Soil moisture data enhancement method and device, medium, equipment and product

Abstract

The invention belongs to the technical field of earth science, and particularly relates to a soil moisture data enhancement method and device, a medium, equipment and a product. The method comprises the following steps: acquiring satellite remote sensing soil moisture data, land assimilation soil moisture data and environment data of a target area within a certain time range; determining a soil moisture reconstruction feature map of the target area according to the satellite remote sensing soil moisture data and the land assimilation soil moisture data within a certain time range; determining a soil moisture multi-source fusion feature map according to the land surface assimilation soil moisture data and the soil moisture reconstruction feature map; and determining enhanced soil moisture data with enhanced resolution according to the soil moisture multi-source fusion feature map and the environmental data. The obtained enhanced soil moisture data and the actually measured soil moisture of the site have high consistency, and time-space continuous and high-resolution soil moisture data support can be provided for related hydro meteorology research.

Description

Technical Field

[0001] The invention belongs to the technical field of earth science, and specifically relates to a method, device, medium, equipment and product for enhancing soil moisture data. Background Art

[0002] Soil moisture inverted from microwave radiation brightness temperature has high accuracy, but its disadvantages are spatiotemporal discontinuity and low spatial resolution, which makes it difficult to meet the needs of fine-grained soil moisture monitoring. Deep learning is a large-scale deep artificial neural network tool that has received widespread attention in recent years. The multi-level network can deeply explore the multi-scale spatial characteristics and potential correlations of various environmental parameters and is widely used to solve various problems in the field of hydrology. However, there is currently a lack of research on using deep learning to invert high-resolution and spatiotemporal continuous soil moisture. Summary of the invention

[0003] In response to the above technical problems, the present invention proposes a method, device, medium, equipment and product for enhancing soil moisture data. The present application obtains satellite remote sensing soil moisture data, land surface assimilated soil moisture data and environmental data of the target area within a certain time range; determines the soil moisture reconstruction characteristic map of the target area according to the satellite remote sensing soil moisture data and the land surface assimilated soil moisture data within a certain time range; determines the soil moisture multi-source fusion characteristic map according to the land surface assimilated soil moisture data and the soil moisture reconstruction characteristic map; determines the enhanced soil moisture data with enhanced resolution according to the soil moisture multi-source fusion characteristic map and the environmental data. The enhanced soil moisture data obtained has a high consistency with the actual soil moisture measured at the site, and can provide spatiotemporally continuous and high-resolution soil moisture data support for related hydrological and meteorological research.

[0004] To solve the above technical problems, the technical solution provided by the present invention includes five aspects.

[0005] In a first aspect, the present application provides a method for enhancing soil moisture data, including: acquiring satellite remote sensing soil moisture data, land surface assimilated soil moisture data and environmental data of a target area within a certain time range; determining a soil moisture reconstruction feature map of the target area based on the satellite remote sensing soil moisture data and the land surface assimilated soil moisture data within a certain time range; determining a soil moisture multi-source fusion feature map based on the land surface assimilated soil moisture data and the soil moisture reconstruction feature map; determining enhanced soil moisture data with enhanced resolution based on the soil moisture multi-source fusion feature map and the environmental data.

[0006] In some embodiments, determining the soil moisture reconstruction feature map of the target area based on the satellite remote sensing soil moisture data and the land surface assimilated soil moisture data within a certain time range includes: extracting effective spatial features of each time phase from the satellite remote sensing soil moisture data through a partial convolutional layer; obtaining the time phase information of each of the effective spatial features; fusing the effective spatial features of each time phase according to the time phase information to obtain a microwave soil moisture continuous feature map of the target area; determining the assimilated soil moisture feature map of the target area based on the land surface assimilated soil moisture data; and determining the soil moisture reconstruction feature map based on the assimilated soil moisture feature map and the microwave soil moisture continuous feature map.

[0007] In some embodiments, determining the soil moisture multi-source fusion feature map based on the land surface assimilated soil moisture data and the soil moisture reconstructed feature map includes: determining the multi-level characteristics of soil moisture in the target area based on the land surface assimilated soil moisture data; determining the multi-scale characteristics of soil moisture in the target area based on the multi-level characteristics of soil moisture; determining the soil moisture multi-source fusion feature map based on the soil moisture multi-scale characteristics and the soil moisture reconstructed feature map.

[0008] In some embodiments, determining the enhanced soil moisture data with resolution enhancement based on the soil moisture multi-source fusion feature map and the environmental data includes: determining the environmental characteristics of the target area based on the environmental data; performing resolution enhancement on the soil moisture multi-source fusion feature map based on the environmental characteristics to obtain a soil moisture resolution enhanced feature map; performing feature enhancement operation on the soil moisture resolution enhanced feature map to obtain enhanced soil moisture data of the target area.

[0009] In some embodiments, determining the soil moisture reconstruction feature map based on the assimilated soil moisture feature map and the microwave soil moisture continuous feature map includes: correcting the assimilated soil moisture feature map according to the microwave soil moisture continuous feature map in a channel attention module to obtain an assimilated soil moisture corrected feature map; enhancing the spatial information of the microwave soil moisture continuous feature map according to the assimilated soil moisture feature map in a spatial attention module to obtain a microwave soil moisture spatially enhanced feature map; and merging the microwave soil moisture spatially enhanced feature map and the assimilated soil moisture corrected feature map through a weighted sum operation to obtain the soil moisture reconstruction feature map.

[0010] In some embodiments, it also includes: the above steps can be implemented in a trained neural network model; the neural network model includes: a missing reconstruction module, a multi-source fusion module and a downscaling module; when training the neural network model, a step-by-step training strategy is used to control the balance between the missing reconstruction module, the multi-source fusion module and the downscaling module.

[0011] In the second aspect, the present application proposes a soil moisture data enhancement device, including: a first acquisition module, used to acquire satellite remote sensing soil moisture data, land surface assimilated soil moisture data and environmental data of a target area within a certain time range; a first determination module, used to determine a soil moisture reconstruction feature map of the target area based on the satellite remote sensing soil moisture data and the land surface assimilated soil moisture data within a certain time range; a second determination module, used to determine a soil moisture multi-source fusion feature map based on the land surface assimilated soil moisture data and the soil moisture reconstruction feature map; a third determination module, used to determine enhanced soil moisture data with enhanced resolution based on the soil moisture multi-source fusion feature map and the environmental data.

[0012] In a third aspect, the present application proposes an electronic device, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of any one of the methods described in the first aspect.

[0013] In a fourth aspect, the present application proposes a readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the methods described in the first aspect.

[0014] In a fifth aspect, the present application proposes a computer program product, comprising a computer program / instructions, which, when executed by a processor, implements the steps of any one of the methods described in the first aspect.

[0015] Beneficial effects created by the present invention: The present application obtains satellite remote sensing soil moisture data, land surface assimilated soil moisture data and environmental data of the target area within a certain time range; determines the soil moisture reconstruction characteristic map of the target area according to the satellite remote sensing soil moisture data and the land surface assimilated soil moisture data within a certain time range; determines the soil moisture multi-source fusion characteristic map according to the land surface assimilated soil moisture data and the soil moisture reconstruction characteristic map; determines the enhanced soil moisture data with enhanced resolution according to the soil moisture multi-source fusion characteristic map and the environmental data. The obtained enhanced soil moisture data has a high consistency with the actual soil moisture measured at the site, and can provide spatiotemporally continuous and high-resolution soil moisture data support for related hydrological and meteorological research. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The scope of the present disclosure may be better understood by reading the following detailed description of exemplary embodiments in conjunction with the accompanying drawings. The drawings included are:

[0017] Figure 1 An overall flow chart of a soil moisture data enhancement method provided in an embodiment of the present application;

[0018] Figure 2 A flowchart of a neural network model for soil moisture data enhancement provided in an embodiment of the present application;

[0019] Figure 3 This is a structural block diagram of a soil moisture data enhancement device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0021] In the following description, reference is made to “some embodiments”, which describe a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict.

[0022] If similar descriptions of "first\second\third" appear in the application documents, the following instructions will be added. In the following description, the terms "first\second\third" involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. It can be understood that "first\second\third" can be interchanged in a specific order or sequence where permitted, so that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0024] Embodiment 1:

[0025] Soil moisture inverted from microwave radiation brightness temperature has high accuracy, but its disadvantages are spatiotemporal discontinuity and low spatial resolution, which makes it difficult to meet the needs of fine-grained soil moisture monitoring. Deep learning is a large-scale deep artificial neural network tool that has received widespread attention in recent years. The multi-level network can deeply explore the multi-scale spatial characteristics and potential correlations of various environmental parameters and is widely used to solve various problems in the field of hydrology. However, there is currently a lack of research on using deep learning to invert high-resolution and spatiotemporal continuous soil moisture.

[0026] In view of the problems existing in the prior art, such as Figure 1 As shown, the present application provides a method for enhancing soil moisture data, and the method is applied to an electronic device, and the electronic device may be a server, a mobile terminal, a computer, a cloud platform, etc. The function implemented by the device data processing provided in the embodiment of the present application can be implemented by calling a program code by a processor of the electronic device, wherein the program code can be stored in a computer storage medium, and the method for enhancing soil moisture data includes:

[0027] Step S1: Obtain satellite remote sensing soil moisture data, land surface assimilated soil moisture data and environmental data of the target area within a certain time range.

[0028] Relevant data of the target area are collected from multiple data sources, including satellite remote sensing soil moisture data, land surface assimilated soil moisture data and environmental data. Satellite remote sensing soil moisture data comes from soil moisture data within a certain time range observed by the Soil Moisture Active and Passive Satellite (SMAP). The land surface assimilated soil moisture data comes from the soil moisture data of the Land Surface Data Assimilation System (CLDAS) of the China Meteorological Administration. The environmental data include: longitude, latitude, altitude, nighttime surface temperature, vegetation index and drought index of the target area.

[0029] Step S2: determining a soil moisture reconstruction characteristic map of the target area according to the satellite remote sensing soil moisture data and the land surface assimilated soil moisture data within a certain time range.

[0030] Due to the existence of factors such as revisit time and seasonal ice age, the soil moisture data inverted by passive microwave remote sensing has serious missing problems, which greatly affects the spatiotemporal continuity of the data. The true situation of soil moisture in the target area cannot be collected in a single acquisition, so it is necessary to collect multiple time phases to form satellite remote sensing soil moisture data of the target area at different times. Therefore, in order to obtain more complete microwave soil moisture data of the target area in this application, it is necessary to obtain satellite remote sensing soil moisture data within a certain time range.

[0031] Therefore, in some embodiments, step S2 "determining the soil moisture reconstruction characteristic map of the target area according to the satellite remote sensing soil moisture data and the land surface assimilated soil moisture data within a certain time range" includes:

[0032] Step S21: extracting effective spatial features of each time phase from the satellite remote sensing soil moisture data through a partial convolutional layer.

[0033] Since the spatial and temporal coverage of satellite remote sensing soil moisture data is low, the direct use of convolutional layers to extract effective spatial features is easily affected by missing areas. Therefore, the present invention first uses partial convolutional layers to process the original input data. The partial convolutional layer automatically updates the mask so that the convolution operation can distinguish between missing and non-missing areas, significantly improving the training efficiency and performance. The extraction expression is as follows:

[0034]

[0035] In the formula, S ( l m,n) Represents the input data of the lth layer at the spatial position (m,n); M represents the mask corresponding to the input data, whose pixel value is 0 or 1, indicating the missing or non-missing point state respectively; 1 is a matrix with all elements of the same shape as M and all 1; W and b are the convolution kernel and the bias term respectively. By performing the operation through this formula, the output value depends entirely on the non-missing area. In addition, some convolutional layers also need to update the mask synchronously. If there is at least one non-missing input value in the convolution operation, the mask at the corresponding position is updated to non-missing. Some convolutional blocks fully take into account the large number of non-missing areas in the input data, which can improve the effectiveness of feature extraction.

[0036] Step S22: using the long short-term memory block to obtain the time phase information of each of the effective spatial features;

[0037] Step S23: The time phase information fuses the effective spatial features of each time phase to obtain a continuous characteristic map of microwave soil moisture in the target area.

[0038] Only using spatial correlation cannot solve the problem of large-scale missing areas in input data. Therefore, the present invention introduces long short-term memory blocks to fill in the missing areas of each other through the effective spatial features of adjacent phase information. The long short-term memory network is the earliest proposed gating algorithm. Each loop module controls the unit state through three gating units, namely the forgetting gate, the input gate and the output gate. After the partial convolution block, the present invention uses convolution-long short-term memory blocks to extract adjacent phase information, further fill in the large-scale missing areas in the input data, and obtain a continuous microwave soil moisture feature map that is continuous in time and space.

[0039] Step S24: determining an assimilated soil moisture characteristic map of the target area according to the land surface assimilated soil moisture data.

[0040] Since there is no missing problem in the assimilated soil moisture data of the land surface, conventional technical means can be directly used in this application to extract the assimilated soil moisture characteristic map.

[0041] Step S25: determining the soil moisture reconstruction characteristic map according to the assimilated soil moisture characteristic map and the microwave soil moisture continuous characteristic map.

[0042] In some embodiments, step S25 “determining the soil moisture reconstruction characteristic map according to the assimilated soil moisture characteristic map and the microwave soil moisture continuous characteristic map” includes:

[0043] Step S251: In the channel attention module, the assimilated soil moisture characteristic map is corrected according to the microwave soil moisture continuous characteristic map to obtain the assimilated soil moisture corrected characteristic map.

[0044] Although the assimilated soil moisture characteristic map has strong spatial continuity, it is obtained through data assimilation, so data deviations may occur in some locations. Therefore, in this application, it is necessary to use the channel attention mechanism combined with the microwave soil moisture continuous characteristic map to correct the assimilated soil moisture characteristic map, thereby obtaining the assimilated soil moisture corrected characteristic map.

[0045] Step S252: In the spatial attention module, the assimilated soil moisture feature map performs spatial information enhancement on the microwave soil moisture continuous feature map to obtain a microwave soil moisture spatially enhanced feature map.

[0046] Since the assimilated soil moisture feature map has good spatial continuity, and although the microwave soil moisture continuous feature map is missing and supplemented by the effective spatial features of the adjacent phases, the supplemented feature map still has some inconsistencies in spatial continuity. Therefore, the spatial attention mechanism is used to spatially enhance the microwave soil moisture continuous feature map in combination with the assimilated soil moisture feature map to obtain the microwave soil moisture spatially enhanced feature map.

[0047] Step S253: merging the microwave soil moisture spatial enhancement feature map and the assimilated soil moisture correction feature map through a weighted sum operation to obtain the soil moisture reconstruction feature map.

[0048] Since the microwave soil moisture spatial enhancement feature map has good data accuracy, and the assimilated soil moisture correction feature map has high spatial continuity, after fusing the two, a more accurate soil moisture reconstruction feature map with high spatial resolution can be obtained. Its expression is as follows:

[0049]

[0050] S CroA =α×S SA +(1-α)×S CA

[0051] In the formula, S SA , S CA and S CroA They represent spatial attention, channel attention, and the output feature map after weighted summation respectively; σ(·) represents the Sigmoid function; W and b represent the convolution kernel and bias term respectively; and are element-wise multiplication and convolution operations respectively; α is the weight, which is determined by the non-missing rate of the same-source spatiotemporal information.

[0052] Step S3: determining a soil moisture multi-source fusion characteristic map according to the land surface assimilated soil moisture data and the soil moisture reconstruction characteristic map.

[0053] Step S2 only reconstructs the missing information of the soil moisture characteristic map. To obtain fully enhanced soil moisture data, the soil moisture reconstruction characteristic map needs to be enhanced in resolution. Therefore, in some embodiments, step S3 "determines the soil moisture multi-source fusion characteristic map based on the land surface assimilated soil moisture data and the soil moisture reconstruction characteristic map" includes:

[0054] Step S31: determining the multi-level characteristics of soil moisture in the target area according to the land surface assimilated soil moisture data.

[0055] Step S32: determining the multi-scale characteristics of soil moisture in the target area according to the multi-level characteristics of soil moisture.

[0056] Step S33: determining the soil moisture multi-source fusion feature map according to the soil moisture multi-scale characteristics and the soil moisture reconstruction feature map.

[0057] Since satellite remote sensing soil moisture data and land surface assimilated soil moisture data have their own advantages in terms of consistency of temporal changes, consistency of spatial distribution and numerical accuracy, both data can reflect the actual situation of soil moisture to a certain extent, but they show significant differences. In order to explore the complex nonlinear correlation implicit in the two data, multi-level features are first extracted from land surface assimilated soil moisture data using residual dense blocks. Then, Laplace attention is used to extract multi-scale features from multi-level features, and then the extracted multi-scale features are fused with the soil moisture reconstruction feature map, so that the fused feature map is initially enhanced in terms of accuracy and resolution.

[0058] Laplace attention consists of a global mean pooling operation, a multi-scale dilated convolutional layer, and a Sigmoid operation, and its expression is as follows:

[0059]

[0060] In the formula, g d represents the result of global mean pooling operation; h and w represent the height and width of the input feature map respectively; S (m,n) Represents the input feature at the spatial position (m,n); S Lap and S input denote the output feature map and input feature map of Laplace attention respectively; D3(·), D5(·), and D7(·) denote convolutional layers with dilation rates of 3, 5, and 7 respectively; and are element-wise multiplication and convolution operations, respectively; σ(·) represents the Sigmoid function; W and b represent the convolution kernel and bias term, respectively.

[0061] Step S4: Determine enhanced soil moisture data with enhanced resolution according to the soil moisture multi-source fusion feature map and the environmental data.

[0062] In this application, the enhanced soil moisture data of the target area is obtained. Although the spatiotemporal continuous soil moisture data is obtained in step S1, its spatial resolution is low and cannot meet the needs of refined monitoring applications. Since the acquired data are all large-scale data, if you want to obtain the refined soil moisture data of the target area, you need to downscale the multi-source fusion feature map.

[0063] Therefore, in some embodiments, step S4 "determining enhanced soil moisture data with enhanced resolution according to the soil moisture multi-source fusion feature map and the environmental data" includes:

[0064] Step S41: determining the environmental characteristics of the target area according to the environmental data.

[0065] In order to further improve the resolution of the soil moisture multi-source fusion feature map in the target area, it is necessary to integrate the environmental characteristics of the target area into the soil moisture multi-source fusion feature map to enhance the detailed information of the soil moisture multi-source feature map in the target area. Therefore, it is necessary to first extract environmental characteristics from environmental data in this application. The environmental data includes: longitude, latitude, altitude, night surface temperature, vegetation index and drought index of the target area. Then, the characteristics of each environmental data, i.e., environmental characteristics, are extracted one by one by conventional means.

[0066] Step S42: performing resolution enhancement on the soil moisture multi-source fusion feature map according to the environmental characteristics to obtain a soil moisture resolution enhanced feature map.

[0067] After obtaining the environmental characteristics, the environmental characteristics are input into the cross-attention module one by one, so that the cross-attention module recalibrates the multi-source fusion feature map of soil moisture according to the environmental characteristics, explores the potential correlation between each environmental data and soil moisture, and effectively enhances the spatial details of soil moisture with the help of high-resolution auxiliary information.

[0068] Step S43: performing a feature enhancement operation on the soil moisture resolution enhancement feature map to obtain enhanced soil moisture data of the target area.

[0069] During the downscaling stage, the dense residual attention block is used to perform feature enhancement operations on the soil moisture resolution enhancement feature map. The dense residual attention block is also called the residual dense convolution block with embedded attention. It consists of a three-layer structure including a residual dense block and a residual attention block. It has been proven to be able to effectively extract high-level information in the feature map and play an important role in the spatial downscaling process. The classic residual dense block has obvious advantages in information extraction and transmission. Therefore, the residual dense block is introduced in the main module to fully extract detail information from various auxiliary factors. In addition, considering that the features from multiple variables are extremely complex, the residual attention is used to recalibrate the feature map to obtain more accurate soil moisture information. It can be specifically expressed as:

[0070]

[0071] In the formula, S input and S RAB Represent the input and output feature maps of the residual attention block, F CA and F SA denote the output feature maps of the channel attention and spatial attention sub-blocks in the dense residual attention block, respectively, and W RAB and b RAB Denote the convolution kernel and bias term of the residual attention block, respectively. In summary, the dense residual attention module extracts and recalibrates multi-level soil moisture magnitude and detail information through alternating residual dense blocks and residual attention blocks, thereby estimating high-resolution soil moisture data.

[0072] In summary, this method can be roughly divided into three steps. The first step is missing reconstruction, from missing to spatiotemporal continuity; the second step of fusion is to initially improve the spatial resolution from 0.36 degrees to 0.06 degrees, and the detailed information comes from the assimilated data of 0.06 degrees; the third step of downscaling is to further improve the spatial resolution from 0.06 degrees to 0.01 degrees, and the detailed information comes from the environmental data of 0.01 degrees. Through the above three steps, the final fine reconstruction results have a high consistency with the measured soil moisture at the site, which can provide spatiotemporal continuity and high-resolution soil moisture data support for related hydrological and meteorological research.

[0073] In some embodiments, the above steps can be implemented by using a trained neural network model. Figure 2 As shown, the neural network model can be roughly divided into "reconstruction module-multi-source fusion module-downscaling module" according to its functions.

[0074] The reconstruction module is mainly used to determine the soil moisture reconstruction feature map of the target area based on the satellite remote sensing soil moisture data and the land surface assimilated soil moisture data within a certain time range. The reconstruction module includes: a partial convolution block, a long short-term memory block, a first cross attention block and a first convolution layer.

[0075] First, extract the effective spatial features of each phase from the satellite remote sensing soil moisture data through partial convolution blocks, then use long short-term memory to extract the phase information of the effective spatial features, reconstruct the missing area through spatiotemporal information, and obtain the microwave soil moisture continuous feature map of the target area. Then obtain the assimilated soil moisture feature map determined according to the land surface assimilated soil moisture data. After obtaining the land surface assimilated soil moisture feature map and the microwave soil moisture continuous feature map, the assimilated soil moisture feature map is corrected according to the microwave soil moisture continuous feature map in the channel attention module of the first cross attention block to obtain the assimilated soil moisture corrected feature map, and the microwave soil moisture continuous feature map is spatially enhanced according to the assimilated soil moisture feature map in the spatial attention module to obtain the microwave soil moisture spatial enhanced feature map; then the microwave soil moisture spatial enhanced feature map and the assimilated soil moisture corrected feature map are merged through weighted sum operation, and then the soil moisture reconstruction feature map is obtained after passing through the convolution layer. The scale of the soil moisture reconstruction feature map is 0.36°, and the scale of the satellite remote sensing soil moisture data is also 0.36°.

[0076] The missing reconstruction process of this application can be expressed as:

[0077]

[0078] Where z represents the variable with a resolution of 0.36°, including the soil moisture reconstruction feature z reco , satellite remote sensing soil moisture data orig and downsampled land surface assimilated soil moisture data z CLDAS ; T represents the time window of the homologous reference data (the previous T days), which is set to 7 days in the present invention; f1(·) represents the reconstructed network.

[0079] The multi-source fusion module is mainly used to determine the soil moisture multi-source fusion feature map according to the land surface assimilated soil moisture data and the soil moisture reconstruction feature map. The multi-source fusion module includes: a dense residual block, a Laplacian attention block, a first upsampling block, a second cross attention block and a second convolution layer.

[0080] The dense residual block is used to determine the multi-level characteristics of soil moisture in the target area based on the land surface assimilated soil moisture data. The Laplace attention block is connected to the dense residual block, and is used to determine the multi-scale characteristics of soil moisture in the target area based on the multi-level characteristics of soil moisture. The first upsampling block is connected to the first convolutional layer and the second cross attention block to obtain the upsampled soil moisture reconstruction feature map. The second cross attention block is used to recalibrate and fuse the soil moisture multi-scale characteristics and the soil moisture reconstruction feature map, and obtain the soil moisture multi-source fusion feature map after the second convolutional layer. The scale of the soil moisture multi-source fusion feature map is 0.06°, and the scale of the land surface assimilated soil moisture data is also 0.06°.

[0081]

[0082] In the formula, y represents the variable with a resolution of 0.06°, including the soil moisture multi-source fusion feature map y fuse Assimilate soil moisture data with land surface CLDAS ; f2(·) represents the fusion network.

[0083] The downscaling module includes: the second upsampling block, the multi-factor cross attention block, the dense residual attention block and the third convolutional layer.

[0084] One end of the second upsampling block is connected to the second convolutional layer, and the other end is connected to the multi-factor cross attention block, so as to obtain the upsampled soil moisture multi-source fusion feature map. The multi-factor cross attention block is used to determine the environmental characteristics of the target area according to the environmental data, and to perform resolution enhancement on the soil moisture multi-source fusion feature map according to the environmental characteristics to obtain a soil moisture resolution enhancement feature map. The dense residual attention block is used to perform feature enhancement on the soil moisture resolution enhancement feature map, and enhanced soil moisture data can be obtained after the third convolutional layer, wherein the scale of the enhanced soil moisture data is 0.01°.

[0085] The multi-source fusion results are spatially downscaled, and this process can be expressed as:

[0086]

[0087] Where x represents the variable with a resolution of 0.01°, including the enhanced soil moisture data x downwith six auxiliary factors: latitude (LAT), longitude (LON), altitude (DEM), land surface temperature at night (LSTN), vegetation index (NDVI) and multi-band drought index (NMDI); f3(·) represents the downscaling network.

[0088] When training the neural network model, a step-by-step training strategy is used to control the balance among the missing reconstruction module, the multi-source fusion module and the downscaling module.

[0089] The network training can be constrained by a multi-task loss function. When the overall loss function converges, the trained network is used for fine reconstruction test of soil moisture, and finally high-resolution spatiotemporal continuous soil moisture data is generated.

[0090] The missing reconstruction loss, multi-source fusion loss and spatial downscaling loss are combined to construct the overall loss function, which is expressed as follows:

[0091]

[0092] Although the multi-task learning framework integrates the functions of missing reconstruction, multi-source fusion and spatial downscaling, the three are in a hierarchical progressive relationship. Only when the loss of the missing reconstruction module is reduced can the multi-source fusion module be guided, and only when the loss of the multi-source fusion module is reduced can the spatial downscaling module be guided. Therefore, the present invention adopts a step-by-step training strategy to control the balance between missing reconstruction, multi-source fusion and spatial downscaling to ensure that the model can complete the three tasks progressively. The step-by-step training strategy is mainly implemented by setting the three coefficients λ1, λ2 and λ3, that is, in the initial stage, a higher λ1 value is set to focus the model training on the missing reconstruction module; in the middle stage, a higher λ2 value is set to focus on training the multi-source fusion module based on more accurate missing reconstruction results; when the training exceeds the specified number of iterations, a higher λ3 value is set to focus on improving the spatial downscaling module with more accurate multi-source fusion results as guidance, thereby obtaining spatiotemporally refined soil moisture.

[0093] This application makes full use of multi-source spatiotemporal and environmental data to achieve continuity and resolution enhancement of soil moisture. The main advantages are as follows: 1) Integrate multiple attention mechanisms and multi-level convolution modules, use land surface assimilation of soil moisture data as a bridge, and gradually enhance the continuity and resolution of soil moisture data; 2) Design multi-task loss functions and step-by-step training strategies to effectively guide the modeling process, making the model more robust.

[0094] Compared with the existing soil moisture data fine reconstruction techniques such as decomposition and random forest based on physical and theoretical scale changes, this model can make full use of the unique advantages of satellite remote sensing soil moisture data and land surface assimilated soil moisture data, while introducing a variety of heterogeneous auxiliary information to achieve high-resolution spatiotemporal continuous soil moisture estimation. Its fine reconstruction results are highly consistent with the measured soil moisture at the site, and can provide spatiotemporal continuous and high-resolution soil moisture data support for related hydrological and meteorological research.

[0095] Embodiment 2:

[0096] Based on the foregoing embodiments, the embodiments of the present application provide a sample acceleration protocol device, and the modules included in the device, as well as the units included in each module, can be implemented by a processor in a computer device; of course, they can also be implemented by a specific logic circuit; in the implementation process, the processor can be a central processing unit (CPU, Central Processing Unit), a microprocessor (MPU, Microprocessor Unit), a digital signal processor (DSP, Digital Signal Processing) or a field programmable gate array (FPGA, Field Programmable Gate Array), etc.

[0097] like Figure 3 As shown, a soil moisture data enhancement device includes: a first acquisition module 1, a first determination module 2, a second determination module 3 and a third determination module 4.

[0098] The first acquisition module 1 is used to obtain satellite remote sensing soil moisture data, land surface assimilated soil moisture data and environmental data of the target area within a certain time range. The first determination module 2 is used to determine the soil moisture reconstruction feature map of the target area based on the satellite remote sensing soil moisture data and the land surface assimilated soil moisture data within a certain time range. The second determination module 3 is used to determine the soil moisture multi-source fusion feature map based on the land surface assimilated soil moisture data and the soil moisture reconstruction feature map. The third determination module 4 is used to determine the enhanced soil moisture data with enhanced resolution based on the soil moisture multi-source fusion feature map and the environmental data.

[0099] Each module in the above-mentioned soil moisture data enhancement device can be implemented in whole or in part by software, hardware and a combination thereof. Each of the above-mentioned modules can be embedded in or independent of the processor in the device in the form of hardware, or can be stored in the memory in the processing device in the form of software, so that the processor can call and execute the operations corresponding to each of the above modules. It should be noted that the division of modules in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0100] Embodiment 3:

[0101] In a third aspect, a computer device is provided, comprising a memory, a processor, and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of any one of the methods described in the first aspect.

[0102] Embodiment 4:

[0103] In a fourth aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the methods described in the first aspect.

[0104] Embodiment 5:

[0105] In a fifth aspect, the present application provides a computer program product, comprising a computer program / instructions, characterized in that when the computer program is executed by a processor, the steps of any one of the methods described in the first aspect are implemented.

[0106] Those of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0107] It should be understood that "one embodiment" or "an embodiment" mentioned throughout the specification means that specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" appearing throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application. The above-mentioned sequence numbers of the embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0108] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0109] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0110] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units. They may be located in one place or distributed over multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0111] In addition, all functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit. The above integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0112] Those skilled in the art can understand that all or part of the steps of the above method embodiments can be completed by hardware related to program instructions, and the above program can be stored in a computer-readable storage medium, and when the program is executed, the steps of the above method embodiments are executed. The above storage medium includes: a mobile storage device, a read-only memory (ROM), a magnetic disk or an optical disk, and other media that can store program codes.

[0113] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application can essentially or in other words, the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium and includes a number of instructions for a controller to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0114] The above is only an implementation method of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for enhancing soil moisture data, characterized in that: include: Obtain satellite remote sensing soil moisture data, land surface assimilated soil moisture data and environmental data of the target area within a certain time range; Determine a soil moisture reconstruction characteristic map of the target area based on the satellite remote sensing soil moisture data and the land surface assimilated soil moisture data within a certain time range; Determine a soil moisture multi-source fusion characteristic map according to the land surface assimilated soil moisture data and the soil moisture reconstruction characteristic map; Enhanced soil moisture data with enhanced resolution is determined according to the soil moisture multi-source fusion feature map and the environmental data.

2. The method according to claim 1, characterized in that The step of determining the soil moisture reconstruction characteristic map of the target area according to the satellite remote sensing soil moisture data and the land surface assimilated soil moisture data within a certain time range includes: Extracting effective spatial features of each time phase from the satellite remote sensing soil moisture data through a partial convolutional layer; Acquiring the time phase information of each of the effective spatial features; The effective spatial features of each time phase are merged according to the time phase information to obtain a continuous characteristic map of microwave soil moisture in the target area; Determine an assimilated soil moisture characteristic map of the target area according to the land surface assimilated soil moisture data; The soil moisture reconstruction characteristic map is determined according to the assimilated soil moisture characteristic map and the microwave soil moisture continuous characteristic map.

3. The method according to claim 1, characterized in that The determining of the soil moisture multi-source fusion characteristic map according to the land surface assimilated soil moisture data and the soil moisture reconstruction characteristic map comprises: Determine the multi-level characteristics of soil moisture in the target area according to the land surface assimilated soil moisture data; Determine the multi-scale characteristics of soil moisture in the target area according to the multi-level characteristics of soil moisture; The soil moisture multi-source fusion characteristic map is determined according to the soil moisture multi-scale characteristics and the soil moisture reconstruction characteristic map.

4. The method according to claim 1, characterized in that: The step of determining the enhanced soil moisture data with enhanced resolution according to the soil moisture multi-source fusion feature map and the environmental data includes: determining environmental characteristics of the target area according to the environmental data; Performing resolution enhancement on the soil moisture multi-source fusion feature map according to the environmental characteristics to obtain a soil moisture resolution enhancement feature map; A feature enhancement operation is performed on the soil moisture resolution enhancement feature map to obtain enhanced soil moisture data of the target area.

5. The method according to claim 2, characterized in that: The determining the soil moisture reconstruction characteristic map according to the assimilated soil moisture characteristic map and the microwave soil moisture continuous characteristic map comprises: In a channel attention module, the assimilated soil moisture characteristic map is corrected according to the microwave soil moisture continuous characteristic map to obtain an assimilated soil moisture corrected characteristic map; In a spatial attention module, spatial information enhancement is performed on the microwave soil moisture continuous feature map according to the assimilated soil moisture feature map to obtain a microwave soil moisture spatial enhancement feature map; The microwave soil moisture spatial enhancement feature map and the assimilated soil moisture correction feature map are combined through a weighted sum operation to obtain the soil moisture reconstruction feature map.

6. The method according to any one of claims 1 to 5, characterized in that: Also includes: The above steps can be implemented in the trained neural network model; The neural network model includes: a missing reconstruction module, a multi-source fusion module and a downscaling module; When training the neural network model, a step-by-step training strategy is used to control the balance among the missing reconstruction module, the multi-source fusion module and the downscaling module.

7. A soil moisture data enhancement device, characterized in that: include: The first acquisition module is used to obtain satellite remote sensing soil moisture data, land surface assimilated soil moisture data and environmental data of the target area within a certain time range; A first determination module is used to determine a soil moisture reconstruction characteristic map of the target area based on the satellite remote sensing soil moisture data and the land surface assimilated soil moisture data within a certain time range; A second determination module is used to determine a soil moisture multi-source fusion characteristic map according to the land surface assimilated soil moisture data and the soil moisture reconstruction characteristic map; The third determination module is used to determine the enhanced soil moisture data with enhanced resolution according to the soil moisture multi-source fusion feature map and the environmental data.

8. An electronic device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method according to any one of claims 1 to 6.

9. A readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images