Ecological biomass estimation method based on fusion of remote sensing image and ground observation data

Abstract

The invention discloses an ecological biomass estimation method based on fusion of a remote sensing image and ground observation data, which relates to the technical field of ecological biomass estimation, and comprises the following steps: obtaining a remote sensing image of an observation area, and carrying out image preprocessing on the remote sensing image to obtain an enhanced image; dividing different ecological regions; extracting ground observation data; dividing the similar ecological areas into an ecological area group; calculating a fluctuation index of the ecological biomass in the similar ecological region; analyzing the ground observation data in the different ecological area groups, and calculating the density relationship of the ecological biomass in the different ecological area groups; the ecological biomass in other areas to be estimated is estimated; the method is used for solving the problem of large deviation of a biomass estimation result caused by insufficient consideration of factors influencing biomass distribution in an existing ecological biomass estimation technology.

Description

Technical Field

[0001] The present invention relates to the technical field of ecological biomass estimation, and in particular to an ecological biomass estimation method based on the fusion of remote sensing images and ground observation data. Background Art

[0002] Ecological biomass estimation technology refers to a methodological system that quantitatively evaluates the total mass of biological organisms or carbon storage in a specific ecosystem by integrating multi-source data and mathematical models. Its core goal is to accurately quantify the productivity, carbon sequestration capacity and resource distribution of the ecosystem, serving climate change research, ecological protection and resource management.

[0003] Existing ecological biomass estimation technologies usually collect biomass in a small area of ​​the ecological region as samples and directly estimate it. However, the degree of vegetation generation, luxuriance and terrain height in different locations within the ecological region are all different. These factors will affect the organisms' choice of habitat and activity range, and therefore will lead to large deviations in the biomass estimation results. For example, in the patent application with publication number CN112348364A, a "Rapid Estimation Method for Grassland Ecosystem Biomass" is disclosed. This solution is to set up reference samples in the study area and construct an estimation manual, and then directly estimate the biomass of other grassland ecosystems. However, different regions have different terrain heights, and their biomass varies greatly. Direct estimation will lead to large deviations in the estimation results. Existing ecological biomass estimation technologies also have the problem of insufficient consideration of factors affecting biomass distribution, resulting in large deviations in the biomass estimation results. Summary of the Invention

[0004] The present invention aims to solve at least one of the technical problems in the prior art to a certain extent. The invention obtains a remote sensing image of the observation area and performs image preprocessing on the remote sensing image to obtain an enhanced image. Then, the pixel grayscale values ​​in the enhanced image are extracted. The enhanced image is divided into ecological regions based on the pixel grayscale values ​​to divide different ecological regions. Ground surveys are conducted according to the divided ecological regions to extract ground observation data. Similar ecological regions are then divided into a group of ecological zone groups. The ground observation data within the ecological zone groups are analyzed to calculate the fluctuation index of the ecological biomass within similar ecological regions. Then, the ground observation data within different ecological zone groups are analyzed to calculate the density relationship of the ecological biomass within different ecological zone groups. Then, an ecological biomass estimation model is constructed. The fluctuation index and density relationship of the ecological zone groups are deeply learned. Finally, the ecological biomass in other areas to be estimated is estimated by the ecological biomass estimation model to solve the problem that the existing ecological biomass estimation technology still has insufficient consideration of factors affecting biomass distribution, resulting in large deviations in the biomass estimation results.

[0005] To achieve the above objectives, in a first aspect, the present application provides an ecological biomass estimation method based on the fusion of remote sensing images and ground observation data, comprising the following steps:

[0006] Acquire remote sensing images of the observation area and perform image preprocessing on the remote sensing images to obtain enhanced images;

[0007] Extracting pixel grayscale values ​​from the enhanced image, dividing the enhanced image into ecological regions based on the pixel grayscale values, and dividing different ecological regions;

[0008] Conduct ground surveys based on the divided ecological regions and extract ground observation data;

[0009] Grouping similar ecological regions into a set of ecoregional groups;

[0010] Analyze ground observation data within ecological region groups and calculate fluctuation indicators of ecological biomass within similar ecological regions;

[0011] Analyze ground observation data in different ecological groups and calculate the density relationship of ecological biomass in different ecological groups;

[0012] Construct an ecological biomass estimation model, conduct in-depth learning on the fluctuation indicators and density relationships of ecological zone groups, and estimate the ecological biomass in other areas to be estimated.

[0013] Furthermore, the image preprocessing includes standardizing and denoising the remote sensing image, and then grayscale processing. After completing the image preprocessing, an enhanced image can be obtained.

[0014] Furthermore, the pixel grayscale values ​​in the enhanced image are extracted, and the enhanced image is divided into ecological regions based on the pixel grayscale values. The division of different ecological regions includes the following sub-steps:

[0015] The pixel at row n and column m in the enhanced image is marked as P(n,m), and the grayscale value of the pixel at P(n,m) is marked as C(n,m);

[0016] Get the vegetation index corresponding to C(n,m), marked as V(n,m);

[0017] Randomly select a P(n,m) in the enhanced image as the starting point of the region, and mark the starting point V(n,m) of the region as the region index;

[0018] Query the V(n,m) of all P(n,m) in the eight neighborhoods of the regional starting point, mark the P(n,m) with the same V(n,m) and regional index as regional points, then query the V(n,m) of all P(n,m) in the eight neighborhoods of the regional point except the regional point and the regional starting point, mark the P(n,m) with the same V(n,m) and regional index as regional points, and execute the loop until there are no more consecutive adjacent P(n,m) with the same V(n,m) and regional index. The continuous regional points form an ecological region.

[0019] Again, any P(n,m) in the enhanced image except the ecological region is used as the starting point of the region for analysis until all pixels in the enhanced image are included in the ecological region, and different ecological regions are obtained.

[0020] Furthermore, ground survey is conducted based on the divided ecological regions, and extracting ground observation data includes the following sub-steps:

[0021] When a ground survey is conducted for any ecological region, it is named as the survey area;

[0022] Preset the survey length, find the highest point and the lowest point in the survey area, and demarcate a square area with a side length of the survey length at the highest point and the lowest point respectively, naming them the highest point survey area and the lowest point survey area;

[0023] The ecological biomass in the highest point survey area and the lowest point survey area were surveyed on the ground and named as the highest point biomass and the lowest point biomass, respectively, and represented by the symbols Hb and Mb;

[0024] Calculate the square of the survey length, named the survey area, represented by the symbol Sa, and calculate Hb / Sa and Mb / Sa to obtain the highest point biomass density and the lowest point biomass density respectively;

[0025] The average value of the terrain height in the survey area is analyzed based on the terrain height in the remote sensing data and is named as the average terrain height;

[0026] A plane rectangular coordinate system is established with the terrain height as the X-axis and the biomass density as the Y-axis, and is named the terrain biomass estimation map. The highest point biomass of the highest point and the lowest point biomass of the lowest point are recorded in the terrain biomass estimation map, and the coordinate points in the terrain biomass estimation map are named terrain biomass points.

[0027] Solve the equation of the line passing through the two points of the geobiotaxis, which is named the geobiotaxis estimation equation;

[0028] Substitute the average terrain into the terrain biological estimation equation to obtain the average biological density of the survey area, and calculate the average biological density of each ecological region. The highest point biological density, the lowest point biological density and the average biological density are the ground observation data.

[0029] Furthermore, dividing similar ecological regions into a group of ecological regions specifically involves dividing ecological regions with the same regional index into the same group of ecological regions.

[0030] Furthermore, the ground observation data within the ecological zone group are analyzed to calculate the fluctuation index of ecological biomass within similar ecological regions, which includes the following sub-steps:

[0031] The ecological regions within the ecological zone group are named as intra-group regions, and the difference between the highest and lowest points in the intra-group region is calculated and named as the relief drop;

[0032] Find the area within the group with the largest relief drop, name it the benchmark area, obtain the relief biomass estimation map of the benchmark area, name it the benchmark estimation map, and name the corresponding relief biomass estimation equation the benchmark equation;

[0033] Enter the highest point biomass density, lowest point biomass density, and average biomass density of all ecoregions within the ecoregion group into the baseline estimation map according to the corresponding terrain height;

[0034] The coordinate points in the benchmark estimation map are named benchmark coordinate points. When analyzing any benchmark coordinate point, the elevation of the benchmark coordinate point is substituted into the benchmark equation to obtain the benchmark density. The difference between the biological density corresponding to the benchmark coordinate point and the benchmark density is then calculated and named the density fluctuation value.

[0035] Analyze all reference coordinate points to obtain different density fluctuation values, find the maximum value of the density fluctuation value, and mark it as the fluctuation index.

[0036] Furthermore, the ground observation data in different ecological groups were analyzed to calculate the density relationship of ecological biomass in different ecological groups;

[0037] The ecological zones are numbered in descending order according to the vegetation index, and the symbol T i Represents, where i is a positive integer and i is the serial number of T;

[0038] Get T i The reference equation, labeled E i , get T i The height of the terrain corresponding to the lowest biomass density is marked as TH i ;

[0039] Calculate and find TH iThe minimum value in is named the lowest terrain, and the lowest terrain is substituted into each E i , find each T i The density of organisms at the lowest elevation, marked as R i ;

[0040] Calculate R i The ratio of the biomass density at the lowest point to the lowest elevation, marked as W i ;

[0041] With i as the horizontal axis, W i Establish a plane rectangular coordinate system for the vertical axis, named the lowest density relationship diagram, and set W i Enter the lowest density relationship diagram according to i.

[0042] Furthermore, the ecological biomass estimation model is constructed, and the relationship between fluctuation indicators and density of ecological zones is deeply studied. The ecological biomass in other areas to be estimated is estimated, which includes the following sub-steps:

[0043] Construct an ecological biomass estimation model and conduct deep learning on the relationship between fluctuation indicators and density of ecological groups;

[0044] The ecological biomass in other areas to be estimated is estimated using the ecological biomass estimation model.

[0045] Furthermore, the construction of an ecological biomass estimation model and the deep learning of the relationship between fluctuation indicators and density of ecological groups include the following sub-steps:

[0046] Constructing ecological biomass estimation models;

[0047] Analyzing the first number of observation areas and obtaining fluctuation indices and density relationships;

[0048] Calculate the average value of all volatility indicators and name it density volatility parameter;

[0049] The minimum density relationship maps of different observation areas are merged into the same map to obtain a comprehensive minimum density relationship map;

[0050] A discrete regression analysis was performed on the lowest density relationship comprehensive graph, and the discrete regression equation with the lowest standard deviation was taken as the lowest density relationship comprehensive equation.

[0051] Furthermore, estimating the ecological biomass in other areas to be estimated using the ecological biomass estimation model includes the following sub-steps:

[0052] The area where ecological biomass estimation is required is marked as the area to be estimated, remote sensing images of the area to be estimated are obtained, and vegetation indicators at the lowest terrain are extracted and named as indicators to be analyzed. At the same time, the lowest terrain is named as terrain to be analyzed.

[0053] Conduct ground survey at the terrain to be analyzed to obtain the biological density at the terrain to be analyzed, which is named as the density to be analyzed;

[0054] Find the value of i corresponding to the indicator to be analyzed in the lowest density relationship comprehensive graph and name it as the value to be analyzed;

[0055] Name the coordinate point with the horizontal axis equal to the value to be analyzed and the vertical axis equal to the density to be analyzed as the coordinate point to be analyzed and enter it into the minimum density relationship comprehensive graph. Move the line corresponding to the minimum density relationship comprehensive equation vertically to the coordinate point to be analyzed to obtain a new equation, which is named the minimum density relationship target equation.

[0056] When analyzing any ecological zone group within the area to be estimated, it is named as the zone group to be analyzed, and the benchmark estimation map and benchmark estimation equation of the ecological zone group with the same regional indicators as the zone group to be analyzed are obtained and named as the benchmark map to be analyzed and the benchmark equation to be analyzed;

[0057] Substitute the i corresponding to the area group to be analyzed into the minimum density relationship target equation, multiply the calculated result by the density to be analyzed to obtain the target minimum density, and at the same time obtain the lowest terrain corresponding to the minimum density relationship target equation, which is named the target minimum terrain;

[0058] Move the straight line corresponding to the benchmark equation to be analyzed vertically to the coordinate point (target lowest terrain, target lowest density) to obtain a new equation, which is named the target benchmark equation;

[0059] Obtain the average value of the terrain height of the block to be analyzed, named as the average height, substitute the average height into the target benchmark equation, solve to obtain the block group average density of the block to be analyzed, increase and decrease the fluctuation index of the block group average density respectively, and obtain the minimum density and maximum density of the block group respectively;

[0060] Calculate the area of ​​the block to be analyzed, named block area, and multiply the block area by the minimum density and maximum density of the block to obtain the minimum biomass and maximum biomass of the block to be analyzed;

[0061] Calculate the minimum and maximum biomass of each ecological zone group in the area to be estimated, and calculate the sum of the minimum biomass to obtain the minimum estimated biomass of the area to be estimated, and calculate the sum of the maximum biomass to obtain the maximum estimated biomass of the area to be estimated.

[0062] The beneficial effects of the present invention are as follows: the present invention obtains a remote sensing image of an observation area, performs image preprocessing on the remote sensing image, obtains an enhanced image, extracts pixel grayscale values ​​in the enhanced image, divides the enhanced image into ecological regions based on the pixel grayscale values, divides different ecological regions, conducts ground surveys according to the divided ecological regions, extracts ground observation data, divides similar ecological regions into a group of ecological region groups, analyzes the ground observation data within the ecological region groups, and calculates the fluctuation index of ecological biomass within similar ecological regions. The advantage is that the calculation of ecological biomass is an estimation, which is an approximate value, and the biomass itself has certain fluctuations. By calculating the fluctuation index, the fluctuation index can be included in the reference range when the biomass is finally estimated, thereby estimating the fluctuation range of the biomass, so as to estimate the biomass more accurately and improve the accuracy and effectiveness of the ecological biomass estimation;

[0063] The present invention analyzes ground observation data in different ecological zone groups, calculates the density relationship of ecological biomass in different ecological zone groups, then constructs an ecological biomass estimation model, conducts deep learning on the fluctuation index and density relationship of the ecological zone group, and finally estimates the ecological biomass in other areas to be estimated through the ecological biomass estimation model. The advantage is that the density relationship fully considers the relationship between biomass in areas with different vegetation indices and terrain heights, thereby estimating the biomass of other ecological areas based on vegetation index and terrain height, and adding fluctuation index makes the estimation result more accurate, thereby improving the accuracy and comprehensiveness of ecological biomass estimation. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 is a flow chart of the steps of the method of the present invention;

[0065] Figure 2 is the enhanced image of the present invention;

[0066] Figure 3 is a schematic diagram of the ecological region of the present invention;

[0067] Figure 4 is a topographic biomass estimation map of the present invention;

[0068] Figure 5 is the benchmark estimation diagram of the present invention;

[0069] Figure 6 This is the minimum density relationship diagram of the present invention. DETAILED DESCRIPTION

[0070] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only 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 making creative efforts are within the scope of protection of the present invention.

[0071] Example 1, please refer to Figure 1 As shown, this application provides an ecological biomass estimation method based on the fusion of remote sensing images and ground observation data, including the following steps:

[0072] See also Figure 2 As shown, in step S1, a remote sensing image of the observation area is obtained, and image preprocessing is performed on the remote sensing image to obtain an enhanced image; the image preprocessing includes standardizing and denoising the remote sensing image, and then performing grayscale processing. After completing the image preprocessing, an enhanced image can be obtained;

[0073] In the specific implementation, the enhanced image is processed as follows Figure 2 shown.

[0074] Step S2, extracting pixel grayscale values ​​from the enhanced image, and dividing the enhanced image into ecological regions based on the pixel grayscale values ​​to identify different ecological regions; Step S2 includes the following sub-steps:

[0075] See also Figure 3 As shown, in step S201, the pixel point located at the nth row and the mth column in the enhanced image is marked as P(n,m) by a symbol, and the grayscale value of the pixel at P(n,m) is marked as C(n,m);

[0076] Step S202, obtaining the vegetation index corresponding to C(n,m), marked as V(n,m);

[0077] Step S203: randomly select a P(n,m) in the enhanced image as the region starting point, and mark the region starting point V(n,m) as the region index;

[0078] Step S204: query the V(n,m) of all P(n,m) within the eight neighborhoods of the region starting point, mark the P(n,m) with the same V(n,m) and region index as region points, then query the V(n,m) of all P(n,m) within the eight neighborhoods of the region point except the region point and the region starting point, mark the P(n,m) with the same V(n,m) and region index as region points, and repeat the process until there are no more consecutive P(n,m) with the same V(n,m) and region index, and the consecutive region points form an ecological region.

[0079] Step S205: Analyze again using any P(n,m) in the enhanced image other than the ecological region as the region starting point until all pixels in the enhanced image are included in the ecological region, thereby obtaining different ecological regions;

[0080] In specific implementation, the vegetation index is usually distinguished by fixed colors in remote sensing images, and the vegetation index represented by each color is a range. In this embodiment, the median of the vegetation index range is taken as the vegetation index for analysis, and different colors have different corresponding grayscale values ​​after grayscale conversion. Therefore, the vegetation index corresponding to the pixel point can be directly obtained based on the grayscale value of the pixel point. The vegetation index is an existing professional term and is obtained based on existing calculation standards. The calculation of the vegetation index is usually completed automatically in remote sensing images. Due to the large number of pixels, it is not convenient to specifically display the pixel grayscale value of each pixel point, and the division process of the ecological area is relatively clear in step S2. It will not be described in detail in this embodiment, and only the ecological area obtained by the final division is given, such as Figure 3 As shown, Figure 3 The figure includes 12 ecological regions. The white curves are the boundaries of the ecological regions, and the numbers are the codes of the ecological regions. In this embodiment, they are called Regions 1 to 12, respectively. The regional indexes of Regions 1 and 2 are both 0.55, while the regional indexes of Regions 3 to 11 are all 0.85, and the regional index of Region 12 is 0.25. Since this embodiment only uses a small part of the observation area as an example to illustrate the ecological biomass estimation method, the remote sensing image only shows a small part of the image to reduce the amount of data displayed. Therefore, there are only three regional indices. In actual applications, when estimating large ecological regions, there will be more regional indices, which will help improve the accuracy of the estimation results.

[0081] Step S3, performing ground survey according to the divided ecological regions and extracting ground observation data; Step S3 includes the following sub-steps:

[0082] Step S301, when performing ground survey on any ecological region, it is named as a survey area;

[0083] Step S302: Preset the survey length, find the highest point and the lowest point in the survey area, and demarcate a square area with a side length of the survey length at each of the highest point and the lowest point, naming them the highest point survey area and the lowest point survey area;

[0084] Step S303, performing ground survey on the ecological biomass in the highest point survey area and the lowest point survey area, naming them as the highest point biomass and the lowest point biomass, represented by symbols Hb and Mb respectively;

[0085] Step S304, calculate the square of the survey length, named the survey area, represented by the symbol Sa, and calculate Hb / Sa and Mb / Sa to obtain the highest point biomass density and the lowest point biomass density respectively;

[0086] Step S305, analyzing the average value of the terrain height in the survey area based on the terrain height in the remote sensing data, and naming it average terrain;

[0087] In specific implementation, the survey length is determined by the surveyor based on the results of the field investigation. The survey length is only used to limit the area of ​​the highest point survey area and the lowest point survey area. The area of ​​each highest point survey area and the lowest point survey area must be equal, and then the biomass density can be calculated. In this embodiment, the survey length is set to 10m, that is, an area of ​​100m2 is selected with the lowest point and the highest point as the center within the survey area, and the amount of biomass therein is surveyed to obtain the highest point biomass Hb and the lowest point biomass Mb respectively. Taking Area 1 as an example, the highest point biomass Hb of Area 1 is 2.9 tons / 100m2, and the lowest point biomass Mb is 2.1 tons / 100㎡, and the corresponding terrain heights are 62m and 16m respectively. The terrain height is actually the difference between the altitude of other areas in the observation area and the lowest altitude when the lowest altitude in the observation area is 0m. The terrain height of 62m means that the highest point in area 1 is 62m higher than the lowest altitude in the observation area. The survey area is 100㎡. The calculated biodensity at the highest point and the lowest point are 29kg / ㎡ and 21kg / ㎡ respectively. Biodensity is actually still biomass, which is just the average value per square meter of the biomass within 100㎡. The average terrain in the survey area is 41m according to the terrain height analysis in the remote sensing data.

[0088] See also Figure 4 As shown, in step S306, a plane rectangular coordinate system is established with the terrain height as the X-axis and the biomass density as the Y-axis, and the system is named a terrain biomass estimation map. The highest point biomass of the highest point in the terrain and the lowest point biomass of the lowest point in the terrain are entered into the terrain biomass estimation map, and the coordinate points in the terrain biomass estimation map are named terrain biomass points;

[0089] Step S307, solving the equation of the line passing through the two topographic biological points, which is named the topographic biological estimation equation;

[0090] Step S308: Substitute the average topography into the topography biomass estimation equation to obtain the average biomass density of the survey area. Calculate the average biomass density of each ecological region. The biomass density at the highest point, the biomass density at the lowest point, and the average biomass density are the ground observation data.

[0091] In the specific implementation, the terrain biomass estimation map is constructed as follows Figure 4As shown, the straight line corresponding to the terrain biological estimation equation is also given. The terrain biological estimation equation is BDY=0.0174×THT+1.8217, where BDY is the biological density and THT is the terrain height. Substituting the average terrain into the terrain biological estimation equation, the average biological density is 2.54kg / ㎡. The calculation result is retained to two decimal places, and finally the highest point biological density is 2.9kg / ㎡, the lowest point biological density is 2.1kg / ㎡, and the average biological density is 2.54kg / ㎡, which is the ground observation data of Area 1. The ground observation data of each ecological area are analyzed and calculated.

[0092] Step S4, dividing similar ecological regions into a group of ecological region groups; specifically, dividing similar ecological regions into a group of ecological region groups includes dividing ecological regions with the same regional index into the same group of ecological region groups;

[0093] In specific implementation, for example Figure 3 Among them, the regional indexes of areas 1 and 2 are both 0.55, while the regional indexes of areas 3 to 11 are all 0.85, and the regional index of area 12 is 0.25. Finally, areas 1 and 2 are divided into one ecological group, areas 3 to 11 are divided into one ecological group, and area 12 is divided into a separate ecological group.

[0094] Step S5, analyzing the ground observation data within the ecological zone group and calculating the fluctuation index of the ecological biomass within similar ecological zones; Step S5 includes the following sub-steps:

[0095] Step S501: Name the ecological regions within the ecological region group as intra-group regions, and calculate the difference between the highest point and the lowest point in the intra-group region, which is named as the terrain drop;

[0096] Step S502, finding the region within the group with the largest terrain drop, naming it the benchmark region, obtaining a terrain biomass estimation map of the benchmark region, naming it the benchmark estimation map, and naming the corresponding terrain biomass estimation equation the benchmark equation;

[0097] In the specific implementation, the ecological zone group consisting of zone 1 and zone 2 is taken as an example. The terrain drop of zone 1 is 46m, while the terrain drop of zone 2 is 28m. Therefore, zone 1 is taken as the benchmark area, and the terrain biomass estimation map of zone 1 is obtained as the benchmark estimation map, that is, Figure 4 For the benchmark estimation map, BDY = 0.0174 × THT + 1.8217 is the benchmark equation. The area with the largest relief difference is used as the benchmark area to ensure that the area can cover the biomass distribution of all relief heights in the ecoregion group, which can be more consistent with the actual situation.

[0098] See also Figure 5As shown, in step S503, the highest point biodensity, the lowest point biodensity, and the average biodensity of all ecological regions in the ecological region group are entered into the benchmark estimation map according to the corresponding terrain height;

[0099] Step S504: Name the coordinate point in the benchmark estimation map as the benchmark coordinate point. When analyzing any benchmark coordinate point, substitute the elevation of the benchmark coordinate point into the benchmark equation to obtain the benchmark density. Then, calculate the difference between the biological density corresponding to the benchmark coordinate point and the benchmark density, which is named the density fluctuation value.

[0100] Step S505: Analyze all reference coordinate points to obtain different density fluctuation values, find the maximum value of the density fluctuation value, and mark it as a fluctuation index;

[0101] In the specific implementation, the highest point biomass density, the lowest point biomass density and the average biomass density of the 2 areas were entered into the benchmark estimation map according to the corresponding terrain height, and the Figure 5 The three gray coordinate points are the coordinate points of Area 2. Calculating the density fluctuation value is actually calculating the residual between the three reference coordinate points entered in Area 2 and the reference equation. This will not be described in detail in this embodiment. The maximum value of the density fluctuation value is the fluctuation index, which represents the fluctuation range of biological density in the area at the same elevation within this ecological zone group. In this embodiment, the fluctuation index of the ecological zone group composed of Areas 1 and 2 is calculated to be 0.2 kg / m2.

[0102] Step S6, analyzing the ground observation data in different ecological zone groups, and calculating the density relationship of ecological biomass in different ecological zone groups; Step S6 includes the following sub-steps:

[0103] Step S601: number the ecological zones in descending order of vegetation index, using the symbol T i Represents, where i is a positive integer and i is the serial number of T;

[0104] Step S602, obtain T i The reference equation, labeled E i , get T i The height of the terrain corresponding to the lowest biomass density is marked as TH i ;

[0105] Step S603, calculate and find TH i The minimum value in is named the lowest terrain, and the lowest terrain is substituted into each E i , find each T i The density of organisms at the lowest elevation, marked as R i ;

[0106] Step S604, calculate R iThe ratio of the biomass density at the lowest point to the lowest elevation, marked as W i ;

[0107] See also Figure 6 As shown, in step S605, i is the horizontal axis, W i Establish a plane rectangular coordinate system for the vertical axis, named the lowest density relationship diagram, and set W i Enter the lowest density relationship diagram according to i;

[0108] In the specific implementation, T1 to T3 are obtained by numbering, which are the ecological zone group consisting of 12 zones, the ecological zone group consisting of zones 1 and 2, and the ecological zone group consisting of zones 3 to 11. At the same time, E1 to E3 are obtained to obtain TH i In fact, the purpose is to obtain the lowest terrain height in the ecological zone group. The values ​​of TH1 to TH3 are 0m, 16m and 12m respectively. The lowest terrain height is 0m. Substitute 0m into each E i , the solution is R1 to R3 are 1.68kg / ㎡, 1.82kg / ㎡ and 1.94kg / ㎡ respectively, the calculation is W1 to W3 are 1, 1.08 and 1.15 respectively, and the minimum density relationship diagram is constructed as follows Figure 6 As shown, since the vegetation included in this embodiment is relatively small, the value of i is small and it is impossible to display every vegetation index. In actual applications, it is necessary to analyze each vegetation, and the legend of vegetation index in remote sensing images is usually below 0.2, 0.2 to 0.3, 0.3 to 0.4, 0.4 to 0.5, 0.5 to 0.6, 0.6 to 0.7 and above 0.7, so the actual value of i is 1 to 7. In this embodiment, the regional index of T1 is 0.25, which is actually the regional index when i=2, the regional index of T2 is 0.55, which is actually the regional index when i=5, and the regional index of T3 is 0.85, which is actually the regional index when i=7. Figure 6 Corrections have been made in all.

[0109] Step S7: construct an ecological biomass estimation model, conduct in-depth learning on the fluctuation index and density relationship of the ecological zone group, and estimate the ecological biomass in other areas to be estimated. Step S7 includes the following sub-steps:

[0110] Step S701, constructing an ecological biomass estimation model and conducting deep learning on the relationship between fluctuation indicators and density of ecological zones;

[0111] Step S701 includes the following sub-steps:

[0112] Step S701.1, constructing an ecological biomass estimation model;

[0113] Step S701.2, analyzing the first number of observation areas and obtaining the fluctuation index and density relationship;

[0114] Step S701.3, calculate the average value of all fluctuation indices and name it density fluctuation parameter;

[0115] Step S701.4, merging the minimum density relationship graphs of different observation areas into a single graph to obtain a comprehensive minimum density relationship graph;

[0116] Step S701.5, performing discrete regression analysis on the minimum density relationship comprehensive graph, and taking the discrete regression equation with the lowest standard deviation as the minimum density relationship comprehensive equation;

[0117] In the specific implementation, the existing convolutional neural network is used to construct the ecological biomass estimation model. The first number is set by the surveyor based on the survey task. The larger the first number, the more accurate the estimation result. Each time the observation area is selected, forests in different climate zones and altitudes can be selected as references. However, the survey task requires funds, and the surveyor decides on his own based on the budget. In this embodiment, the first number is set to 2, that is, two observation areas are analyzed, and finally two fluctuation indicators are obtained, which are 0.2kg / ㎡ and 0.4kg / ㎡ respectively. The calculated density fluctuation parameter is 0.3kg / ㎡. The minimum density relationship diagrams of the two observation areas are merged into the same minimum density relationship comprehensive diagram in order to comprehensively consider the effects of different climates and altitudes on W i The influence of the terrain height is based on the altitude of the observation area itself. For example, the lowest altitude of observation area 1 is 1000m, and the lowest altitude of observation area 2 is 2000m. The terrain height of the area at 1006m and the area at 2006m are both 6m. That is, based on the biological density at the lowest altitude, the change range of biological density is how much when the altitude increases by how many meters. Finally, the comprehensive equation of the minimum density relationship is obtained through discrete regression analysis: W i =0.0017×i 2 +0.015×i+0.9633.

[0118] Step S702, estimating the ecological biomass in other areas to be estimated using an ecological biomass estimation model;

[0119] Step S702 includes the following sub-steps:

[0120] Step S702.1: Mark the area where ecological biomass estimation is required as the estimated area, obtain a remote sensing image of the estimated area, and extract vegetation indicators at the lowest terrain level, naming them the indicators to be analyzed. Also, the lowest terrain level is named the terrain to be analyzed.

[0121] Step S702.2, performing ground survey at the terrain to be analyzed to obtain the biological density at the terrain to be analyzed, which is named as the density to be analyzed;

[0122] Step S702.3, find the value of i corresponding to the indicator to be analyzed in the lowest density relationship comprehensive graph and name it as the value to be analyzed;

[0123] Step S702.4: Name the coordinate point with the horizontal axis equal to the value to be analyzed and the vertical axis equal to the density to be analyzed the coordinate point to be analyzed and enter it into the minimum density relationship comprehensive graph. Vertically move the line corresponding to the minimum density relationship comprehensive equation to the coordinate point to be analyzed to obtain a new equation, named the minimum density relationship target equation.

[0124] In the specific implementation, the terrain to be analyzed is fixed at 0m, the index to be analyzed is extracted as 0.35, the density to be analyzed is 2.3kg / ㎡, and the vegetation index of 0.35 is found to be 3, that is, the value to be analyzed is 3. The density to be analyzed of the terrain to be analyzed is calculated in W i When , it is divided by itself, that is, the calculation result is fixed to 1, so the coordinate point to be analyzed is (value to be analyzed, 1), the coordinate point to be analyzed (3, 1) is entered into the minimum density relationship comprehensive map, and then the straight line corresponding to the minimum density relationship comprehensive equation is vertically moved to the coordinate point to be analyzed. This step is to change the minimum density relationship target equation into a function equation suitable for the area to be estimated under the premise of ensuring that the change law of the ratio relationship of the biological density of different vegetation indicators remains unchanged, and the minimum density relationship target equation is obtained as W i =0.0017×i 2 +0.015×i+0.9397;

[0125] Step S702.5: When analyzing any ecological zone group within the area to be estimated, name it the zone group to be analyzed, obtain a baseline estimation map and a baseline estimation equation for the ecological zone group with the same regional indicators as the zone group to be analyzed, and name them the baseline map to be analyzed and the baseline equation to be analyzed;

[0126] Step S702.6: Substitute the value i corresponding to the block group to be analyzed into the target equation for the minimum density relationship. Multiply the calculated value by the density to be analyzed to obtain the target minimum density. At the same time, obtain the lowest terrain corresponding to the target equation for the minimum density relationship, which is named the target minimum terrain.

[0127] Step S702.7: Move the line corresponding to the benchmark equation to be analyzed vertically to the coordinate point (target lowest terrain, target lowest density) to obtain a new equation, named the target benchmark equation;

[0128] In a specific implementation, for example, when analyzing the to-be-analyzed area group with a regional index of 0.55, the benchmark estimation map and the benchmark estimation equation of the ecological area group with a regional index of 0.55 in this embodiment are obtained as the to-be-analyzed benchmark map and the to-be-analyzed benchmark equation, which are respectively Figure 4 And BDY=0.0174×THT+1.8217, the i corresponding to the block to be analyzed is 5, substitute i=5 into W i =0.0017×i 2 +0.015×i+0.9397 to obtain W5 as 1.06. The result is rounded to two decimal places and then multiplied by the density to be analyzed, 2.3kg / ㎡, to obtain the target minimum density of 2.438kg / ㎡. The target minimum terrain is fixed at 0m. The straight line corresponding to the benchmark equation to be analyzed is vertically moved to the coordinate point (0,2.438). The final target benchmark equation is BDY=0.0174×THT+2.4.

[0129] Step S702.8: Obtain the average value of the terrain height of the block to be analyzed, named average height, and substitute the average height into the target benchmark equation to obtain the block average density of the block to be analyzed. The fluctuation index is increased and decreased by the block average density to obtain the minimum density and maximum density of the block, respectively.

[0130] Step S702.9, calculate the area of ​​the block to be analyzed, named block area, and multiply the block area by the minimum block density and the maximum block density to obtain the minimum biomass and the maximum biomass of the block to be analyzed;

[0131] Step S702.10, calculating the minimum and maximum biomass of each ecological zone group within the estimated area, and summing the minimum biomasses to obtain the estimated minimum biomass of the estimated area, and summing the maximum biomasses to obtain the estimated maximum biomass of the estimated area;

[0132] In the specific implementation, the average height is obtained to be 38m, and the average density of the block group is 3.1kg / ㎡. The calculation result is retained to one decimal place, and the fluctuation index of the average density of the block group is increased and decreased respectively, and the minimum density of the block group is obtained to be 2.9kg / ㎡, and the maximum density of the block group is 3.4kg / ㎡. If the conventional definition of biomass is used for estimation, 2.9kg / ㎡ and 3.4kg / ㎡ are the estimated biomass of the block group to be analyzed. If a specific value is required, the area of ​​the block group is added to obtain the total amount of organic matter actually surviving in the block group to be analyzed. However, under normal circumstances, the total amount of organic matter actually surviving per unit area at a certain moment is estimated, that is, the minimum density and maximum density of the block group in this embodiment. The calculation of the added area is no longer specifically described in this embodiment.

[0133] In Example 2, the present application provides an electronic device, which may include: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus. The memory stores computer-readable instructions, and the processor can call the instructions in the memory. When the computer-readable instructions are executed by the processor, the steps in the ecological biomass estimation method based on the fusion of remote sensing images and ground observation data are executed to achieve the following functions: obtaining remote sensing images of the observation area and performing image preprocessing on the remote sensing images to obtain enhanced images; dividing different ecological regions; extracting ground observation data; dividing similar ecological regions into a group of ecological zone groups; calculating the fluctuation index of ecological biomass within similar ecological regions; analyzing the ground observation data within different ecological zone groups to calculate the density relationship of ecological biomass within different ecological zone groups; and estimating the ecological biomass in other areas to be estimated.

[0134] In addition, the logical instructions in the above-mentioned memory can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0135] Example 3. The present application also provides a computer program product, which includes a computer program stored on a computer-readable storage medium, and the computer program includes program instructions. When the program instructions are executed by a computer, the computer can execute the ecological biomass estimation method based on the fusion of remote sensing images and ground observation data provided by the above methods. The method includes: obtaining remote sensing images of the observation area, and performing image preprocessing on the remote sensing images to obtain enhanced images; dividing different ecological regions; extracting ground observation data; dividing similar ecological regions into a group of ecological zone groups; calculating the fluctuation index of ecological biomass within similar ecological regions; analyzing ground observation data within different ecological zone groups, and calculating the density relationship of ecological biomass within different ecological zone groups; and estimating the ecological biomass in other areas to be estimated.

[0136] Example 4. The present application also provides a computer-readable storage medium. The present application provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the ecological biomass estimation method based on the fusion of remote sensing images and ground observation data are executed to achieve the following functions: obtain remote sensing images of the observation area, and perform image preprocessing on the remote sensing images to obtain enhanced images; divide different ecological regions; extract ground observation data; divide similar ecological regions into a group of ecological zone groups; calculate the fluctuation index of ecological biomass within similar ecological regions; analyze the ground observation data within different ecological zone groups, and calculate the density relationship of ecological biomass within different ecological zone groups; and estimate the ecological biomass in other areas to be estimated.

[0137] Through the description of the above embodiments, the embodiments of the present invention can be provided as methods, systems or computer program products. Based on this understanding, the above technical solutions, in essence or in other words, the part that contributes to the prior art, can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiment.

[0138] In the embodiments provided in this application, it should be understood that the disclosed system or method can be implemented in other ways. The embodiments described above are merely illustrative. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation. For example, multiple modules or units can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of systems, modules and units can be electrical, mechanical or other forms.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application 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 make equivalent replacements for some of the technical features therein. However, 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 embodiments of the present application.

Claims

1. The ecological biomass estimation method based on the fusion of remote sensing images and ground observation data is characterized by: The steps include: Acquire remote sensing images of the observation area and perform image preprocessing on the remote sensing images to obtain enhanced images; Extracting pixel grayscale values ​​from the enhanced image, dividing the enhanced image into ecological regions based on the pixel grayscale values, and dividing different ecological regions; Conduct ground surveys based on the divided ecological regions and extract ground observation data; Grouping similar ecological regions into a set of ecoregional groups; Analyze ground observation data within ecological region groups and calculate fluctuation indicators of ecological biomass within similar ecological regions; Analyze the ground observation data in different ecological groups and calculate the density relationship of ecological biomass in different ecological groups; Construct an ecological biomass estimation model, conduct in-depth learning on the fluctuation indicators and density relationships of ecological zone groups, and estimate the ecological biomass in other areas to be estimated.

2. The ecological biomass estimation method based on the fusion of remote sensing images and ground observation data according to claim 1 is characterized in that: The image preprocessing includes standardizing and denoising the remote sensing image, and then grayscale processing. After the image preprocessing is completed, an enhanced image can be obtained.

3. The ecological biomass estimation method based on the fusion of remote sensing images and ground observation data according to claim 2 is characterized in that: Extracting pixel grayscale values ​​from the enhanced image and dividing the enhanced image into ecological regions based on the pixel grayscale values ​​includes the following sub-steps: The pixel at row n and column m in the enhanced image is marked as P(n,m), and the grayscale value of the pixel at P(n,m) is marked as C(n,m); Get the vegetation index corresponding to C(n,m), marked as V(n,m); Randomly select a P(n,m) in the enhanced image as the starting point of the region, and mark the starting point V(n,m) of the region as the region index; Query the V(n,m) of all P(n,m) in the eight neighborhoods of the regional starting point, mark the P(n,m) with the same V(n,m) and regional index as regional points, then query the V(n,m) of all P(n,m) in the eight neighborhoods of the regional point except the regional point and the regional starting point, mark the P(n,m) with the same V(n,m) and regional index as regional points, and execute the loop until there are no more consecutive adjacent P(n,m) with the same V(n,m) and regional index. The continuous regional points form an ecological region. Again, any P(n,m) in the enhanced image except the ecological region is used as the starting point of the region for analysis until all pixels in the enhanced image are included in the ecological region, and different ecological regions are obtained.

4. The ecological biomass estimation method based on the fusion of remote sensing images and ground observation data according to claim 3 is characterized in that: Conduct ground surveys based on the divided ecological regions and extract ground observation data, including the following sub-steps: When a ground survey is conducted for any ecological region, it is named as the survey area; Preset the survey length, find the highest point and the lowest point in the survey area, and demarcate a square area with a side length of the survey length at the highest point and the lowest point respectively, naming them the highest point survey area and the lowest point survey area; The ecological biomass in the highest point survey area and the lowest point survey area were surveyed on the ground and named as the highest point biomass and the lowest point biomass, respectively, and represented by the symbols Hb and Mb; Calculate the square of the survey length, named the survey area, represented by the symbol Sa, and calculate Hb / Sa and Mb / Sa to obtain the highest point biomass density and the lowest point biomass density respectively; The average value of the terrain height in the survey area is analyzed based on the terrain height in the remote sensing data and is named as the average terrain height; A plane rectangular coordinate system is established with the terrain height as the X-axis and the biomass density as the Y-axis, and is named the terrain biomass estimation map. The highest point biomass of the highest point and the lowest point biomass of the lowest point are recorded in the terrain biomass estimation map, and the coordinate points in the terrain biomass estimation map are named terrain biomass points. Solve the equation of the line passing through the two points of the geobiotaxis, which is named the geobiotaxis estimation equation; Substitute the average terrain into the terrain biological estimation equation to obtain the average biological density of the survey area, and calculate the average biological density of each ecological region. The highest point biological density, the lowest point biological density and the average biological density are the ground observation data.

5. The ecological biomass estimation method based on the fusion of remote sensing images and ground observation data according to claim 4 is characterized in that: Dividing similar ecological regions into a group of ecological groups specifically means dividing ecological regions with the same regional index into the same group of ecological groups.

6. The ecological biomass estimation method based on the fusion of remote sensing images and ground observation data according to claim 5 is characterized in that: Analyzing ground observation data within ecological zone groups and calculating the fluctuation index of ecological biomass within similar ecological zones includes the following sub-steps: The ecological regions within the ecological zone group are named as intra-group regions, and the difference between the highest and lowest points in the intra-group region is calculated and named as the relief drop; Find the area within the group with the largest relief drop, name it the benchmark area, obtain the relief biomass estimation map of the benchmark area, name it the benchmark estimation map, and name the corresponding relief biomass estimation equation the benchmark equation; Enter the highest point biomass density, lowest point biomass density, and average biomass density of all ecoregions within the ecoregion group into the baseline estimation map according to the corresponding terrain height; The coordinate points in the benchmark estimation map are named benchmark coordinate points. When analyzing any benchmark coordinate point, the elevation of the benchmark coordinate point is substituted into the benchmark equation to obtain the benchmark density. The difference between the biological density corresponding to the benchmark coordinate point and the benchmark density is then calculated and named the density fluctuation value. Analyze all reference coordinate points to obtain different density fluctuation values, find the maximum value of the density fluctuation value, and mark it as the fluctuation index.

7. The ecological biomass estimation method based on the fusion of remote sensing images and ground observation data according to claim 6 is characterized in that: Analyze the ground observation data in different ecological groups and calculate the density relationship of ecological biomass in different ecological groups; The ecological zones are numbered in descending order according to the vegetation index, and the symbol T i Represents, where i is a positive integer and i is the serial number of T; Get T i The reference equation, labeled E i , get T i The height of the terrain corresponding to the lowest biomass density is marked as TH i ; Calculate and find TH i The minimum value in is named the lowest terrain, and the lowest terrain is substituted into each E i , find each T i The density of organisms at the lowest elevation, marked as R i ; Calculate R i The ratio of the biomass density at the lowest point to the lowest elevation, marked as W i ; With i as the horizontal axis, W i Establish a plane rectangular coordinate system for the vertical axis, named the lowest density relationship diagram, and set W i Enter the lowest density relationship diagram according to i.

8. The ecological biomass estimation method based on the fusion of remote sensing images and ground observation data according to claim 7 is characterized in that: Constructing an ecological biomass estimation model, conducting deep learning on the relationship between fluctuation indicators and density of ecological zones, and estimating ecological biomass in other areas to be estimated includes the following sub-steps: Construct an ecological biomass estimation model and conduct deep learning on the relationship between fluctuation indicators and density of ecological groups; The ecological biomass in other areas to be estimated is estimated using the ecological biomass estimation model.

9. The ecological biomass estimation method based on the fusion of remote sensing images and ground observation data according to claim 8 is characterized in that: Constructing an ecological biomass estimation model and conducting deep learning on the relationship between fluctuation indicators and density of ecological groups includes the following sub-steps: Constructing ecological biomass estimation models; Analyzing a first number of observation areas and obtaining fluctuation indices and density relationships; Calculate the average value of all volatility indicators and name it density volatility parameter; The minimum density relationship maps of different observation areas are merged into the same map to obtain a comprehensive minimum density relationship map; A discrete regression analysis was performed on the lowest density relationship comprehensive graph, and the discrete regression equation with the lowest standard deviation was taken as the lowest density relationship comprehensive equation.

10. The ecological biomass estimation method based on the fusion of remote sensing images and ground observation data according to claim 9 is characterized in that: Estimating the ecological biomass in other areas to be estimated using the ecological biomass estimation model includes the following sub-steps: The area where ecological biomass estimation is required is marked as the area to be estimated, remote sensing images of the area to be estimated are obtained, and vegetation indicators at the lowest terrain are extracted and named as indicators to be analyzed. At the same time, the lowest terrain is named as terrain to be analyzed. Conduct ground survey at the terrain to be analyzed to obtain the biological density at the terrain to be analyzed, which is named as the density to be analyzed; Find the value of i corresponding to the indicator to be analyzed in the lowest density relationship comprehensive graph and name it as the value to be analyzed; Name the coordinate point with the horizontal axis equal to the value to be analyzed and the vertical axis equal to the density to be analyzed as the coordinate point to be analyzed and enter it into the minimum density relationship comprehensive graph. Move the line corresponding to the minimum density relationship comprehensive equation vertically to the coordinate point to be analyzed to obtain a new equation, which is named the minimum density relationship target equation. When analyzing any ecological zone group within the area to be estimated, it is named as the zone group to be analyzed, and the benchmark estimation map and benchmark estimation equation of the ecological zone group with the same regional indicators as the zone group to be analyzed are obtained and named as the benchmark map to be analyzed and the benchmark equation to be analyzed; Substitute the i corresponding to the area group to be analyzed into the minimum density relationship target equation, multiply the calculated result by the density to be analyzed to obtain the target minimum density, and at the same time obtain the lowest terrain corresponding to the minimum density relationship target equation, which is named the target minimum terrain; Move the straight line corresponding to the benchmark equation to be analyzed vertically to the coordinate point (target lowest terrain, target lowest density) to obtain a new equation, which is named the target benchmark equation; Obtain the average value of the terrain height of the block to be analyzed, named as the average height, substitute the average height into the target benchmark equation, solve to obtain the block group average density of the block to be analyzed, increase and decrease the fluctuation index of the block group average density respectively, and obtain the minimum density and maximum density of the block group respectively; Calculate the area of ​​the block to be analyzed, named block area, and multiply the block area by the minimum density and maximum density of the block to obtain the minimum biomass and maximum biomass of the block to be analyzed; Calculate the minimum and maximum biomass of each ecological zone group in the area to be estimated, and calculate the sum of the minimum biomass to obtain the minimum estimated biomass of the area to be estimated, and calculate the sum of the maximum biomass to obtain the maximum estimated biomass of the area to be estimated.

Images