Remote sensing data-based cultivated land data analysis method, apparatus and device, and medium

Abstract

The invention relates to the field of data analysis and financial science and technology, and discloses a remote sensing data-based cultivated land data analysis method, device, equipment and medium, and the method comprises the steps: respectively obtaining total primary productivity remote sensing data, land utilization data and district and county boundary data; performing boundary cutting on the total primary productivity remote sensing data by using the district and county boundary data to obtain district and county productivity remote sensing data; performing mask operation on the remote sensing data of the productivity of the district and county according to the land utilization data to obtain cultivated land productivity data; performing segmentation center screening on the cultivated land productivity data to obtain a segmentation clustering center set; performing dynamic super-pixel segmentation on the cultivated land productivity data according to the segmentation clustering center set to obtain a regional cultivated land productivity data set; and grading the regional cultivated land productivity data set to obtain a cultivated land productivity data set. And the division fine granularity and accuracy of the cultivated land fertility division are improved.

Description

Technical Field

[0001] The present invention relates to the fields of data analysis and financial technology, and in particular to a method, device, equipment and medium for analyzing cultivated land data based on remote sensing data. Background Art

[0002] In agricultural insurance business, it is necessary to evaluate the soil fertility grade of the insured land. In previous practice, this evaluation is generally based on the national arable land fertility standards. However, the grade is generally divided according to large regions across the country. There is no distinction in a smaller area, which makes it difficult to meet the actual business needs.

[0003] Existing arable land fertility grade assessment schemes require a lot of agricultural background knowledge and rely on many types of data, including soil conditions such as soil quality, various trace elements in the soil, climate conditions, topography, and other data. The distribution of surrounding infrastructure such as water conservancy must also be considered. The acquisition, processing and integration of these data are all very difficult, which makes the granularity of arable land fertility grade classification larger, and thus the accuracy of arable land fertility grade classification is lower. Summary of the invention

[0004] The present invention provides an artificial intelligence-based remote sensing data-based cultivated land data analysis method, device, equipment and medium to solve the technical problem that the granularity of cultivated land fertility classification is large, resulting in low accuracy of cultivated land fertility classification.

[0005] In a first aspect, a method for analyzing cultivated land data based on remote sensing data is provided, comprising:

[0006] Obtain remote sensing data on gross primary productivity, land use data, and district and county boundary data respectively;

[0007] Using the district and county boundary data to perform boundary cutting on the total primary productivity remote sensing data, to obtain district and county productivity remote sensing data;

[0008] Performing a mask operation on the county productivity remote sensing data according to the land use data to obtain cultivated land productivity data;

[0009] Screening the segmentation centers of the cultivated land productivity data to obtain a segmentation cluster center set;

[0010] Performing dynamic superpixel segmentation on the cultivated land productivity data according to the segmentation cluster center set to obtain a regional cultivated land productivity data set;

[0011] The regional cultivated land productivity dataset is graded to obtain a cultivated land fertility dataset.

[0012] In a second aspect, a farmland data analysis device based on remote sensing data is provided, comprising:

[0013] The data acquisition module is used to obtain remote sensing data of total primary productivity, land use data and district and county boundary data respectively;

[0014] A boundary segmentation module is used to perform boundary segmentation on the total primary productivity remote sensing data using the district and county boundary data to obtain district and county productivity remote sensing data;

[0015] A mask segmentation module, used for performing a mask operation on the county productivity remote sensing data according to the land use data to obtain cultivated land productivity data;

[0016] A center screening module, used for screening the segmentation centers of the cultivated land productivity data to obtain a segmentation cluster center set;

[0017] A superpixel segmentation module, used for performing dynamic superpixel segmentation on the cultivated land productivity data according to the segmentation cluster center set to obtain a regional cultivated land productivity data set;

[0018] The grading module is used to grade the regional cultivated land productivity dataset to obtain the cultivated land productivity dataset.

[0019] In a third aspect, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned method for analyzing cultivated land data based on remote sensing data when executing the computer program.

[0020] In a fourth aspect, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the above-mentioned method for analyzing cultivated land data based on remote sensing data is implemented.

[0021] In the scheme implemented by the above-mentioned remote sensing data-based cultivated land data analysis method, device, computer equipment and storage medium, the client can communicate with the server through the network of the client, and the server can obtain the total primary productivity remote sensing data, land use data and district and county boundary data through the client respectively; the district and county boundary data are used to perform boundary cutting on the total primary productivity remote sensing data to obtain the district and county productivity remote sensing data; the district and county productivity remote sensing data are masked according to the land use data to obtain the cultivated land productivity data; the segmentation center of the cultivated land productivity data is screened to obtain the segmentation cluster center set; the cultivated land productivity data is dynamically superpixel segmented according to the segmentation cluster center set to obtain the regional cultivated land productivity data set ; The regional cultivated land productivity data set is graded to obtain a cultivated land productivity data set. In the present invention, for the graded division scheme when the segmentation granularity is relatively small, a dynamic super-resolution division method can be used. First, the cultivated land productivity data is gridded and the center points are screened to obtain a segmentation cluster center set. Then, the productivity data of each pixel in the cultivated land productivity data is aligned using a dynamic time warping method. Subsequently, the cultivated land productivity data is refined and split using an image segmentation algorithm to obtain a super-pixel cultivated land productivity data set. Then, the super-pixel cultivated land productivity data set is divided into regional cultivated land productivity data sets using a clustering method, thereby improving the granularity of cultivated land division and then improving the accuracy of cultivated land grade division. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.

[0023] Figure 1 is a schematic diagram of an application environment of a method for analyzing cultivated land data based on remote sensing data in one embodiment of the present invention;

[0024] Figure 2 is a flow chart of a method for analyzing cultivated land data based on remote sensing data in one embodiment of the present invention;

[0025] Figure 3 yes Figure 1 A schematic flow chart of a specific implementation of step S20;

[0026] Figure 4 yes Figure 1 A schematic flow chart of a specific implementation of step S30;

[0027] Figure 5 yes Figure 4A schematic flow chart of a specific implementation of step S40;

[0028] Figure 6 is a structural schematic diagram of a farmland data analysis device based on remote sensing data in one embodiment of the present invention;

[0029] Figure 7 is a schematic diagram of a structure of a computer device in one embodiment of the present invention;

[0030] Figure 8 It is another structural schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0031] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0032] The method for analyzing cultivated land data based on remote sensing data provided by the embodiment of the present invention can be applied in the following aspects: Figure 1 In the application environment, the client communicates with the server through the network. The server can obtain the total primary productivity remote sensing data, land use data and district and county boundary data through the client; use the district and county boundary data to perform boundary cutting on the total primary productivity remote sensing data to obtain the district and county productivity remote sensing data; perform masking operations on the district and county productivity remote sensing data according to the land use data to obtain the cultivated land productivity data; perform segmentation center screening on the cultivated land productivity data to obtain the segmentation cluster center set; perform dynamic superpixel segmentation on the cultivated land productivity data according to the segmentation cluster center set to obtain the regional cultivated land productivity data set; perform grade classification on the regional cultivated land productivity data set to obtain the cultivated land soil fertility data set, In the present invention, for the grading scheme when the segmentation granularity is relatively small, the dynamic super-resolution segmentation method can be used. First, the cultivated land productivity data is gridded and the center point is screened to obtain the segmentation cluster center set, and then the productivity data of each pixel in the cultivated land productivity data is aligned by the dynamic time warping method, and then the cultivated land productivity data is refined and split by the image segmentation algorithm to obtain a super-pixel cultivated land productivity data set; the super-pixel cultivated land productivity data set is then divided into regional cultivated land productivity data sets by the clustering method, so as to improve the granularity of cultivated land division and then improve the accuracy of cultivated land grading. Among them, the client can be but not limited to various personal computers, laptops, smart phones, tablet computers and portable wearable devices. The server can be implemented by an independent server or a server cluster composed of multiple servers. The present invention is described in detail below through specific embodiments.

[0033] See also Figure 2 As shown, Figure 2 A flow chart of a method for analyzing cultivated land data based on remote sensing data provided by an embodiment of the present invention includes the following steps:

[0034] S10. Obtain remote sensing data on total primary productivity, land use data, and district and county boundary data respectively.

[0035] In detail, the gross primary productivity remote sensing data refers to the remote sensing data of the primary productivity (Gross Primary Productivity, GPP for short) of a certain area. The gross primary productivity remote sensing data is a kind of data that estimates the total amount of carbon fixed by vegetation on the earth's surface during photosynthesis through remote sensing technology. The primary productivity reflects the amount of carbon dioxide absorbed by plants in the ecosystem from the atmosphere through photosynthesis, and then converted into organic matter. The primary productivity is an important indicator for assessing the carbon cycle of the ecosystem, the health of vegetation and biological productivity.

[0036] Specifically, the land use data is data that describes the use and coverage of land on the earth's surface, reflecting how human activities utilize land resources and the natural state of the land. Land use data usually includes different categories of land cover information, such as cultivated land, forest land, grassland, urban construction land, water bodies, wetlands, etc. Through land use data, the use of land can be evaluated, and resource planning, environmental monitoring, ecological assessment and other tasks can be carried out.

[0037] In detail, the sources of the land use data may include remote sensing data, geographic information system data, field survey data, etc. Commonly used land use classification systems such as China's "National Land Use Status Classification Standard", the United States' NLCD, and the global ESA's global land cover dataset.

[0038] Specifically, the district and county boundary data refers to spatial data that describes the geographic boundaries of various districts, counties or similar administrative units within a country or region. The district and county boundary data are usually used in geographic information systems and map applications to help users visualize, analyze and process the geographic scope and shape of specific administrative regions.

[0039] In detail, the total primary productivity remote sensing data can be the primary productivity remote sensing data of the European Space Agency, the data format is GeoTiff format, the time range is the past 8 years, the spatial range is the Chinese region, and the spatial resolution is 300 meters; the land use data can be a 10-meter resolution land use data set in the Chinese region, and the land use data can be obtained from the Fine Resolution Observation and Monitoring of Global Land Cover 2017 (FROM GLC 2017 for short), the data format of the land use data is GeoTiff format, and the data format of the district and county boundary data is shapefile format.

[0040] In the embodiment of the present invention, by respectively acquiring the total primary productivity remote sensing data, the land use data and the district and county boundary data, the land fertility grade classification can be abstracted into a digital image processing scheme, thereby improving the classification efficiency.

[0041] S20. Using the district and county boundary data, the total primary productivity remote sensing data is subjected to boundary cutting to obtain district and county productivity remote sensing data.

[0042] In detail, the district and county productivity remote sensing data refers to the primary productivity remote sensing data corresponding to a district and county in the total primary productivity remote sensing data. For example, when the total primary productivity remote sensing data is the primary productivity remote sensing data in the national area, and the district and county boundary data is the boundary data of district and county A, the district and county productivity remote sensing data refers to the primary productivity remote sensing data corresponding to district and county A. The district and county productivity remote sensing data is ten-day data, with three data points in the upper, middle and lower ten-day periods of each month. In this way, there are a total of 12*3*8=288 data points in each pixel of the district and county productivity remote sensing data.

[0043] In the embodiment of the present invention, referring to Figure 3 As shown, the use of the district and county boundary data to perform boundary cutting on the total primary productivity remote sensing data to obtain district and county productivity remote sensing data includes:

[0044] S21, extracting a boundary scale from the district and county boundary data;

[0045] S22, extracting a remote sensing scale from the total primary productivity remote sensing data;

[0046] S23, calculating the boundary scaling factor between the boundary scale and the remote sensing scale;

[0047] S24, scaling the district and county boundary data according to the boundary scaling factor to obtain a standard district and county boundary;

[0048] S25. Using the standard district and county boundaries to perform boundary cutting on the total primary productivity remote sensing data, to obtain district and county productivity remote sensing data.

[0049] In detail, the boundary scale refers to the scale data of the district and county boundary data. The district and county boundary data and boundary scale can be extracted using Python's geopandas package and Rasterio library or the open source geographic information system tool QGIS. The remote sensing scale refers to the scale data corresponding to the total primary productivity remote sensing data.

[0050] In detail, the boundary scaling factor between the boundary scale and the remote sensing scale is calculated. For example, when the remote sensing scale is 1:100 and the boundary scale is 1:300, the scaling factor is 3, that is, the boundary vector of the district and county boundary data needs to be magnified three times to obtain the standard district and county boundary.

[0051] Specifically, the gdalwarp tool of the open source geographic information system tool QGIS can be used to scale the district and county boundary data using the remote sensing scale and the boundary scale to obtain a standard district and county boundary.

[0052] In detail, the method of using the standard county boundary to perform boundary cutting on the total primary productivity remote sensing data to obtain the county productivity remote sensing data includes taking the total primary productivity remote sensing data as a raster and the standard county boundary as a mask layer, using the open source geographic information system tool QGIS to perform an operation of clipping the raster to the mask layer, and filling areas with no data with no data values, saving the results, and obtaining the county productivity remote sensing data, wherein the primary productivity remote sensing data can be clipped using the gdalwarp or gdal_translate tool.

[0053] In an embodiment of the present invention, by using the district and county boundary data to perform boundary cutting on the total primary productivity remote sensing data, the district and county productivity remote sensing data is obtained, and the productivity remote sensing data of the districts and counties that need to be divided into cultivated land productivity can be screened out, thereby achieving accurate cultivated land productivity division.

[0054] S30. Perform a masking operation on the county productivity remote sensing data according to the land use data to obtain cultivated land productivity data.

[0055] In detail, the cultivated land productivity data refers to the primary productivity data of the cultivated land area of ​​the district and county corresponding to the district and county productivity remote sensing data, and the land use data includes the land type of each area, wherein the land type includes cultivated land type, forestry type and residential type, etc. The cultivated land productivity data refers to the primary productivity data of the land area whose land type is cultivated land type in the district and county productivity remote sensing data.

[0056] In the embodiment of the present invention, referring to Figure 4 As shown, the masking operation is performed on the district and county productivity remote sensing data according to the land use data to obtain the cultivated land productivity data, including:

[0057] S31, extracting cultivated land type land data from the land use data;

[0058] S32, extracting the cultivated land scale from the cultivated land type land data;

[0059] S33, extracting a remote sensing scale from the cultivated land productivity data;

[0060] S34, calculating the cultivated land scaling factor between the cultivated land scale and the remote sensing scale;

[0061] S35, scaling the cultivated land boundary data according to the cultivated land scaling factor to obtain scaled cultivated land data;

[0062] S36, generating farmland mask data according to the scaled farmland data;

[0063] S37. Use the cultivated land mask data to perform a mask operation on the cultivated land productivity data to obtain the cultivated land productivity data.

[0064] In detail, the cultivated land type land data refers to the land area data whose land use type is cultivated land type in the land use data. The cultivated land type land data can be extracted using the LandUse function in the open source geographic information system tool QGIS. The LandUse function is a function used for land classification in the land use data. For example, "LandUse" = 1 means extracting data whose land type is cultivated land type in the land use data.

[0065] Specifically, the cultivated land scale refers to the scale data corresponding to the land use data, and the remote sensing scale is the same as the remote sensing scale in the above step S20, which will not be repeated here.

[0066] In detail, the command line gdal_calc.py in the open source geographic information system tool QGIS can be used to generate farmland mask data according to the scaled farmland data. The use of the farmland mask data to perform a mask operation on the farmland productivity data to obtain the farmland productivity data means taking the farmland productivity data as a raster and the base mask data as a mask layer, using the open source geographic information system tool QGIS to perform an operation of clipping the raster to the mask layer, and filling the area without data with a no data value, saving the result, and obtaining the farmland productivity data, wherein the farmland productivity data can be clipped using the gdalwarp or gdal_translate tool.

[0067] In an embodiment of the present invention, by performing a masking operation on the district and county productivity remote sensing data according to the land use data, the cultivated land productivity data can be obtained, and the primary productivity data of the cultivated land area in the target district and county area can be removed, so as to facilitate the subsequent cultivated land productivity division operation.

[0068] S40, screening the segmentation centers of the cultivated land productivity data to obtain a segmentation cluster center set.

[0069] In detail, the segmentation cluster center set refers to the cluster centers selected when segmenting the cultivated land productivity data into productivity data of multiple sub-regions.

[0070] In the embodiment of the present invention, referring to Figure 5 As shown, the segmentation center screening of the cultivated land productivity data is performed to obtain a segmentation cluster center set, including:

[0071] S41, performing area statistics on the cultivated land productivity data to obtain the total cultivated land area;

[0072] S42, calculating the area of ​​the sub-cultivated land according to the total area of ​​the cultivated land and the preset number of divisions;

[0073] S43, initializing a segmentation grid according to the sub-cultivated land area;

[0074] S44, gridding the cultivated land productivity data according to the segmentation grids to obtain a segmented cultivated land grid data set;

[0075] S45, gathering the grid midpoints corresponding to the segmented cultivated land grid data set into a segmentation cluster center set.

[0076] In detail, performing area statistics on the cultivated land productivity data to obtain the total cultivated land area includes extracting resolution data of the cultivated land productivity data; calculating the single-pixel area of ​​the cultivated land productivity data; and calculating the total cultivated land area of ​​the cultivated land productivity data based on the single-pixel area, wherein the resolution data refers to data such as the length and width of a single pixel in the cultivated land productivity data. The resolution data of the cultivated land productivity data can be extracted using the gdalinfo tool of the open source geographic information system tool QGIS, wherein the output of the Pixel Size field is the resolution data, the single-pixel area is the product data of the length greater than the width in the resolution data, and calculating the total cultivated land area of ​​the cultivated land productivity data based on the single-pixel area refers to counting the total number of pixels in the cultivated land productivity data, and multiplying the total number of pixels by the single-pixel area to obtain the total cultivated land area.

[0077] Specifically, the number of divisions refers to the number of areas that need to be divided, and the number of divisions may be 1000. Calculating the sub-cultivated land area based on the total cultivated land area and the preset number of divisions refers to dividing the total cultivated land area by the number of divisions to obtain the sub-cultivated land area.

[0078] In detail, initializing the segmentation grid according to the sub-arable land area refers to generating a uniformly distributed grid with the sub-arable land area as the grid size, and gridding the arable land productivity data according to the segmentation grid to obtain a segmented arable land grid data set refers to segmenting the arable land productivity data using each grid, and each segmented arable land grid data in the segmented arable land grid data set corresponds to partial data of the arable land productivity data segmented by a grid in the segmentation grid.

[0079] Specifically, the method of gathering the grid midpoints corresponding to the segmented farmland grid data set into a segmentation cluster center set refers to taking the grid midpoints of some grids corresponding to each segmented farmland grid data in the style farmland grid data set as the segmentation cluster center. If the pixel corresponding to the grid midpoint is not in the segmented farmland grid data, then the pixel point in the segmented farmland grid data that is closest to the grid midpoint is taken as the segmentation cluster center.

[0080] In an embodiment of the present invention, by screening the segmentation centers of the cultivated land productivity data to obtain a segmentation cluster center set, the initial cluster center can be screened out, thereby facilitating the subsequent uniform segmentation of the cultivated land productivity data through clustering methods, thereby improving the accuracy of soil fertility grade division.

[0081] S50, performing dynamic superpixel segmentation on the cultivated land productivity data according to the segmentation cluster center set to obtain a regional cultivated land productivity data set.

[0082] In detail, the regional farmland productivity data set refers to the farmland productivity data of each sub-region obtained after the farmland productivity data is spatially subdivided.

[0083] In an embodiment of the present invention, the farmland productivity data is dynamically segmented into superpixels according to the segmentation cluster center set to obtain a regional farmland productivity data set, including: performing time series data conversion on the farmland productivity data to obtain a productivity sequence set; aggregating the productivity sequences corresponding to the segmentation cluster center set in the productivity sequence set into a center productivity sequence set; selecting productivity sequences in the productivity sequence set one by one as target productivity sequences, and selecting center productivity sequences in the center productivity sequence set one by one as target center sequences; calculating the productivity distance between the target productivity sequence and the target center sequence; updating the productivity sequence set into a primary productivity sequence group set according to the nearest allocation method based on all productivity distances; iteratively updating the primary productivity sequence group set to obtain a productivity sequence class set; regionally merging each productivity sequence class in the productivity sequence class set to obtain a superpixel farmland productivity data set; clustering the superpixel farmland productivity data set to obtain a regional farmland productivity data set.

[0084] In detail, the time series data conversion refers to extracting the time series data of productivity corresponding to each pixel in the cultivated land productivity data, and each productivity sequence in the productivity series set corresponds to the primary productivity data of a pixel point in the cultivated land productivity data within a period of time.

[0085] Specifically, each pixel point in the cultivated land productivity data may correspond to a change sequence of primary productivity data of cultivated land of the corresponding pixel area in the district or county within the past eight years. For example, when the cultivated land productivity data is ten-day data, there are three data points in the upper, middle and lower ten-day periods each month. In this way, each pixel has a total of 12*3*8=288 data points, and the length of the productivity sequence is 288.

[0086] Specifically, the calculation of the productivity distance between the target productivity sequence and the target center sequence includes: constructing a cost matrix for the target productivity sequence and the target center sequence to obtain a productivity cost matrix; performing dynamic programming calculation on the productivity cost matrix to obtain a cumulative cost matrix; performing path screening on the cumulative cost matrix to obtain an alignment path; and generating a productivity distance based on the alignment path.

[0087] In detail, each element D(i,j) of the productivity cost matrix represents the cost between the i-th element of the target productivity sequence and the j-th element of the target center sequence, wherein D(i,j) is the element value of the i-th row and j-th column of the productivity cost matrix, and i and j are sequence indices, wherein the cost can be calculated using the Euclidean distance algorithm.

[0088] Specifically, the cumulative cost matrix represents the minimum cost from the beginning of the target productivity sequence and from the beginning of the target center sequence to the current position, and the cumulative cost matrix can provide a basis for finding the best matching path.

[0089] In detail, the dynamic programming calculation of the productivity cost matrix to obtain the cumulative cost matrix refers to the use of dynamic programming methods to update the minimum cumulative cost of each position in the cumulative cost matrix, and the cumulative cost refers to the sum of the minimum cumulative elements in the cost matrix from the starting point of the sequence to the target position.

[0090] Specifically, the alignment path refers to the path consisting of the sum of the shortest elements tracing back from the lower right corner element of the cumulative cost matrix to the upper left corner element, and the productivity distance refers to the element value of the lower right corner element of the cumulative cost matrix, that is, the element value of the first element of the alignment path.

[0091] In detail, updating the productivity sequence set into the primary productivity sequence group set according to the method of nearest allocation based on all productivity distances means allocating each productivity sequence in the productivity sequence set to the group where each central productivity sequence in the central productivity sequence set is located according to the method of nearest allocation based on productivity distances, so as to obtain the primary productivity sequence group set, wherein each primary productivity sequence group in the primary productivity sequence group set corresponds to each central productivity sequence in the central productivity sequence set.

[0092] Specifically, the KNN clustering algorithm can be used to iteratively update and perform clustering operations on the primary productivity sequence group set. The region merging refers to merging the pixels in the cultivated land productivity data corresponding to each productivity sequence class in the productivity sequence class set into sub-regions, wherein each sub-region corresponds to each super-pixel cultivated land productivity data in the super-pixel cultivated land productivity data set.

[0093] In an embodiment of the present invention, by dynamically superpixel segmenting the cultivated land productivity data according to the segmentation cluster center set, a regional cultivated land productivity data set is obtained, and the cultivated land productivity data can be divided according to the crop's acoustic field cycle and its accumulated organic matter data in different periods, thereby improving the interpretability and accuracy of the soil fertility grade division.

[0094] S60, classifying the regional cultivated land productivity dataset to obtain a cultivated land productivity dataset.

[0095] In detail, each piece of cultivated land soil fertility data in the cultivated land soil fertility data set corresponds to the regional location information and soil fertility grade information of the corresponding sub-region in the cultivated land productivity data.

[0096] In an embodiment of the present invention, the regional cultivated land productivity data set is graded to obtain a cultivated land soil fertility data set, including: calculating the productivity mean of each regional cultivated land productivity data in the regional cultivated land productivity data set to obtain a mean productivity set; sorting the mean productivity set in descending order to obtain a mean productivity sequence; grading the mean productivity sequence according to a preset sequence number range to obtain a productivity grade sequence; and grade-labeling the regional cultivated land productivity data using the productivity grade sequence to obtain a cultivated land soil fertility data set.

[0097] In detail, the productivity mean calculation refers to calculating the mean productivity data of cultivated land productivity data in each region during the growth cycle. The serial number range, for example, when the length of the mean productivity sequence is 20, 1 to 5 is a high level of soil fertility, 6 to 10 is a medium level of soil fertility, 11 to 15 is a low level of soil fertility, and 16 to 20 is a poor level of soil fertility.

[0098] In an embodiment of the present invention, by classifying the regional cultivated land productivity dataset, a cultivated land soil fertility dataset is obtained, which can be combined with the productivity data of the crop growth cycle to achieve detailed soil fertility regional division and soil fertility grade division, thereby improving the accuracy of soil fertility grade division.

[0099] It can be seen that in the above scheme, by respectively acquiring the total primary productivity remote sensing data, land use data and district and county boundary data, the land arable land productivity grade division can be abstracted into a digital image processing scheme, thereby improving the division efficiency. By using the district and county boundary data to perform boundary cutting on the total primary productivity remote sensing data, the district and county productivity remote sensing data is obtained, and the productivity remote sensing data of the counties and districts that need to be divided into arable land productivity can be screened out, thereby realizing accurate arable land productivity division. By performing masking operations on the district and county productivity remote sensing data according to the land use data, the arable land productivity data is obtained, and the primary productivity data of the arable land area in the target district and county area can be removed, so as to facilitate the subsequent arable land productivity division operation.

[0100] By screening the segmentation centers of the cultivated land productivity data to obtain a segmentation cluster center set, the initial cluster center can be screened out, so as to facilitate the subsequent uniform segmentation of the cultivated land productivity data by clustering method, thereby improving the accuracy of soil fertility grade division; by dynamically superpixel segmenting the cultivated land productivity data according to the segmentation cluster center set, a regional cultivated land productivity data set is obtained, and the cultivated land productivity data can be divided according to the sound field cycle of the crop and the organic matter data accumulated in different periods, thereby improving the interpretability and accuracy of the soil fertility grade division; by performing grade division on the regional cultivated land productivity data set, an cultivated land soil fertility data set is obtained, and detailed soil fertility area division and soil fertility grade division can be realized in combination with the productivity data of the crop growth cycle, thereby improving the accuracy of soil fertility grade division.

[0101] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0102] In one embodiment, a farmland data analysis device based on remote sensing data is provided, and the farmland data analysis device based on remote sensing data corresponds one-to-one to the farmland data analysis method based on remote sensing data in the above embodiment. Figure 6 As shown, the farmland data analysis device based on remote sensing data includes a data acquisition module 101, a boundary segmentation module 102, a mask segmentation module 103, a center screening module 104, a super pixel segmentation module 105 and a level classification module 106. The functional modules are described in detail as follows:

[0103] The data acquisition module 101 is used to respectively acquire the total primary productivity remote sensing data, the land use data and the district and county boundary data;

[0104] The boundary segmentation module 102 is used to perform boundary segmentation on the total primary productivity remote sensing data using the district and county boundary data to obtain district and county productivity remote sensing data;

[0105] The mask segmentation module 103 is used to perform a mask operation on the county productivity remote sensing data according to the land use data to obtain cultivated land productivity data;

[0106] A center screening module 104 is used to screen the segmentation centers of the cultivated land productivity data to obtain a segmentation cluster center set;

[0107] A superpixel segmentation module 105 is used to perform dynamic superpixel segmentation on the farmland productivity data according to the segmentation cluster center set to obtain a regional farmland productivity data set;

[0108] The classification module 106 is used to classify the regional cultivated land productivity dataset to obtain a cultivated land productivity dataset.

[0109] In one embodiment, when the boundary segmentation module 102 uses the district and county boundary data to perform boundary segmentation on the total primary productivity remote sensing data to obtain the district and county productivity remote sensing data, it is used to:

[0110] Extracting a boundary scale from the district and county boundary data;

[0111] extracting a remote sensing scale from the total primary productivity remote sensing data;

[0112] Calculating a boundary scaling factor between the boundary scale and the remote sensing scale;

[0113] Scaling the district and county boundary data according to the boundary scaling factor to obtain a standard district and county boundary;

[0114] The standard county boundaries are used to perform boundary cutting on the total primary productivity remote sensing data to obtain county productivity remote sensing data.

[0115] In one embodiment, when performing a masking operation on the county productivity remote sensing data according to the land use data to obtain the cultivated land productivity data, the mask segmentation module 103 is used to:

[0116] Extracting cultivated land type land data from the land use data;

[0117] Extracting the cultivated land scale from the cultivated land type land data;

[0118] extracting remote sensing scale from the farmland productivity data;

[0119] Calculating the cultivated land scaling factor between the cultivated land scale and the remote sensing scale;

[0120] Scaling the cultivated land boundary data according to the cultivated land scaling factor to obtain scaled cultivated land data;

[0121] generating cultivated land mask data according to the scaled cultivated land data;

[0122] The cultivated land mask data is used to perform a mask operation on the cultivated land productivity data to obtain the cultivated land productivity data.

[0123] In one embodiment, when performing segmentation center screening on the cultivated land productivity data to obtain a segmentation cluster center set, the center screening module 104 is used to:

[0124] Performing area statistics on the cultivated land productivity data to obtain the total cultivated land area;

[0125] Calculate the area of ​​sub-cultivated land according to the total area of ​​the cultivated land and the preset number of divisions;

[0126] Initialize the segmentation grid according to the sub-cultivated land area;

[0127] Gridding the cultivated land productivity data according to the segmentation grids to obtain a segmented cultivated land grid data set;

[0128] The grid midpoints corresponding to the segmented farmland grid data set are gathered into a segmentation cluster center set.

[0129] In one embodiment, when the superpixel segmentation module 105 performs dynamic superpixel segmentation on the farmland productivity data according to the segmentation cluster center set to obtain a regional farmland productivity dataset, it is used to:

[0130] Performing time series data conversion on the cultivated land productivity data to obtain a productivity series set;

[0131] Gathering the productivity sequences corresponding to the segmentation cluster center set in the productivity sequence set into a center productivity sequence set;

[0132] Selecting productivity sequences in the productivity sequence set one by one as target productivity sequences, and selecting central productivity sequences in the central productivity sequence set one by one as target central sequences;

[0133] Calculating the productivity distance between the target productivity sequence and the target center sequence;

[0134] According to all productivity distances, the productivity sequence set is updated into a primary productivity sequence group set in accordance with the method of nearest allocation;

[0135] Iteratively updating the primary productivity sequence group set to obtain a productivity sequence class set;

[0136] Performing regional merging on each productivity sequence class in the productivity sequence class set to obtain a super-pixel cultivated land productivity dataset;

[0137] A clustering operation is performed on the super-pixel cultivated land productivity dataset to obtain a regional cultivated land productivity dataset.

[0138] In one embodiment, when calculating the productivity distance between the target productivity sequence and the target center sequence, the superpixel segmentation module 105 is used to:

[0139] Constructing a cost matrix for the target productivity sequence and the target center sequence to obtain a productivity cost matrix;

[0140] Performing dynamic programming calculation on the productivity cost matrix to obtain a cumulative cost matrix;

[0141] Performing path screening on the cumulative cost matrix to obtain an alignment path;

[0142] Generate productivity distances according to the alignment paths.

[0143] In one embodiment, when the classification module 106 performs classification on the regional cultivated land productivity dataset to obtain the cultivated land productivity dataset, it is used to:

[0144] Calculating the productivity mean of each regional cultivated land productivity data in the regional cultivated land productivity data set to obtain a mean productivity set;

[0145] Sorting the mean productivity set in descending order to obtain a mean productivity sequence;

[0146] Classifying the mean productivity sequence according to a preset sequence number range to obtain a productivity grade sequence;

[0147] The productivity grade sequence is used to grade the regional cultivated land productivity data to obtain a cultivated land productivity data set.

[0148] The present invention provides a cultivated land data analysis device based on remote sensing data. The device first acquires the total primary productivity remote sensing data, land use data and district and county boundary data respectively, and can abstract the land cultivated land productivity grade division into a digital image processing scheme, thereby improving the division efficiency. The district and county boundary data are used to perform boundary cutting on the total primary productivity remote sensing data to obtain the district and county productivity remote sensing data, and the productivity remote sensing data of the district and county that need to be divided into cultivated land productivity can be screened out, thereby realizing accurate cultivated land productivity division. The cultivated land productivity data is obtained by performing a mask operation on the district and county productivity remote sensing data according to the land use data, and the primary productivity data of the cultivated land area in the target district and county area can be removed, thereby facilitating the subsequent cultivated land productivity division operation.

[0149] By screening the segmentation centers of the cultivated land productivity data to obtain a segmentation cluster center set, the initial cluster center can be screened out, so as to facilitate the subsequent uniform segmentation of the cultivated land productivity data by clustering method, thereby improving the accuracy of soil fertility grade division; by dynamically superpixel segmenting the cultivated land productivity data according to the segmentation cluster center set, a regional cultivated land productivity data set is obtained, and the cultivated land productivity data can be divided according to the sound field cycle of the crop and the organic matter data accumulated in different periods, thereby improving the interpretability and accuracy of the soil fertility grade division; by performing grade division on the regional cultivated land productivity data set, an cultivated land soil fertility data set is obtained, and detailed soil fertility area division and soil fertility grade division can be realized in combination with the productivity data of the crop growth cycle, thereby improving the accuracy of soil fertility grade division.

[0150] For the specific definition of the cultivated land data analysis device based on remote sensing data, please refer to the definition of the intelligent question and answer method above, which will not be repeated here. Each module in the above-mentioned cultivated land data analysis device based on remote sensing data can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.

[0151] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 7 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile and / or volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external client through a network connection. When the computer program is executed by the processor, it realizes the functions or steps of a server-side method for analyzing cultivated land data based on remote sensing data.

[0152] In one embodiment, a computer device is provided. The computer device may be a client, and its internal structure diagram may be as follows: Figure 8As shown. The computer device includes a processor, a memory, a network interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external server through a network connection. When the computer program is executed by the processor, the functions or steps on the client side of a cultivated land data analysis method based on remote sensing data are implemented.

[0153] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following steps when executing the computer program:

[0154] Obtain remote sensing data on gross primary productivity, land use data, and district and county boundary data respectively;

[0155] Using the district and county boundary data to perform boundary cutting on the total primary productivity remote sensing data, to obtain district and county productivity remote sensing data;

[0156] Performing a mask operation on the county productivity remote sensing data according to the land use data to obtain cultivated land productivity data;

[0157] Screening the segmentation centers of the cultivated land productivity data to obtain a segmentation cluster center set;

[0158] Performing dynamic superpixel segmentation on the cultivated land productivity data according to the segmentation cluster center set to obtain a regional cultivated land productivity data set;

[0159] The regional cultivated land productivity dataset is graded to obtain a cultivated land fertility dataset.

[0160] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0161] Obtain remote sensing data on gross primary productivity, land use data, and district and county boundary data respectively;

[0162] Using the district and county boundary data to perform boundary cutting on the total primary productivity remote sensing data, to obtain district and county productivity remote sensing data;

[0163] Performing a mask operation on the county productivity remote sensing data according to the land use data to obtain cultivated land productivity data;

[0164] Screening the segmentation centers of the cultivated land productivity data to obtain a segmentation cluster center set;

[0165] Performing dynamic superpixel segmentation on the cultivated land productivity data according to the segmentation cluster center set to obtain a regional cultivated land productivity data set;

[0166] The regional cultivated land productivity dataset is graded to obtain a cultivated land fertility dataset.

[0167] It should be noted that the above functions or steps that can be implemented by the computer-readable storage medium or computer device can refer to the relevant descriptions on the server side and the client side in the aforementioned method embodiment. To avoid repetition, they will not be described one by one here.

[0168] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0169] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.

[0170] The above-described embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some of the technical features therein by equivalents; and these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention, and should be included in the protection scope of the present invention. It should be noted that if software tools or components other than those of the company appear in the embodiments of this application, they are only used for example introduction and do not represent actual use.

Claims

1. A method for analyzing cultivated land data based on remote sensing data, characterized in that: include: Obtain remote sensing data on gross primary productivity, land use data, and district and county boundary data respectively; Using the district and county boundary data to perform boundary cutting on the total primary productivity remote sensing data, to obtain district and county productivity remote sensing data; Performing a mask operation on the county productivity remote sensing data according to the land use data to obtain cultivated land productivity data; Screening the segmentation centers of the cultivated land productivity data to obtain a segmentation cluster center set; Performing dynamic superpixel segmentation on the cultivated land productivity data according to the segmentation cluster center set to obtain a regional cultivated land productivity data set; The regional cultivated land productivity dataset is graded to obtain a cultivated land fertility dataset.

2. The method for analyzing cultivated land data based on remote sensing data according to claim 1, characterized in that: The method of using the district and county boundary data to perform boundary cutting on the total primary productivity remote sensing data to obtain district and county productivity remote sensing data includes: Extracting a boundary scale from the district and county boundary data; extracting a remote sensing scale from the total primary productivity remote sensing data; Calculating a boundary scaling factor between the boundary scale and the remote sensing scale; Scaling the district and county boundary data according to the boundary scaling factor to obtain a standard district and county boundary; The standard county boundaries are used to perform boundary cutting on the total primary productivity remote sensing data to obtain county productivity remote sensing data.

3. The method for analyzing cultivated land data based on remote sensing data according to claim 1, characterized in that: The masking operation is performed on the county productivity remote sensing data according to the land use data to obtain the cultivated land productivity data, including: Extracting cultivated land type land data from the land use data; Extracting the cultivated land scale from the cultivated land type land data; extracting a remote sensing scale from the farmland productivity data; Calculating the cultivated land scaling factor between the cultivated land scale and the remote sensing scale; Scaling the cultivated land boundary data according to the cultivated land scaling factor to obtain scaled cultivated land data; generating cultivated land mask data according to the scaled cultivated land data; The cultivated land mask data is used to perform a mask operation on the cultivated land productivity data to obtain the cultivated land productivity data.

4. The method for analyzing cultivated land data based on remote sensing data according to claim 1, characterized in that: The segmentation center screening of the cultivated land productivity data to obtain a segmentation cluster center set includes: Performing area statistics on the cultivated land productivity data to obtain the total cultivated land area; Calculate the area of ​​sub-cultivated land according to the total area of ​​the cultivated land and the preset number of divisions; Initialize the segmentation grid according to the sub-cultivated land area; Gridding the cultivated land productivity data according to the segmentation grids to obtain a segmented cultivated land grid data set; The grid midpoints corresponding to the segmented farmland grid data set are gathered into a segmentation cluster center set.

5. The method for analyzing cultivated land data based on remote sensing data according to claim 1, characterized in that: The method of performing dynamic superpixel segmentation on the cultivated land productivity data according to the segmentation cluster center set to obtain a regional cultivated land productivity data set includes: Performing time series data conversion on the cultivated land productivity data to obtain a productivity series set; Gathering the productivity sequences corresponding to the segmentation cluster center set in the productivity sequence set into a center productivity sequence set; Selecting productivity sequences in the productivity sequence set one by one as target productivity sequences, and selecting central productivity sequences in the central productivity sequence set one by one as target central sequences; Calculating the productivity distance between the target productivity sequence and the target center sequence; According to all productivity distances, the productivity sequence set is updated into a primary productivity sequence group set in accordance with the method of nearest allocation; Iteratively updating the primary productivity sequence group set to obtain a productivity sequence class set; Performing regional merging on each productivity sequence class in the productivity sequence class set to obtain a super-pixel cultivated land productivity dataset; A clustering operation is performed on the super-pixel cultivated land productivity dataset to obtain a regional cultivated land productivity dataset.

6. The method for analyzing cultivated land data based on remote sensing data according to claim 5, characterized in that: The calculating the productivity distance between the target productivity sequence and the target center sequence comprises: Constructing a cost matrix for the target productivity sequence and the target center sequence to obtain a productivity cost matrix; Performing dynamic programming calculation on the productivity cost matrix to obtain a cumulative cost matrix; Performing path screening on the cumulative cost matrix to obtain an alignment path; Generate productivity distances according to the alignment paths.

7. The method for analyzing cultivated land data based on remote sensing data according to claim 1, characterized in that: The step of classifying the regional cultivated land productivity dataset to obtain the cultivated land productivity dataset includes: Calculating the productivity mean of each regional cultivated land productivity data in the regional cultivated land productivity data set to obtain a mean productivity set; Sorting the mean productivity set in descending order to obtain a mean productivity sequence; Classifying the mean productivity sequence according to a preset sequence number range to obtain a productivity grade sequence; The productivity grade sequence is used to grade the regional cultivated land productivity data to obtain a cultivated land productivity data set.

8. A farmland data analysis device based on remote sensing data, characterized in that: include: The data acquisition module is used to obtain remote sensing data of total primary productivity, land use data and district and county boundary data respectively; A boundary segmentation module is used to perform boundary segmentation on the total primary productivity remote sensing data using the district and county boundary data to obtain district and county productivity remote sensing data; A mask segmentation module, used for performing a mask operation on the county productivity remote sensing data according to the land use data to obtain cultivated land productivity data; A center screening module, used for screening the segmentation centers of the cultivated land productivity data to obtain a segmentation cluster center set; A superpixel segmentation module, used for performing dynamic superpixel segmentation on the cultivated land productivity data according to the segmentation cluster center set to obtain a regional cultivated land productivity data set; The grading module is used to grade the regional cultivated land productivity dataset to obtain the cultivated land productivity dataset.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method for analyzing cultivated land data based on remote sensing data as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method for analyzing cultivated land data based on remote sensing data as described in any one of claims 1 to 7 is implemented.

Images