A method and system for reconstructing ecosystem biomass based on sky-ground big data

Abstract

The present invention discloses a method and system for reconstructing ecosystem biomass based on sky-ground big data. The control method selects grass survey data, UAV remote sensing data and satellite remote sensing vegetation index as research objects, and uses regression-assimilation coupling analysis method and machine learning to establish a biomass reconstruction mathematical model. Classification modeling is performed considering conditions such as different ecosystem types, regions and vegetation growth periods. UAV and satellite remote sensing vegetation indexes are selected, and the biomass of different ecosystems is estimated through the established mathematical regression model of vegetation index and biomass. Ground-observed vegetation index and biomass data are used for dynamic assimilation to optimize the model to improve the estimation accuracy and prediction accuracy of the model. This method reduces the unpredictable noise caused by factors such as observation angle and sensor performance. The spatiotemporal resolution and estimation prediction accuracy of ecological biomass data are improved, thereby more scientifically guiding the study of the evolution laws of different ecosystems.

Description

Technical Field

[0001] The present invention relates to the field of ecological remote sensing technology, and in particular to a method and system for reconstructing ecosystem biomass based on sky-ground big data. Background Art

[0002] Ecosystem biomass is an important indicator of animal husbandry production, and is closely related to agricultural and animal husbandry production, soil erosion, land desertification, and degradation of ecosystem service functions. Therefore, obtaining information on ecosystem biomass and its changes at a higher temporal and spatial resolution is of great practical significance for revealing the laws of animal husbandry production space and animal husbandry output changes, exploring the driving factors of changes, and analyzing and evaluating the regional ecological environment. Summary of the invention

[0003] In view of the above-mentioned deficiencies in the prior art, the present invention provides a method and system for reconstructing ecosystem biomass based on sky-ground big data.

[0004] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0005] A method for reconstructing ecosystem biomass based on sky-ground big data comprises the following steps:

[0006] S1. Obtain satellite images, UAV remote sensing images and ground observation data of vegetation coverage in the study area, and calculate the corresponding satellite image vegetation index products and UAV image vegetation index products;

[0007] S2. Temporal and spatial arrangement of satellite image vegetation index products, drone image vegetation index products and ground observation data;

[0008] S3. Calculate the mean of the satellite image vegetation index product and the drone image vegetation index product and construct a vegetation index data set, and use the ground observation data to verify the data results;

[0009] S4, reconstructing the vegetation index data set constructed in step S3;

[0010] S5, the ecosystem biomass and the vegetation index data set after data reconstruction and lag in step S4 are used to build a regression model to analyze the changing trends of vegetation coverage and aboveground biomass and the proportion of each changing trend in the total.

[0011] The beneficial effect of the above scheme is that, based on field survey data and raster data sets, the spatiotemporal resolution and estimation accuracy of ecosystem biomass data are improved, providing more accurate data support for refined regional ecological environmental changes and its regional climate model.

[0012] Furthermore, the step S2 specifically includes:

[0013] If satellite image vegetation index products, drone image vegetation index products, and ground observation data are stored in the selected scene at the same time, the resolution of the drone image vegetation index product will be resampled to the resolution of the satellite image vegetation index product.

[0014] The beneficial effect of the above further solution is that multiple data sources are integrated to achieve higher calculation accuracy.

[0015] Furthermore,

[0016] The method for verifying the data results using ground observation data in step S3 is:

[0017] S31. Calculate the difference between the mean of vegetation index products of satellite images and drone images and the vegetation index products of ground observation data;

[0018] S32, judging the range of the difference calculated in step S31, if the value is between ±0.2, it is considered that the data set is correct, otherwise, the data set is not used, and the specific calculation formula is as follows:

[0019] 0.2≤|AB|

[0020] Among them, A is the average of the vegetation index product of satellite images and the vegetation index product of drone images, and B is the vegetation index product of ground observation data.

[0021] The beneficial effect of the above further solution is that multiple data sources are integrated to achieve higher calculation accuracy.

[0022] Furthermore, the step S4 specifically includes:

[0023] S41, performing interval extraction on the vegetation index data set constructed in step S3, finding the maximum and minimum values ​​of the image NDVI time series curve, dividing the NDVI time series into multiple intervals using the obtained maximum and minimum values, and obtaining local fitting intervals of multiple time dimensions;

[0024] S42, using a Gaussian fitting function to perform local fitting on the local fitting intervals of the multiple time dimensions extracted in step S41;

[0025] S43, reconstructing and connecting the local fitting intervals of multiple time dimensions that have been locally fitted in step S42 as a whole to obtain a Gaussian fitting curve of the time series.

[0026] The beneficial effect of the above further scheme is to fuse multi-source remote sensing data from air, space and ground to calculate vegetation index products with higher accuracy.

[0027] Furthermore, in step S42, an interval model function is used for local fitting, and the interval model function is expressed as:

[0028] f(t)=f(t;c,d)=c1+c2g(t;d)

[0029] Where c is the linear coefficient, (c1,c2)∈c, c1 is the base value of the fitting curve, c2 is the amplitude of the fitting curve, d is the nonlinear coefficient and d=(d1,d2...,d i ), g(t; d) is the shape of the basis function, expressed as:

[0030]

[0031] Among them, d1 is the location parameter of the variable t corresponding to the maximum or minimum value, and d2, d3, d4 and d5 are the width and steepness of the left and right half curves of the extreme value respectively.

[0032] Furthermore, in step S43, the local fitting intervals of multiple time dimensions are reconstructed and connected as a whole using a reconstruction connection function, and the reconstruction connection function is expressed as:

[0033]

[0034] Among them, t L to R The interval between represents the sliding interval to be fitted, f L (t), f R (t) and f C (t) are the local fitting functions of the minimum values ​​on the left and right sides and the maximum value in the middle of the sliding interval to be fitted, a(t) and b(t) are shear coefficients, with values ​​ranging from 0 to 1.

[0035] Furthermore, the regression model in step S5 is a linear regression model, which is expressed as:

[0036] y=wx+D

[0037] Among them, x is the vegetation coverage parameter, y is the biomass parameter, w is the straight line intercept, and D is the straight line slope.

[0038] The beneficial effect of the above further scheme is to fuse multi-source remote sensing data from the air, space and ground to calculate vegetation index products with higher accuracy, and then calculate the final biomass estimation method.

[0039] A system for reconstructing ecosystem biomass based on sky-ground big data is also proposed, including:

[0040] It includes a data acquisition module, a data processing module and a data output module;

[0041] The data acquisition module is used to obtain satellite images, UAV remote sensing images and ground observation data of vegetation coverage in the study area, and calculate the corresponding satellite image vegetation index products and UAV image vegetation index products;

[0042] The data processing module is used to perform temporal and spatial sorting of satellite image vegetation index products, drone image vegetation index products and ground observation data, calculate the mean of satellite image vegetation index products and drone image vegetation index products and construct a vegetation index data set, and use the ground observation data to verify the data results, reconstruct the vegetation index data set, and construct a regression model for the ecosystem biomass and the vegetation index data set after data reconstruction, and analyze the change trends of vegetation coverage and aboveground biomass and the proportion of each change trend in the total;

[0043] The data output module is used to output the calculated biomass data. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Figure 1 The figure is a flow chart of the ecosystem biomass reconstruction method based on sky-ground big data of the present invention.

[0045] Figure 2 It is a schematic diagram of the structure of the ecosystem biomass reconstruction system based on sky-ground big data of the present invention. DETAILED DESCRIPTION

[0046] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0047] A method for reconstructing ecosystem biomass based on sky-ground big data, such as Figure 1 As shown, the following steps are included:

[0048] S1. Obtain satellite images, UAV remote sensing images and ground observation data of vegetation coverage in the study area, and calculate the corresponding satellite image vegetation index products and UAV image vegetation index products;

[0049] Specifically, in this embodiment, the vegetation coverage product in the global land surface characteristic parameter GLASS of the study area (the spatial position in this paper is 37°24′~53°23′N, 97°12′-126°04′E) is obtained to obtain the satellite estimated NDVI product; and at the same time, the UAV remote sensing image in the study area is obtained and the corresponding NDVI product is calculated.

[0050] S2. Temporal and spatial arrangement of satellite image vegetation index products, drone image vegetation index products and ground observation data;

[0051] Specifically, the satellite NDVI products, UAV NDVI products and ground observation data are sorted according to time and space. For scenarios where the three types of data exist at the same time and space, the resolution of the UAV NDVI product is resampled to the resolution of the satellite NDVI product. The mean of the satellite NDVI product and the UAV NDVI is calculated as the constructed data set, and the results are verified with ground observation data. If the error is less than 0.2, the constructed day-scale data set is considered correct. Subsequently, the maximum value synthesis method MVC is used to construct data sets of different time scales from day to week to month to year for the day-scale data set.

[0052] Among them, if there are satellite image vegetation index products, drone image vegetation index products and ground observation data in the selected scene, the resolution of the drone image vegetation index product will be resampled to the resolution of the satellite image vegetation index product.

[0053] S3. Calculate the mean of the satellite image vegetation index product and the drone image vegetation index product and construct a vegetation index data set, and use the ground observation data to verify the data results;

[0054] This embodiment uses a data reconstruction method and combines the asymmetric Gaussian function fitting method (AG) to reconstruct the NDVI data. The asymmetric Gaussian function fitting method uses a combination of segmented Gaussian functions (curves) to simulate the seasonal growth (phenology) law of vegetation. One combination represents a vegetation prosperity and decline process. Finally, the time series reconstruction is achieved by smoothly connecting each Gaussian fitting curve. The specific process includes: interval extraction, that is, selecting a maximum or minimum interval in the time dimension as a local fitting interval, local fitting, that is, using a Gaussian fitting function to fit the local interval data, and overall connection, that is, merging the local fitting results.

[0055] Specifically, the mean of the vegetation index products of satellite images and drone images is calculated, and the average value calculation formula is:

[0056] X=(x1+x2) / 2

[0057] Among them, for satellite images and drone images with the same resolution, x1 is the vegetation index of any pixel in the satellite image, x2 is the vegetation index of the drone image of the corresponding pixel, and X is the mean of the vegetation index of the satellite image and the vegetation index of the drone image of the corresponding pixel.

[0058] The method of using ground observation data to verify data results is:

[0059] S31. Calculate the difference between the mean of vegetation index products of satellite images and drone images and the vegetation index products of ground observation data;

[0060] S32, judging the range of the difference calculated in step S31, if the value is between ±0.2, it is considered that the data set is correct, otherwise, the data set is not used, and the specific calculation formula is as follows:

[0061] 0.2≤|AB|

[0062] Among them, A is the average of the vegetation index product of satellite images and the vegetation index product of drone images, and B is the vegetation index product of ground observation data.

[0063] S4, reconstructing the vegetation index data set constructed in step S3;

[0064] In this embodiment, a method for estimating ecosystem biomass with high spatiotemporal resolution is constructed, a regression analysis model is constructed based on NDVI parameters and ecosystem biomass, and the vegetation coverage and aboveground biomass change trends and the proportion of each change trend in the total are analyzed at the pixel scale, and finally a method for spatiotemporal reconstruction of ecosystem biomass data based on sky-ground big data is realized. Specifically,

[0065] S41, performing interval extraction on the vegetation index data set constructed in step S3, finding the maximum and minimum values ​​of the image NDVI time series curve, dividing the NDVI time series into multiple intervals using the obtained maximum and minimum values, and obtaining local fitting intervals of multiple time dimensions.

[0066] S42, using a Gaussian fitting function to perform local fitting on the local fitting intervals of the multiple time dimensions extracted in step S41;

[0067] In this embodiment, an interval model function is used for local fitting, and the interval model function is expressed as:

[0068] f(t)=f(t;c,d)=c1+c2g(t;d)

[0069] Where c is the linear coefficient, (c1,c2)∈c, c1 is the base value of the fitting curve, c2 is the amplitude of the fitting curve, d is the nonlinear coefficient and d=(d1,d2...,d i ), g(t; d) is the shape of the basis function, expressed as:

[0070]

[0071] Among them, d1 is the location parameter of the variable t corresponding to the maximum or minimum value, and d2, d3, d4 and d5 are the width and steepness of the left and right half curves of the extreme value respectively.

[0072] S43, reconstructing and connecting the local fitting intervals of multiple time dimensions that have been locally fitted in step S42 as a whole to obtain a Gaussian fitting curve of the time series.

[0073] In this embodiment, the reconstruction connection function is used to reconstruct the connection of the local fitting intervals of multiple time dimensions as a whole. The reconstruction connection function is expressed as:

[0074]

[0075] Among them, t L to R The interval between represents the sliding interval to be fitted, f L (t), f R (t) and f C (t) are the local fitting functions of the minimum values ​​on the left and right sides and the maximum value in the middle of the sliding interval to be fitted. a(t) and b(t) are shear coefficients, with values ​​ranging from 0 to 1.

[0076] S5, the ecosystem biomass and the vegetation index data set after data reconstruction and lag in step S4 are used to build a regression model to analyze the changing trends of vegetation coverage and aboveground biomass and the proportion of each changing trend in the total.

[0077] The regression model is a linear regression model, expressed as:.

[0078] y=wx+D

[0079] Among them, x is the vegetation coverage parameter, y is the biomass parameter, w is the straight line intercept, and D is the straight line slope.

[0080] This proposal also proposes an ecosystem biomass reconstruction system based on sky-ground big data, such as Figure 2 As shown, it includes a data acquisition module, a data processing module and a data output module;

[0081] The data acquisition module is used to obtain satellite images, UAV remote sensing images and ground observation data of vegetation coverage in the study area, and calculate the corresponding satellite image vegetation index products and UAV image vegetation index products;

[0082] The data processing module is used to perform temporal and spatial arrangement of satellite image vegetation index products, drone image vegetation index products and ground observation data, calculate the mean of satellite image vegetation index products and drone image vegetation index products and construct a vegetation index data set, use the ground observation data to verify the data results, reconstruct the vegetation index data set, and construct a regression model for the ecosystem biomass and the vegetation index data set after data reconstruction, analyze the vegetation coverage and aboveground biomass change trends and the proportion of each change trend in the total; the data output module is used to output the calculated biomass data.

[0083] The system is built using the method constructed in this paper. It only requires inputting satellite images, UAV images, and ground-observed NDVI values, and the biomass estimation can be completed using the regression analysis model fitted by the method in this paper.

[0084] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0085] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0086] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0087] The present invention uses specific embodiments to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

[0088] Those skilled in the art will appreciate that the embodiments described herein are intended to help readers understand the principles of the present invention, and should be understood that the protection scope of the present invention is not limited to such specific statements and embodiments. Those skilled in the art can make various other specific variations and combinations that do not deviate from the essence of the present invention based on the technical revelations disclosed by the present invention, and these variations and combinations are still within the protection scope of the present invention.

Claims

1. A method for reconstructing ecosystem biomass based on sky-ground big data, characterized in that: The steps include: S1. Obtain satellite images, UAV remote sensing images and ground observation data of vegetation coverage in the study area, and calculate the corresponding satellite image vegetation index products and UAV image vegetation index products; S2. Temporal and spatial arrangement of satellite image vegetation index products, drone image vegetation index products and ground observation data; S3. Calculate the mean of the satellite image vegetation index product and the drone image vegetation index product and construct a vegetation index data set, and use the ground observation data to verify the data results. The specific method is as follows: S31. Calculate the difference between the mean of vegetation index products of satellite images and drone images and the vegetation index products of ground observation data; S32, determine the range of the difference calculated in step S31, if the value is within If the value is between 0.0 and 0.2, it is considered to be the correct data set. Otherwise, the data set will not be used. The specific calculation formula is as follows: Among them, A is the average of the vegetation index product of satellite images and the vegetation index product of drone images, and B is the vegetation index product of ground observation data; S4, reconstructing the vegetation index data set constructed in step S3, specifically including: S41, performing interval extraction on the vegetation index data set constructed in step S3, finding the maximum and minimum values ​​of the image NDVI time series curve, dividing the NDVI time series into multiple intervals using the obtained maximum and minimum values, and obtaining local fitting intervals of multiple time dimensions; S42, using a Gaussian fitting function to perform local fitting on the multiple time dimension local fitting intervals extracted in step S41 using an interval model function, the interval model function is expressed as: Where c is the linear coefficient, , c1 is the base value of the fitting curve, c2 is the amplitude of the fitting curve, d is the nonlinear coefficient and d=(d1,d 2. ..,d i ), g(t; d) is the shape of the basis function, expressed as: Among them, d1 is the location parameter of the variable t corresponding to the maximum or minimum value, d2, d3, d4 and d5 are the width and steepness of the left and right half curves of the extreme value respectively; S43, the local fitting intervals of multiple time dimensions that are locally fitted in step S42 are reconstructed and connected as a whole to obtain a Gaussian fitting curve of the time series, and the local fitting intervals of multiple time dimensions are reconstructed and connected as a whole using a reconstruction connection function, and the reconstruction connection function is expressed as: Among them, t L to R The interval between represents the sliding interval to be fitted, f L (t), f R (t) and f C (t) are the local fitting functions of the minimum values ​​on the left and right sides and the maximum value in the middle of the sliding interval to be fitted, a(t) and b(t) are shear coefficients, with values ​​ranging from 0 to 1; S5. The ecosystem biomass and the vegetation index data set after data reconstruction in step S4 are used to construct a regression model to analyze the changing trends of vegetation coverage and aboveground biomass and the proportion of each changing trend in the total.

2. The method for reconstructing ecosystem biomass based on sky-ground big data according to claim 1, characterized in that: The step S2 specifically includes: If satellite image vegetation index products, drone image vegetation index products, and ground observation data exist in the selected scene, the resolution of the drone image vegetation index product will be resampled to the resolution of the satellite image vegetation index product.

3. The ecosystem biomass reconstruction method based on sky-ground big data according to claim 2 is characterized in that: In step S3, the mean values ​​of the satellite image vegetation index product and the drone image vegetation index product are calculated, and the mean value calculation formula is: Among them, for satellite images and drone images with the same resolution, x 1 is the vegetation index of any pixel satellite image, x 2 is the vegetation index of the drone image corresponding to the pixel, X It is the mean of the vegetation index of satellite images and the vegetation index of drone images of the corresponding pixels.

4. The method for reconstructing ecosystem biomass based on sky-ground big data according to claim 1, characterized in that: The regression model in step S5 is a linear regression model, which is expressed as: in, is the vegetation coverage parameter, is the biomass parameter, is the straight line intercept, is the slope of the straight line.

5. An ecosystem biomass reconstruction system based on sky-ground big data based on any one of the ecosystem biomass reconstruction methods of claims 1-4, characterized in that: It includes a data acquisition module, a data processing module and a data output module; The data acquisition module is used to obtain satellite images, UAV remote sensing images and ground observation data of vegetation coverage in the study area, and calculate the corresponding satellite image vegetation index products and UAV image vegetation index products; The data processing module is used to perform temporal and spatial sorting of satellite image vegetation index products, drone image vegetation index products and ground observation data, calculate the mean of satellite image vegetation index products and drone image vegetation index products and construct a vegetation index data set, and use the ground observation data to verify the data results, reconstruct the vegetation index data set, and construct a regression model for the ecosystem biomass and the vegetation index data set after data reconstruction, and analyze the change trends of vegetation coverage and aboveground biomass and the proportion of each change trend in the total; The data output module is used to output the calculated biomass data.

Images