driving factor data that is not determined in advance to be able to affect the vegetation index data.
Step S102, carrying out local spatial autocorrelation analysis on vegetation index data to obtain analysis results containing multiple clustering types. That is, the analysis result includes related data of a plurality of cluster types.
In this step S102, in the case where the vegetation index data of the research area is raster data, as shown in fig. 2, step S102 may include step S1021 of performing format conversion on the vegetation index data after preprocessing. The format conversion may be converting vegetation index data from raster data to vector data.
For example, in the case where the vegetation index data is NPP raster data, the NPP raster data is imported into ArcGIS software, and is converted into NPP point data by a raster point tool in the ArcGIS software. The ArcGIS software is a platform for organizing, managing, sharing and analyzing spatial data, which is not limited in the present application.
It should be understood that, although the data format conversion tool using ArcGIS software as the vegetation index data is specifically exemplified in this step, those skilled in the art may select other format conversion tools to perform data format conversion according to actual needs, so long as it is ensured that raster data of the vegetation index data can be converted into vector data, which is not limited in the present application.
In addition, in the case where the acquired vegetation index data of the investigation region is vector data, step S1022 may be directly performed.
Step S102 may include step S1022 of importing the vegetation index data into the geographic information system.
For example, when the vegetation index data is NPP point data, all NPP point data related to the research area may be stored in advance in a format recognizable by ArcGIS software such as CSV (common-Separated Values), and the user may import the data in a format recognizable by ArcGIS software such as CSV in which NPP point data is stored into the geographic information system, and then may implement local spatial autocorrelation analysis of NPP point data through tools in the geographic information system.
Correspondingly, the processing procedure of other vegetation index data similar to the NPP is similar to that of the NPP data, and the following is not exemplified one by one, and can be seen in particular.
In addition, step S102 may include step S1023, i.e. obtaining an analysis result including a plurality of cluster types by performing local spatial autocorrelation analysis in the geographic information system.
Specifically, after NPP point data is imported into ArcGIS software, local spatial autocorrelation analysis is performed on the NPP point data by using a clustering and outlier analysis tool in the ArcGIS software, wherein the clustering and outlier analysis tool can identify spatial clusters of elements with high values or low values, i.e., the ArcGIS software can perform local spatial autocorrelation analysis on the input NPP point data. The ArcGIS software obtains the clustering and outlier analysis results containing various clustering types by calculating Local Moran's I (Local Moran index), z score and P value. The cluster types include a high-value cluster HH, a low-value cluster LL, an outlier HL with a high value mainly surrounded by a low value, an outlier LH with a low value mainly surrounded by a high value, and an insignificant point (No signalicant, NS for short), which the present application is not limited to.
In addition, although the geographic information system is specifically defined as ArcGIS software in this step, those skilled in the art may set a specific type of the geographic information system according to actual needs, so long as the input vegetation index data can be subjected to local spatial autocorrelation analysis, which is not limited in the present application.
The geographic information system may also be, for example, the spatial statistics software Geoda software.
In addition, when the geographic information system outputs analysis results of different cluster types, the analysis results include, in addition to the different cluster types, combined data composed of all the cluster types combined together.
It should be appreciated that the combined data may also be referred to as a population of data that is not clustered, and embodiments of the application are not so limited.
For example, the analysis results obtained by ArcGIS software include, in addition to the high-value cluster HH, the low-value cluster LL, the outlier HL whose high value is mainly surrounded by low values, the outlier LH whose low value is mainly surrounded by high values, and the insignificant point (NS), the combined data of the high-value cluster HH, the low-value cluster LL, the outlier HL whose high value is mainly surrounded by low values, the outlier LH whose low value is mainly surrounded by high values, and the insignificant point (NS).
In addition, the application can also determine the aggregation and heterogeneity of the spatial distribution of vegetation indexes through the analysis results comprising a plurality of cluster types.
For example, in the case that the vegetation index data is NPP data and NPP data for 2000, 2005 and 2010 of the Guizhou province, by using the above step S102, it can be determined that the high-value cluster HH points of the NPPs of 2000, 2005 and 2010 of the Guizhou province are mainly distributed in the eastern portion where the average altitude is low and the stony desertification degree is relatively low, while the low-value cluster LL points are mainly distributed in the western portion where the average altitude is high and the stony desertification degree is relatively high, and also that there are relatively small abnormal value HL points whose high values are mainly surrounded by low values and abnormal value LH points whose low values are mainly surrounded by high values are respectively distributed in the western portion and the eastern portion, which is not limited by the present application.
And step S103, extracting at least two driving factor data of the positions of all the element data of the vegetation index according to the analysis result.
In step S103, driving factor data corresponding to each element position of vegetation index data of the research area may be extracted by the geographic information system.
For example, in the case where the vegetation index data is NPP point data, after performing the local spatial autocorrelation analysis, the result of the local spatial autocorrelation is that the input NPP point data (or NPP point data in step S1021) is classified. However, the geographical location (or coordinates) of each point is unchanged, where all NPP point data and raster data of driving factors of at least two NPPs in the result of the local spatial autocorrelation analysis may be used as input data, the input data may be imported into ArcGIS software, and the raster pixel values of the driving factors of the NPPs at the positions of NPP point elements are extracted by multi-value extraction of the ArcGIS software to a point tool, so as to obtain element class data including NPP value fields and fields of the driving factors of at least two NPPs, where element classes in the ArcGIS software refer to element sets having the same geometric features, such as a point set, which is not limited in the present application.
In addition, in the present embodiment, the extraction process of the vegetation index data and the driving factor data of at least two vegetation indexes of the research area is specifically exemplified, but it should be understood by those skilled in the art that the above extraction process may be set according to actual requirements, and the present application is not limited thereto.
Step S104, combining different cluster types according to a preset combination rule to obtain at least one combined cluster type.
It should be understood that the combined cluster type may refer to a cluster type corresponding to the remaining combinations except for the combinations corresponding to the combined data. That is, the combined cluster type may be a cluster type that does not contain the remaining new combinations other than the cluster type to which the combined data corresponds.
In step S104, after the driving factor data corresponding to each element position of the vegetation index data is acquired, a combination of different cluster types may be determined based on the analysis result of the multiple cluster types in step S102, and the combination may be a combination of at least two different cluster types performed on the basis of step S102.
For example, by selecting the element class of the type (cluster type) or each combination type (at least one combination cluster type obtained by different cluster type combinations) respectively by attribute-by-attribute selection tools in ArcGIS software, so that on the basis of a plurality of cluster types (including a high-value cluster HH, a low-value cluster LL, an outlier HL whose high value is mainly surrounded by a low value, an outlier LH whose low value is mainly surrounded by a high value, and an insignificant point), at least one combination cluster type may include at least one of the following cluster types: the first, second, third, fourth, and fifth combined cluster types.
Wherein combining different cluster types to obtain at least one combined cluster type comprises: combining the high-value cluster HH and the low-value cluster LL to obtain a first combined cluster type; and/or combining the high-value cluster HH and the abnormal value HL with the high value mainly surrounded by the low value to obtain a second combined cluster type; and/or combining the low-value cluster LL and the abnormal value LH with the low value mainly surrounded by the high value to obtain a third combined cluster type; and/or combining the abnormal value HL with the high value mainly surrounded by the low value and the abnormal value LH with the low value mainly surrounded by the high value to obtain a fourth combined cluster type; and/or combining the high-value cluster HH, the low-value cluster LL, the outlier HL with the high value mainly surrounded by the low value, and the outlier LH with the low value mainly surrounded by the high value to obtain a fifth combined cluster type, which is not limited in the present application.
In the case where the combined data, the cluster type, and the combined cluster type are acquired, a plurality of data sets including each type and driving factor data thereof are acquired based on the combined data, the plurality of cluster types, and at least one combined cluster type. The plurality of data sets comprise vegetation index data corresponding to the combined data and driving factor data of vegetation indexes belonging to the combined data, vegetation index data corresponding to each of a plurality of cluster types and driving factor data of vegetation indexes belonging to each of a plurality of cluster types, and vegetation index data corresponding to each of at least one combined cluster type and driving factor data of vegetation indexes belonging to each of at least one combined cluster type.
It should be understood that one data set here is a set of vegetation index data and driving factor data thereof corresponding to the same, or the same cluster type, or the same combined cluster type.
For example, one data set may be combination data and driving factor data thereof, high-value cluster HH and driving factor data thereof, low-value cluster LL and driving factor data thereof, high-value outlier HL and driving factor data thereof mainly surrounded by low values, outlier LH and driving factor data thereof mainly surrounded by high values, insignificant point and driving factor data thereof, first combination cluster type and driving factor data thereof, second combination cluster type and driving factor data thereof, third combination cluster type and driving factor data thereof, fourth combination cluster type and driving factor data thereof, and fifth combination cluster type and driving factor data thereof, and embodiments of the present application are not limited thereto.
Specifically, under the condition that vegetation index data is NPP data, respective element classes are respectively derived by ArcGIS software according to the combined data, the cluster type and the combined cluster type, the element classes comprise numerical value fields of NPP and element classes of fields of driving factors of at least two NPPs, the element classes are respectively stored into dbf (database file) formats, then an Excel tool can be utilized to convert the dbf format of each element class into csv format, and finally data in each file in the csv format are respectively standardized by R software to eliminate the influence of different dimensions so as to obtain the csv file of standardized data, and the application is not limited to the above.
In addition, although the combination cluster types included in at least one combination cluster type are exemplified in the present embodiment, it should be understood by those skilled in the art that the combination cluster types included in at least one combination cluster type may also be set according to actual requirements, which is not limited in the present application.
Step S105, carrying out importance analysis on driving factors of vegetation indexes belonging to each combination cluster type in at least one combination cluster type by utilizing a random forest model.
It should be appreciated that the random forest model may also be referred to as a random forest algorithm, as the application is not limited in this regard. It should be appreciated that the random forest model may be a random forest code implemented by programming, or may be software capable of running the random forest model, as the application is not limited in this regard.
In addition, according to the random forest model and the plurality of data sets, an importance ranking result of the driving factors of the vegetation indexes belonging to the combined data, an importance ranking result of the driving factors of the vegetation indexes of each of the plurality of cluster types, and an importance ranking result of the driving factors of the vegetation indexes of each of the at least one combined cluster type are respectively obtained.
For example, in the case that the vegetation index data is NPP data, the standardized csv file obtained in step S104 is imported into Weka (Waikato Environment for Knowledge Analysis, wycark intelligent analysis environment) software, the NPP data is taken as a dependent variable, the data of at least two driving factors of NPP are taken as independent variables, and by setting related parameters and executing Random Forest tools in the Weka software, the driving factors of the vegetation index of the combined data, the driving factors of the vegetation index of the multiple cluster types, and the driving factors of the vegetation index of the at least one combined cluster type are respectively subjected to importance analysis, so that importance ranking results can be obtained by the Weka software, and the importance analysis results include: the ranking result of the drivers for each cluster type (e.g., high-value cluster HH), the ranking result of the drivers for each combined cluster type (e.g., first combined cluster type, etc.), and the ranking result of the drivers for the combined data.
Furthermore, while in the present embodiment, the driver importance analysis by the random forest model is specifically exemplified, it should be understood by those skilled in the art that the driver importance analysis may be performed by other suitable models or methods, and the present application is not limited thereto.
In addition, the software that implements the random forest model is specifically exemplified in this step as Weka software, but those skilled in the art will also appreciate that it may be implemented by other software that can implement random forests (or other suitable driving factor analysis methods).
For example, the method can also be implemented by Matlab software, and the application is not limited to the method.
In addition, using the random forest model and the plurality of data sets, a combined precision evaluation index corresponding to the combined data, a first precision evaluation index corresponding to each of the plurality of cluster types, and a second precision evaluation index corresponding to each of the at least one combined cluster type may also be obtained, respectively.
It should be understood that the output results of the random forest may include a combined precision evaluation index, a first precision evaluation index, a second precision evaluation index, and an importance ranking result, i.e., the combined precision evaluation index, the first precision evaluation index, the second precision evaluation index, and their respective corresponding importance ranking results may be the result data in the same output result of each dataset.
For example, in the case where the vegetation index data is NPP data, the standardized csv file obtained in step S104 is imported into Weka (Waikato Environment for Knowledge Analysis, wycark intelligent analysis environment) software, NPP data is used as a dependent variable, data of driving factors of at least two NPPs are used as independent variables, and by setting related parameters and executing Random Forest tools in the Weka software, an output result of the Weka software includes a combined precision evaluation index corresponding to the combined data, a first precision evaluation index corresponding to each of a plurality of cluster types, and a second precision evaluation index corresponding to each of at least one combined cluster type.
Wherein, the combined precision evaluation index or the first precision evaluation index or the second precision evaluation index can comprise a correlation coefficient (Correlation coefficient), an average absolute error (Mean absolute error), a root mean square error (Root mean squared error), a relative absolute error (Relative absolute error) and a relative root mean square error (Root relative squared error), and the 5 model evaluation indexes can be automatically calculated in the result of the Weka software executing a random forest algorithm, wherein, the correlation coefficient is used for verifying the deviation degree of a true value of data and a model predicted value, and the closer to 1, the higher the precision of the model is, and the lower the contrary is; the average absolute error and the root mean square error are used for measuring the difference between the predicted value and the actual result, and the smaller the better the difference is; the relative absolute error and the relative root mean square error are relative errors, and the magnitude of the errors is reflected by reflecting the proportion of the error to the true value, which is not limited by the application.
In addition, in the case where the combination accuracy evaluation index, the first accuracy evaluation index, and the second accuracy evaluation index are acquired, the difference in accuracy of the combination data and the various cluster types can be analyzed by comparing the combination accuracy evaluation index, the first accuracy evaluation index, and the second accuracy evaluation index.
For example, in the case where a plurality of first precision evaluation indexes corresponding to a plurality of cluster types are acquired, and also a second precision evaluation index corresponding to each of at least one combined cluster type, and a combined precision evaluation index corresponding to NPP data of the entirety not clustered are acquired, the correlation coefficient of the first combined cluster type is the highest, the values of the remaining four error indexes are the lowest, and the like, so that the precision of the first combined cluster type is determined to be the highest.
For another example, the combined cluster type in the recombined partial combined cluster type can be determined by comparing the combined precision evaluation index corresponding to the combined data, the first precision evaluation index corresponding to the multiple cluster types and the second precision evaluation index corresponding to the combined cluster type, so that the precision is improved compared with the cluster type and the combined data.
In addition, when the importance ranking result, the combined precision evaluation index, the first precision evaluation index and the second precision evaluation index are obtained, key driving factors of vegetation indexes of a dataset (belonging to the whole of non-clustering, the clustering type or the combined clustering type) meeting the precision requirement can be screened according to the importance ranking result and the precision evaluation indexes corresponding to the three types of data.
For example, under the condition that the vegetation index data is NPP data and the importance ranking result and the accuracy of the driving factors of NPPs in the Guizhou province in 2000, 2005 and 2010 are obtained, the key driving factors are selected according to the 8 driving factors of NPPs of the first combined cluster type in the Guizhou province in 2000, 2005 and 2010, and after comparison, the driving factor with the highest importance in 2000 and 2005 and the driving factor with the highest importance in 2010 population density are determined.
In addition, although the number of key driving factors is specifically defined in the present embodiment, it should be understood by those skilled in the art that the number of key driving factors may be set according to actual requirements, which is not limited in the present application.
It should be noted that, although step S105 shows that the importance analysis is performed on the driving factors of the overall vegetation index data that is not clustered, the driving factors of the vegetation index belonging to the cluster type, and the driving factors of the vegetation index belonging to the combined cluster type, those skilled in the art may also perform the setting according to the actual requirements.
For example, the driving factors of the vegetation indexes belonging to the cluster type and the combined cluster type may be analyzed for importance, respectively, and the embodiment of the present application is not limited thereto.
Compared with the prior art, the method provided by the embodiment of the application has the advantages that firstly, the spatial autocorrelation analysis is carried out on the whole to obtain the cluster type, then, the result of the cluster type is recombined to obtain the combined cluster type, then, the output result of the random forest modeling can be used for determining, and compared with the combined data and the cluster type, the recombined part of the combined cluster type can improve the accuracy of the result.
It should be noted that, the specific cluster types corresponding to the partial combined cluster types may be different in different scenes.
In addition, the importance difference of different driving factors of vegetation indexes of a data set meeting the precision requirement can be obtained through the random forest model, so that scientific basis can be provided for researching regional vegetation monitoring and conservation, ecological engineering effect evaluation, forest management scheme programming and management policy formulation.
In addition, the scheme of the application has stronger applicability, and solves the problems that certain schemes in the prior art are low in precision and only suitable for specific scenes.
It should be noted that although the operations of the method of the present application are depicted in the drawings in a particular order, this does not require or imply that the operations must be performed in that particular order or that all of the illustrated operations be performed in order to achieve desirable results. The steps depicted in the flow diagrams may change the order of execution, as the case may be. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step to perform, and/or one step decomposed into multiple steps to perform.
It should be understood that the above-described driving factor analysis method of vegetation index is merely exemplary, and those skilled in the art can vary according to the specific circumstances according to the above-described method.
As shown in fig. 3, the embodiment of the application further provides a driving factor analysis device 300 for the regional vegetation index. It should be understood that the apparatus 300 corresponds to the method embodiment of fig. 1 to 2, and is capable of performing the steps involved in the method embodiment, and specific functions of the apparatus 300 may be referred to in the foregoing description, and detailed descriptions thereof are omitted herein as appropriate to avoid redundancy. The device 300 includes at least one software functional module that can be stored in memory in the form of software or firmware (firmware) or cured in an Operating System (OS) of the device 300. Specifically, the driving factor analysis device 300 includes:
a first obtaining module 310, configured to obtain vegetation index data and at least two driving factor data of a vegetation index of the research area; the first analysis module 320 is configured to perform local spatial autocorrelation analysis on the vegetation index data to obtain an analysis result including multiple cluster types; a second obtaining module 330, configured to combine different cluster types according to a preset combination rule to obtain at least one combined cluster type; the second analysis module 340 is configured to perform importance analysis on driving factors of vegetation indexes belonging to each of at least one combined cluster type by using a random forest model.
In one possible embodiment, the vegetation index data comprises net primary productivity data; driving factor data for net primary productivity including at least two of normalized vegetation index data, population density data, domestic total production data, night light brightness data, digital elevation model data, annual average rainfall data, annual average air temperature data, and grade data.
In one possible embodiment, an importing module (not shown) for importing vegetation index data into a geographic information system; the first analysis module 320 is further configured to obtain, by using the geographic information system, an analysis result including a plurality of cluster types, where the cluster types include a high-value cluster HH, a low-value cluster LL, an outlier HL with a high value mainly surrounded by a low value, an outlier LH with a low value mainly surrounded by a high value, and an insignificant point.
In one possible embodiment, the at least one combined cluster type comprises at least one of the following types: the first, second, third, fourth, and fifth combined cluster types, the second acquisition module 330 is further configured to: combining the high-value cluster HH and the low-value cluster LL to obtain a first combined cluster type; and/or combining the high-value cluster HH and the abnormal value HL with the high value mainly surrounded by the low value to obtain a second combined cluster type; and/or combining the low-value cluster LL and the abnormal value LH with the low value mainly surrounded by the high value to obtain a third combined cluster type; and/or combining the abnormal value HL with the high value mainly surrounded by the low value and the abnormal value LH with the low value mainly surrounded by the high value to obtain a fourth combined cluster type; and/or combining the high-value cluster HH, the low-value cluster LL, the outlier HL with the high value mainly surrounded by the low value and the outlier LH with the low value mainly surrounded by the high value to obtain a fifth combined cluster type.
In a possible embodiment, the first analysis module 320 is further configured to obtain combined data formed by combining all the cluster types.
In one possible embodiment, a third obtaining module (not shown) is configured to obtain a plurality of data sets according to the combined data, the plurality of cluster types, and the at least one combined cluster type, where the plurality of data sets includes driving factor data of vegetation index data corresponding to the combined data and vegetation index belonging to the combined data, driving factor data of vegetation index data corresponding to each of the plurality of cluster types and vegetation index belonging to each of the plurality of cluster types, and driving factor data of vegetation index data corresponding to each of the at least one combined cluster type and vegetation index belonging to each of the at least one combined cluster type.
In a possible embodiment, the second analysis module 340 is further configured to obtain, using the random forest model and the plurality of data sets, an importance ranking result of the driving factors of the vegetation indexes belonging to the combined data, an importance ranking result of the driving factors of the vegetation indexes of each of the plurality of cluster types, and an importance ranking result of the driving factors of the vegetation indexes of each of the at least one combined cluster type, respectively.
In a possible embodiment, the second analysis module 340 is further configured to obtain, using the random forest model and the plurality of data sets, a combined precision evaluation index corresponding to the combined data, a first precision evaluation index corresponding to each of the plurality of cluster types, and a second precision evaluation index corresponding to each of the at least one combined cluster type, respectively.
It should be appreciated that the units or modules described in the driving factor analysis device 300 of the vegetation index correspond to the respective steps in the method described with reference to fig. 1. Thus, the operations and features described above with respect to the method are equally applicable to the driving factor analysis device 300 for vegetation indexes and the units contained therein, and are not described herein. The driving factor analysis device 300 of the vegetation index may be implemented in a browser of the electronic device or other security applications in advance, or may be loaded into the browser of the electronic device or the security applications thereof by downloading or the like. The corresponding units in the driving factor analysis device 300 of the vegetation index may cooperate with the units in the electronic device to implement the solution of the embodiment of the present application.
The embodiment of the application also provides a storage medium, wherein the storage medium is stored with a driving factor analysis program of the regional vegetation index, and the driving factor analysis program of the regional vegetation index realizes the steps of the driving factor analysis method of the regional vegetation index shown in fig. 1 when being executed by a processor.
It is to be understood that the embodiments described herein may be implemented in hardware, software, firmware, middleware, microcode, or a combination thereof. For a hardware implementation, the processor may be implemented within one or more application specific integrated circuits (Application Specific Integrated Circuit, ASIC), digital signal processors (Digital Signal Processor, DSP), programmable logic devices (Programmable Logic Device, PLD), field programmable gate arrays (Field-Programmable Gate Array, FPGA), general purpose processors, controllers, microcontrollers, microprocessors, other electronic units designed to perform the functions of the application, or a combination thereof.
For a software implementation, the techniques herein may be implemented by means of units that perform the functions herein. The software codes may be stored in a memory and executed by a processor. The memory may be implemented within the processor or external to the processor.
It should be noted that the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
Claims (6)
1. A driving factor analysis method for a regional vegetation index, comprising:
acquiring vegetation index data and at least two driving factor data of vegetation indexes of a research area;
carrying out local space autocorrelation analysis on the vegetation index data to obtain analysis results containing multiple clustering types;
combining different cluster types according to a preset combination rule to obtain at least one combined cluster type;
carrying out importance analysis on driving factors of vegetation indexes belonging to each combination cluster type in the at least one combination cluster type by utilizing a random forest model;
the performing local spatial autocorrelation analysis on the vegetation index data to obtain an analysis result including a plurality of cluster types, including: importing the vegetation index data into a geographic information system; obtaining analysis results containing a plurality of cluster types through the geographic information system, wherein the cluster types comprise a high-value cluster HH, a low-value cluster LL, an abnormal value HL with a high value mainly surrounded by a low value, an abnormal value LH with a low value mainly surrounded by a high value and an insignificant point or obtaining combined data formed by combining all the cluster types;
Before the importance analysis of the driving factors of the vegetation indexes belonging to each of the at least one combined cluster type by using the random forest model, the driving factor analysis method further comprises the steps of: acquiring a plurality of data sets according to the combination data, the plurality of cluster types and the at least one combination cluster type, wherein the plurality of data sets comprise vegetation index data corresponding to the combination data and driving factor data of vegetation indexes belonging to the combination data, vegetation index data corresponding to each cluster type in the plurality of cluster types and driving factor data of vegetation indexes belonging to each cluster type in the plurality of cluster types, vegetation index data corresponding to each combination cluster type in the at least one combination cluster type and driving factor data of vegetation indexes belonging to each combination cluster type in the at least one combination cluster type;
the importance analysis of the driving factors of the vegetation indexes belonging to each of the at least one combined cluster type by using the random forest model comprises the following steps: and respectively acquiring an importance ranking result of the driving factors of the vegetation indexes belonging to the combined data, an importance ranking result of the driving factors of the vegetation indexes of each cluster type in the plurality of cluster types and an importance ranking result of the driving factors of the vegetation indexes of each combined cluster type in the at least one combined cluster type by utilizing the random forest model and the plurality of data sets.
2. The driving factor analysis method according to claim 1, wherein the vegetation index data comprises net primary productivity data;
the driving factor data for net primary productivity includes at least two of normalized vegetation index data, population density data, domestic total production data, night light brightness data, digital elevation model data, annual average rainfall data, annual average air temperature data, and grade data.
3. The driving factor analysis method according to claim 1, wherein the at least one combined cluster type includes at least one of the following types: the method comprises the steps of combining different cluster types according to a preset combination rule to obtain at least one combined cluster type, wherein the first combined cluster type, the second combined cluster type, the third combined cluster type, the fourth combined cluster type and the fifth combined cluster type comprise the following steps:
combining the high-value cluster HH and the low-value cluster LL to obtain the first combined cluster type; and/or the number of the groups of groups,
combining the high-value cluster HH and an abnormal value HL of which the high value is mainly surrounded by a low value to obtain the second combined cluster type; and/or the number of the groups of groups,
Combining the low-value cluster LL and the abnormal value LH of which the low value is mainly surrounded by a high value to obtain the third combined cluster type; and/or the number of the groups of groups,
combining the abnormal value HL of which the high value is mainly surrounded by the low value and the abnormal value LH of which the low value is mainly surrounded by the high value to obtain the fourth combined cluster type; and/or the number of the groups of groups,
and combining the high-value cluster HH, the low-value cluster LL, the outlier HL with the high value mainly surrounded by the low value and the outlier LH with the low value mainly surrounded by the high value to obtain the fifth combined cluster type.
4. The driving factor analysis method according to claim 1, characterized in that the driving factor analysis method further comprises:
and respectively acquiring a combined precision evaluation index corresponding to the combined data, a first precision evaluation index corresponding to each of the plurality of cluster types and a second precision evaluation index corresponding to each of the at least one combined cluster type by utilizing the random forest model and the plurality of data sets.
5. A driving factor analysis device for a regional vegetation index, comprising:
the first acquisition module is used for acquiring vegetation index data and at least two driving factor data of vegetation indexes of the research area;
The first analysis module is used for carrying out local spatial autocorrelation analysis on the vegetation index data so as to obtain analysis results containing multiple clustering types;
the second acquisition module is used for combining different cluster types according to a preset combination rule to acquire at least one combined cluster type;
the second analysis module is used for carrying out importance analysis on driving factors of vegetation indexes belonging to each combination cluster type in the at least one combination cluster type by utilizing a random forest model;
the first analysis module is specifically configured to: importing the vegetation index data into a geographic information system; obtaining analysis results containing a plurality of cluster types through the geographic information system, wherein the cluster types comprise a high-value cluster HH, a low-value cluster LL, an abnormal value HL with a high value mainly surrounded by a low value, an abnormal value LH with a low value mainly surrounded by a high value and an insignificant point or obtaining combined data formed by combining all the cluster types;
the apparatus further comprises: a third obtaining module, configured to obtain, by the second analyzing module, a plurality of data sets according to the combined data, the plurality of cluster types, and the at least one combined cluster type before performing importance analysis on driving factors of vegetation indexes belonging to each of the at least one combined cluster type using a random forest model, where the plurality of data sets include driving factor data of vegetation index data corresponding to the combined data and vegetation indexes belonging to the combined data, driving factor data of vegetation index data corresponding to each of the plurality of cluster types and vegetation index data corresponding to each of the plurality of cluster types, and driving factor data of vegetation index data corresponding to each of the at least one combined cluster type and vegetation index data corresponding to each of the at least one combined cluster type;
The second analysis module is specifically configured to: and respectively acquiring an importance ranking result of the driving factors of the vegetation indexes belonging to the combined data, an importance ranking result of the driving factors of the vegetation indexes of each cluster type in the plurality of cluster types and an importance ranking result of the driving factors of the vegetation indexes of each combined cluster type in the at least one combined cluster type by utilizing the random forest model and the plurality of data sets.
6. A storage medium, wherein a driving factor analysis program for a regional vegetation index is stored on the storage medium, and the driving factor analysis program for a regional vegetation index, when executed by a processor, implements the steps of the driving factor analysis method for a regional vegetation index according to any one of claims 1 to 4.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2019113748920 | 2019-12-27 | ||
| CN201911374892 | 2019-12-27 |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111259963A CN111259963A (en) | 2020-06-09 |
| CN111259963B true CN111259963B (en) | 2023-10-13 |
Family
ID=70950728
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010051514.5A Active CN111259963B (en) | 2019-12-27 | 2020-01-16 | Driving factor analysis method and device for regional vegetation index and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111259963B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113076796B (en) * | 2021-02-08 | 2022-09-23 | 中国科学院地理科学与资源研究所 | Karst stony desertification remote sensing mapping method and device |
| CN113095285A (en) * | 2021-04-30 | 2021-07-09 | 郑州大学 | Method for quantitatively analyzing vegetation space-time evolution driving mechanism based on pixel scale |
| CN112990782B (en) * | 2021-04-30 | 2021-08-06 | 浙江甲骨文超级码科技股份有限公司 | Crop growth environment intelligent adjusting method and device based on time-space autocorrelation analysis |
| CN117612038A (en) * | 2023-11-22 | 2024-02-27 | 中国矿业大学(北京) | Mining area vegetation carbon sink fine calculation method based on unmanned aerial vehicle image |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106199557A (en) * | 2016-06-24 | 2016-12-07 | 南京林业大学 | A kind of airborne laser radar data vegetation extracting method |
| CN108053072A (en) * | 2017-12-22 | 2018-05-18 | 中国科学院地理科学与资源研究所 | The structure of desertification dynamic simulation model and application |
| CN109033599A (en) * | 2018-07-18 | 2018-12-18 | 福州大学 | A kind of Influencing Factor of Soil Erosion importance analysis method based on random forest |
| CN109885600A (en) * | 2019-02-25 | 2019-06-14 | 甘肃省祁连山水源涵养林研究院 | Forest cover growth rate analysis method and system |
| CN110135368A (en) * | 2019-05-20 | 2019-08-16 | 太原理工大学 | Based on spatial autocorrelation regional vegetation net primary productivity spatio-temporal variability detection method |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10121106B2 (en) * | 2014-06-30 | 2018-11-06 | Digitalglobe, Inc. | Spectral data analytic cube classifier (SPADACC) for continuous wide area geospatial modeling using normalized and highly dimensional multispectal raster data |
-
2020
- 2020-01-16 CN CN202010051514.5A patent/CN111259963B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106199557A (en) * | 2016-06-24 | 2016-12-07 | 南京林业大学 | A kind of airborne laser radar data vegetation extracting method |
| CN108053072A (en) * | 2017-12-22 | 2018-05-18 | 中国科学院地理科学与资源研究所 | The structure of desertification dynamic simulation model and application |
| CN109033599A (en) * | 2018-07-18 | 2018-12-18 | 福州大学 | A kind of Influencing Factor of Soil Erosion importance analysis method based on random forest |
| CN109885600A (en) * | 2019-02-25 | 2019-06-14 | 甘肃省祁连山水源涵养林研究院 | Forest cover growth rate analysis method and system |
| CN110135368A (en) * | 2019-05-20 | 2019-08-16 | 太原理工大学 | Based on spatial autocorrelation regional vegetation net primary productivity spatio-temporal variability detection method |
Non-Patent Citations (1)
| Title |
|---|
| 姬盼盼 ; 高敏华 ; 杨晓东 ; .中国西北部干旱区NPP驱动力分析――以新疆伊犁河谷和天山山脉部分区域为例.生态学报.2019,(08),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111259963A (en) | 2020-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111259963B (en) | Driving factor analysis method and device for regional vegetation index and storage medium | |
| Stehman | Sampling designs for accuracy assessment of land cover | |
| Breidenbach et al. | Comparison of nearest neighbour approaches for small area estimation of tree species-specific forest inventory attributes in central Europe using airborne laser scanner data | |
| CN111862292B (en) | Data rendering method and device for transmission line corridor and computer equipment | |
| Konstantelos et al. | Using vine copulas to generate representative system states for machine learning | |
| CN106651025B (en) | Traffic condition prediction method and device | |
| US11676375B2 (en) | System and process for integrative computational soil mapping | |
| CN109543907B (en) | Complex terrain wind resource assessment method and device | |
| Rust et al. | Quantifying differences in circulation patterns based on probabilistic models: IPCC AR4 multimodel comparison for the North Atlantic | |
| CN108959187A (en) | A kind of variable branch mailbox method, apparatus, terminal device and storage medium | |
| US11790410B2 (en) | System and method for natural capital measurement | |
| CN103824223A (en) | Crop yield remote sensing estimation method based on MapReduce and neural network | |
| CN108830554A (en) | The outcome data information quality intelligent detecting method and system of task based access control model | |
| Sačkov et al. | Integration of tree allometry rules to treetops detection and tree crowns delineation using airborne lidar data | |
| Vauhkonen et al. | Selecting the trees to be harvested based on the relative value growth of the remaining trees | |
| Ho et al. | Amic: An adaptive information theoretic method to identify multi-scale temporal correlations in big time series data | |
| CN111932146A (en) | Method and device for analyzing pollution cause, computer equipment and readable storage medium | |
| CN116416884A (en) | Testing device and testing method for display module | |
| Miettinen et al. | Demonstration of large area forest volume and primary production estimation approach based on Sentinel-2 imagery and process based ecosystem modelling | |
| CN115545331A (en) | Control strategy prediction method and device, equipment and storage medium | |
| CN117235524A (en) | Learning training platform of automatic valuation model | |
| Dutta et al. | Performance evaluation of south Esk hydrological sensor web: unsupervised machine learning and semantic linked data approach | |
| CN117493798A (en) | Meteorological environment data analysis method and system | |
| CN116822366A (en) | Construction of runoff pollution load calculation model and runoff pollution load calculation method | |
| Schuh et al. | Machine learning and generalized linear model techniques to predict aboveground biomass in Amazon rainforest using LiDAR data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |