A carbon flux monitoring method and system based on soil drought remote sensing data

Abstract

The present invention provides a method and system for monitoring carbon flux based on soil drought remote sensing data. The method includes: dividing the monitoring area into multiple monitoring sub-areas according to the categories of planted crops; measuring the drought degree and carbon flux of sampling points in each monitoring sub-area in different months; fitting a change curve of carbon flux under different drought degrees of the monitoring sub-area according to the drought degree and carbon flux of the monitoring sub-area; setting identification codes for each monitoring sub-area respectively; associating the identification codes of the monitoring sub-area with the change curve; obtaining the real-time remote sensing data of the monitoring sub-area; calculating the real-time drought degree of the monitoring sub-area according to the real-time remote sensing data; and predicting the real-time carbon flux of the monitoring sub-area according to the change curve associated with the identification code of the monitoring sub-area. This method can use remote sensing data to predict the carbon flux of a large-area monitoring area, and solve the defects of the traditional carbon flux measurement method, which is time-consuming, laborious and inefficient.

Description

Technical Field

[0001] The present invention belongs to the technical field of agriculture, and particularly relates to a method and system for monitoring carbon flux based on soil drought remote sensing data. Background Art

[0002] Traditional methods for measuring carbon flux include the alkali absorption method and the greenhouse method. The alkali absorption method is to cover the upper surface with a container, inside which there is an open vessel filled with an alkali absorption solution. After a period of time, the amount of soil CO2 release is calculated from the change in the pH value of the alkali solution. The greenhouse method is to tightly cover the soil surface with a plastic greenhouse or a corresponding container of appropriate size, and the amount of soil CO2 release is calculated using the difference in CO2 concentration in the greenhouse over a certain period of time. These methods have the following defects: they require manual measurement, are time-consuming, and are only applicable to the measurement of carbon flux in a small area. When the monitoring area is large, the traditional measurement methods are inefficient. Summary of the Invention

[0003] Aiming at the defects in the prior art, the present invention provides a method and system for monitoring carbon flux based on soil drought remote sensing data, which solves the defects of the traditional carbon flux measurement methods being time-consuming, laborious, and inefficient.

[0004] In a first aspect, a method for monitoring carbon flux based on soil drought remote sensing data includes:

[0005] Dividing the monitoring area into multiple monitoring sub-areas according to the types of planted crops;

[0006] Measuring the drought degree and carbon flux of the sampling points in each monitoring sub-area in different months;

[0007] According to the drought degree and carbon flux of the monitoring sub-area, fitting the change curve of carbon flux under different drought degrees of the monitoring sub-area;

[0008] Setting identification codes for each monitoring sub-area respectively, and setting the same identification code for the monitoring sub-areas with the same change curve;

[0009] Associating the identification code of the monitoring sub-area with the change curve;

[0010] Obtaining the real-time remote sensing data of the monitoring sub-area;

[0011] Calculating the real-time drought degree of the monitoring sub-area according to the real-time remote sensing data;

[0012] Predicting the real-time carbon flux of the monitoring sub-area according to the change curve associated with the identification code of the monitoring sub-area.

[0013] Further, the method for measuring the drought degree includes:

[0014] Setting weights for rainfall and soil water content respectively according to the types of planted crops;

[0015] Obtain the rainfall from the ground station;

[0016] Obtain SAR satellite images and analyze the SAR satellite images to obtain soil moisture content;

[0017] Multiply the rainfall and the soil moisture content by their corresponding weights respectively to obtain the drought degree.

[0018] Furthermore, the real-time remote sensing data includes remote sensing images;

[0019] Calculating the real-time drought degree of the monitoring sub-region based on the real-time remote sensing data specifically includes:

[0020] Create a neural network model;

[0021] Use the neural network model to extract the low-level features of the remote sensing image;

[0022] Enhance the low-level features using attention mechanisms of different scales;

[0023] Fuse the features enhanced at all scales to obtain the high-level features of the remote sensing image;

[0024] Fuse the low-level features and the high-level features, and the fused features;

[0025] Restore the fused features to the size of the remote sensing image to obtain a feature image;

[0026] Calculate the real-time drought degree of the monitoring sub-region based on the feature image.

[0027] Furthermore, the real-time remote sensing data includes climate information; the climate information includes temperature, humidity, and sunshine duration;

[0028] Calculating the real-time drought degree of the monitoring sub-region based on the real-time remote sensing data specifically includes:

[0029] After denoising and smoothing the climate information, substitute it into a preset vegetation index to obtain the real-time drought degree.

[0030] Furthermore, it also includes:

[0031] Set up a display interface;

[0032] Display a map including the monitoring sub-region on the display interface;

[0033] Set different display colors for different carbon fluxes;

[0034] Set the color of the monitoring sub-region in the map to the display color corresponding to its carbon flux; among them, the monitoring sub-regions with the same display color are connected to each other;

[0035] When a display color is selected on the map, all connected monitored sub-regions corresponding to the display color are highlighted.

[0036] In a second aspect, a carbon flux monitoring system based on soil drought remote sensing data includes:

[0037] A sampling unit: used to divide the monitored area into multiple monitored sub-regions according to the categories of planted crops; measure the drought degree and carbon flux of sampling points in each monitored sub-region in different months; and also used to obtain real-time remote sensing data of the monitored sub-region;

[0038] A fitting unit: used to fit the change curve of carbon flux under different drought degrees of the monitored sub-region according to the drought degree and carbon flux of the monitored sub-region;

[0039] An identification unit: used to set identification codes for each monitored sub-region respectively, where the monitored sub-regions with the same change curve are set with the same identification code;

[0040] An association unit: used to associate the identification code of the monitored sub-region with the change curve;

[0041] A prediction unit: used to calculate the real-time drought degree of the monitored sub-region according to the real-time remote sensing data; predict the real-time carbon flux of the monitored sub-region according to the change curve associated with the identification code of the monitored sub-region.

[0042] Further, the sampling unit is specifically used for:

[0043] Set weights for rainfall and soil moisture content respectively according to the categories of planted crops;

[0044] Obtain rainfall from the ground station;

[0045] Obtain SAR satellite images and analyze the SAR satellite images to obtain soil moisture content;

[0046] Multiply rainfall and soil moisture content by their corresponding weights respectively to obtain the drought degree.

[0047] Further, the real-time remote sensing data includes remote sensing images;

[0048] The prediction unit is specifically used for:

[0049] Create a neural network model;

[0050] Use the neural network model to extract low-level features of the remote sensing image;

[0051] Use attention mechanisms of different scales to enhance the low-level features;

[0052] Fuse the features enhanced at all scales to obtain high-level features of the remote sensing image;

[0053] Fuse low-level features and high-level features, and use the fused features;

[0054] Restore the fused features to the size of the remote sensing image to obtain a feature image;

[0055] Calculate the real-time drought degree of the monitored sub-region based on the feature image.

[0056] Furthermore, the real-time remote sensing data includes climate information; the climate information includes temperature, humidity, and sunshine duration;

[0057] The prediction unit is specifically used for:

[0058] After denoising and smoothing the climate information, substitute it into a preset vegetation index to obtain the real-time drought degree.

[0059] Furthermore, it further includes: The display unit is used for:

[0060] Set up a display interface;

[0061] Display a map including the monitored sub-region on the display interface;

[0062] Set different display colors for different carbon fluxes;

[0063] Set the color of the monitored sub-region in the map to the display color corresponding to its carbon flux; where the monitored sub-regions with the same display color are connected to each other;

[0064] When a display color is selected in the map, highlight all the connected monitored sub-regions corresponding to the display color.

[0065] As can be seen from the above technical solutions, the carbon flux monitoring method and system based on soil drought remote sensing data provided by the present invention can use remote sensing data to predict the carbon fluxes of large-scale monitoring regions, and solve the defects of the traditional carbon flux measurement method, which is time-consuming, laborious, and has low efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for the description of the specific embodiments or the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0067] Figure 1 It is a flowchart of the carbon flux monitoring method provided for the embodiment.

[0068] Figure 2 It is a method for measuring the drought degree provided for the embodiment.

[0069] Figure 3Another method for measuring the degree of drought provided for the embodiment.

[0070] Figure 4 Flow chart showing the method provided for the embodiment.

[0071] Figure 5 Block diagram of the carbon flux monitoring system provided for the embodiment. Detailed implementation manners

[0072] The embodiments of the technical solution of the present invention will be described in detail below with reference to the accompanying drawings. The following embodiments are only used to illustrate the technical solution of the present invention more clearly, so they are only examples and cannot be used to limit the protection scope of the present invention. It should be noted that unless otherwise specified, the technical terms or scientific terms used in this application should be the ordinary meanings understood by those skilled in the art to which the present invention belongs.

[0073] It should be understood that when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, wholes, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or their combinations.

[0074] It should also be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in this specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0075] As used in this specification and the appended claims, the term "if" can be interpreted as "when", "once", "in response to determining" or "in response to detecting" according to the context. Similarly, the phrase "if determined" or "if [the described condition or event] is detected" can be interpreted as meaning "once determined", "in response to determining", "once [the described condition or event] is detected" or "in response to detecting [the described condition or event]" according to the context.

[0076] Embodiment:

[0077] A carbon flux monitoring method based on soil drought remote sensing data, see Figure 1 , including:

[0078] Dividing the monitoring area into multiple monitoring sub-areas according to the categories of the planted crops;

[0079] Measuring the drought degree and carbon flux of the soil at the sampling points in each monitoring sub-area in different months;

[0080] According to the drought degree and carbon flux of the monitoring sub-region, the change curve of carbon flux under different drought degrees in the monitoring sub-region is fitted;

[0081] Identification codes are set for each monitoring sub-region respectively, and the monitoring sub-regions with the same change curve are set with the same identification code;

[0082] The identification code of the monitoring sub-region is associated with the change curve;

[0083] Obtain the real-time remote sensing data of the monitoring sub-region;

[0084] Calculate the real-time drought degree of the monitoring sub-region according to the real-time remote sensing data;

[0085] Predict the real-time carbon flux of the monitoring sub-region according to the change curve associated with the identification code of the monitoring sub-region.

[0086] In this embodiment, since drought can cause plant water stress, resulting in a decrease in photosynthetic rate and leaf area, and thus a decrease in carbon flux. And the impact of drought on the carbon flux of crops with different growth habits is also different. Therefore, when monitoring carbon flux, the monitoring area can be first divided into multiple monitoring sub-regions according to the types of crops planted, for example, the monitoring area is divided into monitoring sub-regions according to crops such as wheat and corn. Then, samples are taken from each monitoring sub-region to monitor the change relationship between drought and carbon flux, and then the real-time carbon flux of the large-area monitoring region is predicted. This method can set multiple sampling points in the monitoring sub-region, and comprehensively evaluate the change relationship between drought and carbon flux in the monitoring sub-region according to multiple sampling points. For example, the sampling points are set at the corners or middle positions of the monitoring sub-region. After determining the sampling points, measure the drought degree and carbon flux of the soil at the sampling points in different months, so that the situation of the sampling points throughout the year can be sampled, and the carbon flux at different times in the monitoring sub-region can be predicted later.

[0087] In this embodiment, the method can fit the drought degree and carbon flux measured at multiple sampling points in the monitored sub-region to obtain the change curve of the carbon flux under different drought degrees in the monitored sub-region. The carbon flux can be calculated by traditional methods, and the change curve reflects the change of the drought degree and carbon flux in the monitored sub-region. Then, the method sets an identification code for each monitored sub-region. For example, assuming that the monitored region is divided into 10 monitored sub-regions, Arabic numerals can be used to identify the monitored sub-regions. If the change curves of 2 out of 10 monitored sub-regions are the same, then Arabic numerals 1-9 are used to identify these 10 monitored sub-regions, where the 2 monitored sub-regions with the same change curve are identified by the same Arabic numeral, and the other 8 are identified by different Arabic numerals. And the identification code of the monitored sub-region is associated with the change curve. In this way, when calculating the carbon flux of a certain monitored sub-region, its change curve can be obtained through the identification code of the monitored sub-region and predicted according to the change curve.

[0088] In this embodiment, when the area of the monitored region is large, the traditional method for measuring carbon flux is time-consuming and laborious. The method obtains real-time remote sensing data of the large-area monitored region, and according to the division of the monitored region, obtains the real-time remote sensing data of each monitored sub-region, calculates the real-time drought degree of the monitored sub-region according to the real-time remote sensing data, and predicts the real-time carbon flux of the monitored sub-region according to the change curve associated with the identification code of the monitored sub-region.

[0089] The carbon flux monitoring method based on soil drought remote sensing data can use remote sensing data to predict the carbon flux of a large-area monitored region, and solve the defects of the traditional carbon flux measurement method, which is time-consuming, laborious and has low efficiency.

[0090] Further, in some embodiments, refer to Figure 2 , the method for measuring the drought degree includes:

[0091] Set weights for rainfall and soil moisture content respectively according to the types of crops planted;

[0092] Obtain the rainfall from the ground station;

[0093] Obtain SAR satellite images, and analyze the SAR satellite images to obtain the soil moisture content;

[0094] Multiply the rainfall and soil moisture content by the corresponding weights respectively to obtain the drought degree.

[0095] In this embodiment, the method can determine the drought degree based on the rainfall and soil moisture content. The method can directly obtain the rainfall through a ground station or measure it by radar. The radar emits microwave signals and measures the rainfall by receiving the echo information reflected by the rainfall layer. The method obtains SAR satellite images and analyzes the SAR satellite images to obtain the soil moisture content. Since the drought characteristics of different crops are different, the method sets weights for each drought influencing factor, and finally multiplies the rainfall and soil moisture content by the corresponding weights to obtain the drought degree.

[0096] Further, in some embodiments, referring to Figure 3 , the real-time remote sensing data includes remote sensing images;

[0097] Calculating the real-time drought degree of the monitoring sub-region based on the real-time remote sensing data specifically includes:

[0098] Create a neural network model;

[0099] Use the neural network model to extract the low-level features of the remote sensing image;

[0100] Enhance the low-level features using attention mechanisms of different scales;

[0101] Fuse the features enhanced at all scales to obtain the high-level features of the remote sensing image;

[0102] Fuse the low-level features and the high-level features, and the fused features;

[0103] Restore the fused features to the size of the remote sensing image to obtain a feature image;

[0104] Calculate the real-time drought degree of the monitoring sub-region based on the feature image.

[0105] In this embodiment, the method can also determine the drought degree through remote sensing images. The method first creates a neural network model and uses the neural network model to extract the low-level features of the remote sensing image. For example, the method can use activation functions and normalization methods to reduce the number of parameters of the remote sensing image, and change the dimension and depth of the neural network model through convolution and residuals to extract features at different dimensions and depths, obtaining low-level features. Then the method uses attention mechanisms of different scales to enhance the low-level features to obtain features enhanced at different scales, and then fuses the features enhanced at all scales to obtain the high-level features of the remote sensing image. Finally, the high-level features are restored to the size of the original image to obtain an effect image containing drought features, and finally the real-time drought degree of the monitoring sub-region is calculated based on the feature image.

[0106] Further, in some embodiments, the real-time remote sensing data includes climate information; the climate information includes temperature, humidity, and sunshine duration;

[0107] Calculating the real-time drought degree of the monitored sub-region based on real-time remote sensing data specifically includes:

[0108] After denoising and smoothing the climate information, substitute it into a preset vegetation index to obtain the real-time drought degree.

[0109] In this embodiment, this method can determine the real-time drought degree according to climate information. This method first obtains the temperature, humidity, and sunshine duration of the monitored area. After denoising and smoothing the temperature, humidity, and sunshine duration, substitute them into a preset vegetation index to obtain the real-time drought degree. The vegetation index is determined according to the planted crops, and different planted crops can select different vegetation indices to measure the drought degree.

[0110] Further, in some embodiments, refer to Figure 4 , and further includes:

[0111] Set up a display interface;

[0112] Display a map including the monitored sub-region in the display interface;

[0113] Set different display colors for different carbon fluxes;

[0114] Set the color of the monitored sub-region in the map to the display color corresponding to its carbon flux; among them, the monitored sub-regions with the same display color are connected to each other;

[0115] When a display color is selected in the map, highlight all the connected monitored sub-regions corresponding to the display color.

[0116] In this embodiment, this method can also display the measured real-time carbon flux. This method can display the display interface through a display screen and display the real-time carbon flux in the form of a map. This method displays a map including the monitored sub-region in the display interface, where the map can display the entire monitored area, and when displaying, it can display the monitored sub-regions divided from the monitored area. In order to facilitate customers to view the carbon fluxes of each monitored sub-region, different colors are used to fill the display areas of the monitored sub-regions, where different colors represent different carbon fluxes. In order to improve the user experience of the display, this method can connect the monitored sub-regions with the same carbon flux, so that when a user selects a monitored sub-region, other monitored sub-regions with the same carbon flux as that in the selected monitored sub-region can be highlighted and displayed simultaneously. When a user selects a display color in the map, this method can also highlight all the connected monitored sub-regions corresponding to the display color.

[0117] A carbon flux monitoring system based on soil drought remote sensing data, refer to Figure 5 , includes:

[0118] Sampling unit: used to divide the monitoring area into multiple monitoring sub-areas according to the categories of planted crops; measure the drought degree and carbon flux of soil at sampling points in each monitoring sub-area in different months; also used to obtain real-time remote sensing data of the monitoring sub-area;

[0119] Fitting unit: used to fit the change curve of carbon flux under different drought degrees of the monitoring sub-area according to the drought degree and carbon flux of the monitoring sub-area;

[0120] Identification unit: used to set identification codes for each monitoring sub-area respectively, and the monitoring sub-areas with the same change curve are set with the same identification code;

[0121] Association unit: used to associate the identification code of the monitoring sub-area with the change curve;

[0122] Prediction unit: used to calculate the real-time drought degree of the monitoring sub-area according to the real-time remote sensing data; predict the real-time carbon flux of the monitoring sub-area according to the change curve associated with the identification code of the monitoring sub-area.

[0123] Furthermore, in some embodiments, the sampling unit is specifically used for:

[0124] Set weights for rainfall and soil water content respectively according to the categories of planted crops;

[0125] Obtain rainfall from the ground station;

[0126] Obtain SAR satellite images, and analyze the SAR satellite images to obtain soil water content;

[0127] Multiply rainfall and soil water content by their corresponding weights respectively to obtain the drought degree.

[0128] Furthermore, in some embodiments, the real-time remote sensing data includes remote sensing images;

[0129] The prediction unit is specifically used for:

[0130] Create a neural network model;

[0131] Use the neural network model to extract low-level features of the remote sensing image;

[0132] Use attention mechanisms of different scales to enhance the low-level features;

[0133] Fuse the features enhanced at all scales to obtain high-level features of the remote sensing image;

[0134] Fuse the low-level features and high-level features, and the fused features;

[0135] Restore the fused features to the size of the remote sensing image to obtain a feature image;

[0136] Calculate the real-time drought degree of the monitored sub-region based on the characteristic image.

[0137] Furthermore, in some embodiments, the real-time remote sensing data includes climate information; the climate information includes temperature, humidity, and sunshine duration;

[0138] The prediction unit is specifically used for:

[0139] After denoising and smoothing the climate information, substitute it into the preset vegetation index to obtain the real-time drought degree.

[0140] Furthermore, in some embodiments, it further includes: The display unit is used for:

[0141] Set up a display interface;

[0142] Display a map including the monitored sub-region on the display interface;

[0143] Set different display colors for different carbon fluxes;

[0144] Set the color of the monitored sub-region in the map to the display color corresponding to its carbon flux; where the monitored sub-regions with the same display color are connected to each other;

[0145] When a display color is selected in the map, highlight all the connected monitored sub-regions corresponding to the display color.

[0146] For the system provided by the embodiments of the present invention, for the sake of brief description, for the parts not mentioned in the embodiment part, reference may be made to the corresponding content in the foregoing embodiments.

[0147] Finally, it should be noted that: The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: It is still possible to modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; And these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered within the scope of the claims and the description of the present invention.

Claims

1. A carbon flux monitoring method based on soil drought remote sensing data, characterized in that Including: Dividing the monitoring area into multiple monitoring sub-areas according to the categories of planted crops; Measuring the drought degree and carbon flux of soil at sampling points in each monitoring sub-area in different months; Fitting the change curve of carbon flux under different drought degrees in the monitoring sub-area according to the drought degree and carbon flux of the monitoring sub-area; Setting identification codes for each monitoring sub-area respectively, where the monitoring sub-areas with the same change curve are set with the same identification code; Associating the identification code of the monitoring sub-area with the change curve; Obtaining the real-time remote sensing data of the monitoring sub-area; Calculating the real-time drought degree of the monitoring sub-area according to the real-time remote sensing data; Predicting the real-time carbon flux of the monitoring sub-area according to the change curve associated with the identification code of the monitoring sub-area; Wherein, the real-time remote sensing data includes remote sensing images; The calculating the real-time drought degree of the monitoring sub-area according to the real-time remote sensing data specifically includes: Creating a neural network model; Using the neural network model to extract the low-level features of the remote sensing image; Enhancing the low-level features by using attention mechanisms of different scales; Fusing the features enhanced at all scales to obtain the high-level features of the remote sensing image; Fusing the low-level features and high-level features, and the fused features; Restoring the fused features to the size of the remote sensing image to obtain a feature image; Calculating the real-time drought degree of the monitoring sub-area according to the feature image.

2. The carbon flux monitoring method based on remote sensing data of soil drought according to claim 1, characterized in that The measuring method of the drought degree includes: Setting weights for rainfall and soil water content respectively according to the categories of planted crops; Obtaining the rainfall from the ground station; Obtaining SAR satellite images and analyzing the SAR satellite images to obtain the soil water content; Multiplying the rainfall and soil water content by the corresponding weights respectively to obtain the drought degree.

3. The carbon flux monitoring method based on soil drought remote sensing data according to claim 2, wherein The real-time remote sensing data includes climate information; the climate information includes temperature, humidity, and sunshine duration; The calculating the real-time drought degree of the monitoring sub-area according to the real-time remote sensing data specifically includes: Denosing and smoothing the climate information and substituting it into a preset vegetation index to obtain the real-time drought degree.

4. The carbon flux monitoring method based on soil drought remote sensing data according to claim 1, characterized in that Also including: Setting a display interface; Displaying a map including the monitoring sub-areas in the display interface; Setting different display colors for different carbon fluxes; Setting the color of the monitoring sub-areas in the map to the display color corresponding to their carbon fluxes; where the monitoring sub-areas with the same display color are connected to each other; When a display color is selected in the map, highlighting all the connected monitoring sub-areas corresponding to the display color.

5. A carbon flux monitoring system based on soil drought remote sensing data, characterized in that, Including: Sampling unit: used for dividing the monitoring area into multiple monitoring sub-areas according to the categories of planted crops; measuring the drought degree and carbon flux of soil at sampling points in each monitoring sub-area in different months; and also used for obtaining the real-time remote sensing data of the monitoring sub-area; Fitting unit: used for fitting the change curve of carbon flux under different drought degrees in the monitoring sub-area according to the drought degree and carbon flux of the monitoring sub-area; Identification unit: used for setting identification codes for each monitoring sub-area respectively, where the monitoring sub-areas with the same change curve are set with the same identification code; Association unit: used for associating the identification code of the monitoring sub-area with the change curve; Prediction unit: used to calculate the real-time drought degree of the monitored sub-region according to the real-time remote sensing data; predict the real-time carbon flux of the monitored sub-region according to the change curve associated with the identification code of the monitored sub-region; Among them, the real-time remote sensing data includes remote sensing images; calculating the real-time drought degree of the monitored sub-region according to the real-time remote sensing data specifically includes: Create a neural network model; Use the neural network model to extract the low-level features of the remote sensing image; Enhance the low-level features using attention mechanisms of different scales; Fuse the features enhanced at all scales to obtain the high-level features of the remote sensing image; Fuse the low-level features and high-level features, and the fused features; Restore the fused features to the size of the remote sensing image to obtain a feature image; Calculate the real-time drought degree of the monitored sub-region according to the feature image.

6. The carbon flux monitoring system based on soil drought remote sensing data according to claim 5, characterized in that The sampling unit is specifically used for: Set weights for rainfall and soil moisture content respectively according to the types of planted crops; Obtain the rainfall from the ground station; Obtain SAR satellite images and analyze the SAR satellite images to obtain the soil moisture content; Multiply the rainfall and soil moisture content by their corresponding weights respectively to obtain the drought degree.

7. The carbon flux monitoring system based on soil drought remote sensing data according to claim 5, characterized in that The real-time remote sensing data includes climate information; the climate information includes temperature, humidity, and sunshine duration; The prediction unit is specifically used for: After denoising and smoothing the climate information, substitute it into a preset vegetation index to obtain the real-time drought degree.

8. The carbon flux monitoring system based on soil drought remote sensing data according to claim 7, wherein It further includes: The display unit is used for: Set up a display interface; Display a map containing the monitored sub-region on the display interface; Set different display colors for different carbon fluxes; Set the color of the monitored sub-region in the map to the display color corresponding to its carbon flux; where the monitored sub-regions with the same display color are connected to each other; When a display color is selected in the map, highlight all the connected monitored sub-regions corresponding to the display color.

Images