Prediction method of field crop biomass in cloud-covered areas using satellite images in rainy season

Abstract

The present invention discloses a method for predicting field crop biomass in cloud-obscured areas using satellite imagery during rainy seasons. The method comprises: obtaining data on key growth periods for field crops within a study area; extracting farmland from satellite imagery and inverting crop biomass data; removing potentially affected farmland, obtaining satellite imagery and crop biomass data for the actual predicted area obscured by clouds and a reference area not obscured by clouds; and comparing these images to obtain a crop biomass distribution map for the predicted date, thereby enabling crop biomass prediction for cloud-obscured areas. By utilizing integrated air-space-ground monitoring technology, the present invention achieves precise inversion of satellite-scale crop biomass data through the multi-dimensional collaboration of satellites, drones, and ground-based sensor networks. This method addresses the lack of satellite-scale data on key crop growth periods due to cloud obscuration, enabling continuous satellite-scale monitoring of crop growth.

Description

Technical Field

[0001] The invention relates to a method for predicting the biomass of field crops, and in particular to a method for predicting the biomass of field crops in a cloud-blocked area of ​​satellite images during rainy seasons. Background Art

[0002] Satellites can cover vast areas and are suitable for monitoring large-scale farmland. Satellite remote sensing enables large-scale analysis and assessment of crop growth conditions. Accurate crop growth forecasts can help optimize agricultural management practices such as irrigation, fertilization, and pest and disease control. This not only increases crop yields but also reduces resource waste and environmental pollution. Optical satellite data (such as Landsat and Sentinel-2) are one of the most commonly used data sources for crop growth monitoring using remote sensing technology. However, this data is susceptible to cloud cover and cloud shadows, especially in cloudy or rainy areas. Cloud obstruction can result in missing or incomplete data, hindering accurate and continuous crop growth forecasts. Therefore, developing methods that can effectively predict crop growth in the presence of cloud obstruction is particularly important. By developing crop growth forecasting methods applicable to cloud-obstructed areas, the accuracy and reliability of forecasts can be significantly improved, enabling farmers, agricultural managers, and policymakers to make informed decisions.

[0003] In recent years, with the development of remote sensing technology, machine learning algorithms, and big data processing capabilities, researchers have begun to explore new methods to solve the problem of cloud obstruction. For example, Synthetic Aperture Radar (SAR) data is not affected by clouds and can provide all-weather observation data. Although SAR data has obvious advantages in solving the problem of cloud obstruction, its geometric distortion, speckle noise, low contrast, polarization complexity, cost, and technical barriers cannot be ignored. For example, compared with optical data, especially in flat and uniform areas, SAR data usually has lower contrast. Low contrast may make it difficult to distinguish different types of land features, thereby affecting crop type identification and growth index estimation.

[0004] In addition, multi-source data fusion and time series analysis methods offer new approaches to addressing this problem. Data from different sensors (such as optical satellites, SAR, drones, and ground-based observation stations) can be combined to provide more comprehensive and accurate information. By fusing data from different sources, the strengths of each data source can be leveraged, overcoming the limitations of a single data source. Time series analysis is a statistical method used to analyze data points that change over time and extract useful information and patterns from them. In agricultural monitoring, time series analysis can help fill data gaps caused by cloud cover and capture dynamic changes during crop growth. However, existing time series analysis methods are primarily used to predict the morphological evolution and motion of remote sensing satellite cloud images, enabling effective monitoring and rapid prediction of catastrophic weather conditions such as thunderstorms, high winds, and heavy rain. These methods are not suitable for crop prediction in farmland because they fail to account for the specific characteristics of farmland crops and the potential for abnormal growth or even death due to precipitation. Summary of the Invention

[0005] To address the problems presented in the previous article, the present invention provides a method for predicting field crop biomass in cloud-obscured areas using satellite imagery during the rainy season. Inspired by multi-source data fusion and time series analysis, the present invention proposes a simple and rapid method for predicting crop biomass in cloud-obscured areas using satellite imagery during the rainy season. This method addresses the challenges posed by climate change and improves the adaptability and resilience of agricultural monitoring systems.

[0006] The technical solution adopted in the present invention is:

[0007] The method for predicting crop biomass in cloud-blocked areas of satellite images during rainy seasons of the present invention comprises:

[0008] S1: Construct a farmland image segmentation model, obtain historical satellite images of the key growth period of field crops in the study area that are not blocked by clouds, and train the farmland image segmentation model to obtain a trained image segmentation model.

[0009] S2: Obtain satellite images, drone images, and ground sampling data for the key growth periods of field crops in the study area. Input the satellite images to be predicted into the trained farmland image segmentation model for processing and output the farmland mask. Process the satellite images to be predicted based on the farmland mask and combined with elevation information to obtain the satellite images to be predicted that only contain the farmland area and its elevation.

[0010] S3: Construct a crop biomass inversion model at different growth stages based on the UAV imagery and ground sampling data in step S2, thereby obtaining crop biomass data by inverting the crop biomass inversion model at different growth stages, the UAV imagery, and the satellite imagery to be predicted that only contains the farmland area and its elevation.

[0011] S4: The date blocked by clouds in the satellite image containing only the farmland area and its elevation is used as the date to be predicted. Based on the precipitation data in the preset time period before the date to be predicted and the elevation data of the farmland area, the farmland that may be affected in the study area is determined and removed to obtain the satellite image of the actual prediction area blocked by clouds and the reference area not blocked by clouds.

[0012] S5: Perform the inversion operation in step S3 on the satellite images of the actual prediction area blocked by clouds within a preset time period before the predicted date and the reference area not blocked by clouds within the predicted date and the preset time period before the predicted date, respectively, to obtain their respective crop biomass data, and then compare them to obtain the crop biomass data of the actual prediction area blocked by clouds on the predicted date, thereby obtaining a crop biomass distribution map for the predicted date and realizing crop biomass prediction in the cloud-blocked area.

[0013] If cloud-free satellite images are obtained for a later date, the predicted crop biomass value of the actual predicted area for the predicted date should be immediately corrected based on the later data.

[0014] The step S1 is specifically as follows:

[0015] S11: Replace the ResNet module in the backbone layer of the Pyramid Attention Network (PAN) with a lightweight MobileNetV1 network to build a farmland image segmentation model.

[0016] S12: For each historical satellite image of the key growth period of field crops in the study area that is not obscured by clouds, it is uniformly cropped into several square cropped satellite images of the same size. The labels of the farmland coverage area and other areas are marked in each cropped satellite image and the images are divided into training and validation sets according to the preset ratio.

[0017] S13: Binary cross entropy loss and Dice coefficient loss are combined into a hybrid loss function to construct a farmland image segmentation model. The training set and validation set are input into the farmland image segmentation model for training until the hybrid loss function converges, thereby obtaining a trained image segmentation model.

[0018] In step S2, the satellite imagery is panchromatic and multispectral imagery from the Gaofen series of satellites; the drone imagery is RGB and multispectral imagery; and the ground sampling data is biomass data for field crops at pre-set sampling points within typical fields within the study area. The key growth period for field crops includes at least two different growth periods, each of which is sorted by phenological phase. All imagery and ground sampling data are stored in a database by date and are referred to as historical dated imagery. Ground sampling data can be collected from typical fields within the study area, and the biomass data specifically refers to the fresh weight of the aboveground portion of wheat.

[0019] In step S2, farmland extraction processing is performed on the satellite image to be predicted based on the farmland mask, and then the average elevation information of each farmland in the satellite image to be predicted is obtained in combination with the satellite digital elevation model (DEM), thereby obtaining a satellite image to be predicted that only contains the farmland area and its elevation.

[0020] In the step S3, first, the various vegetation indices and visible light image texture features at the preset sampling points are obtained according to the drone image in step S2. For each preset sampling point, the various vegetation indices at the preset sampling points and the biomass data of the field crops in the ground sampling data are subjected to correlation regression analysis, and several vegetation indices with correlations higher than a preset threshold are screened out. According to the various screened vegetation indices and visible light image texture features, an integrated learning regression algorithm is used to construct an inversion model for the biomass of crops in different growth periods. The inversion model for the biomass of crops in different growth periods obtains the various different growth periods in the key growth periods of field crops at the drone scale according to the inversion of the drone image. The biomass distribution of field crops in the key growth period is used as the first crop biomass data, and then the first crop biomass data is interpolated and upscaled to match the resolution of the satellite image. The upscaled first crop biomass data is used as the true value of the satellite-scale biomass inversion, and the satellite image to be predicted that only contains the farmland area and its elevation in step S2 and the true value of the satellite-scale biomass inversion are subjected to the same operation as the biomass data of field crops in the UAV image and ground sampling data in step S3, so as to invert the biomass distribution of field crops in different growth periods in the key growth period of field crops at the satellite scale as the final crop biomass data.

[0021] In step S4, precipitation data on the predicted date and the preset time period before it in the study area is obtained, including precipitation amount, precipitation time and precipitation duration, and the disaster value of each farmland in the study area is determined based on the elevation data of the farmland area. The farmland whose disaster value exceeds the disaster threshold value is regarded as the possible disaster-stricken farmland, and the possible disaster-stricken farmland is cropped and removed from the satellite image to be predicted that only contains the farmland area and its elevation. The remaining area in the study area that is blocked by clouds is used as the actual prediction area, and the area not blocked by clouds is used as the reference area.

[0022] The damage value R of the farmland is as follows:

[0023]

[0024] Among them, α1, α2, α3, and α4 are the weights of the first, second, third, and fourth factors, respectively, determined based on historical flooding events; β1 and β2 are the first and second nonlinear exponents, respectively, used to simulate the nonlinear effects of soil permeability and topography on the water accumulation cutoff; P is the cumulative precipitation within the study area during the predicted date and the preset time period before it; T is the duration of precipitation within the study area during the predicted date and the preset time period before it; K is the soil permeability of the farmland within the study area, determined based on the soil type of the farmland; G represents the topography influence coefficient, determined based on the elevation data of the farmland area. The higher the elevation, the lower the topography influence coefficient, and the less susceptible it is to waterlogging. The critical value of the farmland damage value R and the magnitude of each parameter are determined based on historical flooding events. The generated empirical formula is then used to determine the waterlogging status of each plot in the study area.

[0025] In step S5, for each pixel in the satellite image, the crop biomass data obtained from the satellite image of the reference area not blocked by clouds during the preset time period before the predicted date are arranged in order of growth period to obtain a first biomass array A. i [a1, a2, …, a n ], where a n , …, a2 and a1 represent the crop biomass data of the i-th pixel in the satellite image on the to-be-predicted date, …, n-1 days before the to-be-predicted date, and n days before the to-be-predicted date, respectively; the crop biomass data obtained from the satellite image of the actual prediction area blocked by clouds within the preset time period before the to-be-predicted date are arranged in order of growth period to obtain the second biomass array B i[ b1, b2, …, b n-1 ], where b n-1, …, b2 and b1 represent the crop biomass data of the i-th pixel in the satellite image 1 day before the predicted date, …, n-1 days before the predicted date, and n days before the predicted date, respectively; the second biomass array of each pixel in the satellite image is compared with the first n-1 elements in the first biomass array. If the first n-1 elements in one of the first biomass arrays are exactly the same as those in the second biomass array, the n-th element of the first biomass array is used as the n-th element of the second biomass array, that is, as the crop biomass data of the actual predicted area blocked by clouds on the predicted date. The crop biomass data of the actual predicted area blocked by clouds on the predicted date are fused with the actual prediction area to obtain the crop biomass distribution map of the predicted date.

[0026] The beneficial effects of the present invention are:

[0027] The present invention adopts integrated air-space-ground monitoring technology, and realizes the accurate inversion of satellite-scale crop biomass data through the multi-dimensional cooperation of satellites, drones and ground sensor networks. It can solve the problem of lack of satellite-scale data during the critical growth period of crops due to cloud cover, and realize continuous monitoring of crop growth at the satellite scale. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 is a flow chart of the method of the present invention;

[0029] Figure 2 This is an example of satellite imagery during the cloud cover period;

[0030] Figure 3 This is the structure diagram of the farmland image segmentation model;

[0031] Figure 4 It is the label map of satellite image classification results;

[0032] Figure 5 is the extracted farmland area map;

[0033] Figure 6 is the relationship diagram between wheat biomass and vegetation index at seedling and tillering stages, where: Figure 6 (a) is the relationship diagram between a certain day in the seedling stage and the vegetation index. Figure 6 (b) is the relationship diagram between a certain day in the tillering period and the vegetation index;

[0034] Figure 7 This is the inversion result diagram of the first biomass data of wheat in the seedling and tillering stages. Figure 7 (a) is the inversion result of the first biomass data of wheat seedling stage. Figure 7 (b) is the inversion result of the first biomass data of wheat in the tillering stage;

[0035] Figure 8This is a diagram showing the statistical distribution of the AGB estimation accuracy of each level of the Stacking model during the testing phase, where: Figure 8 (a) is the seedling model R 2 Schematic diagram of the statistical distribution of AGB estimation accuracy of various models in the testing phase, Figure 8 (b) is a diagram showing the statistical distribution of the RMSE of the seedling model at the test stage for the AGB estimation accuracy of each model. Figure 8 (c) is the tillering period model R 2 Schematic diagram of the statistical distribution of AGB estimation accuracy of various models in the testing phase, Figure 8 (d) is a schematic diagram of the statistical distribution of the RMSE of the tillering period model and the AGB estimation accuracy of each level of models during the testing phase. DETAILED DESCRIPTION

[0036] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0037] like Figure 1 and Figure 2 As shown, in a specific implementation, the present invention studies the critical growth period of wheat in a certain region in a certain month. The satellite imagery of the study region in that month has high cloud cover, and this period is a critical growth period for crops, such as wheat in the tillering-heading stage. The lack of crop growth monitoring during this period will result in a lack of key data for monitoring this growth cycle. The present invention's method for predicting crop biomass in cloud-obscured areas of satellite imagery during rainy seasons is specifically as follows:

[0038] S1: Construct a farmland image segmentation model. Obtain historical satellite images of the key growth period of field crops in the study area that are not obscured by clouds and train the farmland image segmentation model to obtain a trained image segmentation model. The details are as follows:

[0039] S11: Replace the ResNet module in the backbone layer of the pyramid attention network PAN with a lightweight MobileNetV1 network to construct a farmland image segmentation model.

[0040] S12: For each historical satellite image of field crops in the study area that is not obscured by clouds during their critical growth period, uniformly crop it into several equally sized square cropped satellite images. Label each cropped satellite image with the labels of the farmland-covered areas and other areas, and divide it into training and validation sets according to a preset ratio. Specifically, the original historical satellite image is cropped to a pixel size of 512×512, and then labeled. The field is named "farm" with a pixel value of 1, and other coverage types are labeled "other" with a pixel value of 255. The training, validation, and test sets can be divided into a ratio of 7:2:1 to obtain the dataset used for the semantic segmentation model.

[0041] S13: The binary cross entropy loss and the Dice coefficient loss are combined into a hybrid loss function to construct a farmland image segmentation model. The training set and the validation set are input into the farmland image segmentation model for training until the hybrid loss function converges, thereby obtaining a trained image segmentation model.

[0042] like Figure 3 The results are shown in Table 1. The results are shown in Table 1. The evaluation index in the table is the result of the 50th round of training. It can be seen that the model constructed by PAN using MobileNet2 as the backbone network performs better than the model constructed by using Resnet as the backbone network. In addition, considering the training time and prediction accuracy, the model selected by the present invention is also the best performing model. The trained model is used to classify satellite images. The results are shown in Table 1. Figure 4 As shown. According to the classification results, farmland is extracted from satellite images. The extraction results are shown as follows. Figure 5 The farmland label map is superimposed on the satellite image DEM to extract the average elevation data of each field.

[0043] Table 1

[0044]

[0045]

[0046] S2: Obtain satellite images, drone images, and ground sampling data for the key growth periods of field crops in the study area. Input the satellite images to be predicted into the trained farmland image segmentation model for processing and output the farmland mask. Process the satellite images to be predicted based on the farmland mask and combined with elevation information to obtain the satellite images to be predicted that only contain the farmland area and its elevation.

[0047] Satellite imagery uses the Gaofen series of panchromatic and multispectral images; drone imagery uses RGB and multispectral images. Ground sampling data includes biomass data for field crops at pre-set sampling points within typical fields within the study area. The critical growth period for field crops includes at least two distinct growth periods, each organized by phenological phase. All imagery and ground sampling data are stored in a database categorized by date and referred to as historical dated imagery. Ground sampling data can be collected from selected typical fields within the study area, and biomass data specifically refers to the fresh weight of the aboveground portion of wheat.

[0048] Farmland extraction is performed on the satellite image to be predicted based on the farmland mask, and then the average elevation information of each farmland in the satellite image to be predicted is obtained in combination with the satellite digital elevation model (DEM), thereby obtaining a satellite image to be predicted that only contains the farmland area and its elevation.

[0049] S3: Construct a crop biomass inversion model at different growth stages based on the UAV imagery and ground sampling data in step S2, thereby obtaining crop biomass data by inverting the crop biomass inversion model at different growth stages, the UAV imagery, and the satellite imagery to be predicted that only contains the farmland area and its elevation.

[0050] First, the vegetation index and visible light image texture features at each preset sampling point are obtained according to the drone image in step S2. For each preset sampling point, the vegetation index at the preset sampling point and the biomass data of the field crops in the ground sampling data are subjected to correlation regression analysis, and several vegetation indices with correlations higher than a preset threshold are screened out. According to the screened vegetation index and visible light image texture features, an integrated learning regression algorithm is used to construct a crop biomass inversion model at different growth stages. The integrated learning regression algorithm specifically adopts a Stacking model, and a total of two layers of learners are designed. Among them, the primary learner uses a total of 5 machine learning algorithms, namely Gaussian regression, support vector regression, random forest, multivariate linear regression and Cubist regression. The secondary learner selects a linear ridge regression algorithm. The biomass data set is input into the Stacking model for training according to a five-fold cross method. A total of 200 rounds of training are trained. The data set is re-divided in each round of training to avoid repeated training. The crop biomass inversion model at different growth stages obtains the field crop biomass distribution of each different growth stage in the key growth period of the field crops at the drone scale according to the drone image inversion as the first crop biomass data, such as Figure 7 (a) and Figure 7 As shown in (b), the first biomass data inversion results of wheat seedling stage and wheat tillering stage can be obtained; then the first crop biomass data is interpolated and upscaled to match the resolution of the satellite image, and the upscaled first crop biomass data is used as the true value of the satellite-scale biomass inversion, and the satellite image to be predicted containing only the farmland area and its elevation in step S2 and the true value of the satellite-scale biomass inversion are subjected to the same operation as the biomass data of the field crops in the drone image and ground sampling data in step S3, thereby inverting the field crop biomass distribution of each different growth period in the key growth period of the satellite-scale field crops as the final crop biomass data.

[0051] The spectral vegetation index includes 8 types: normalized difference vegetation index, improved simple ratio vegetation index, green normalized difference vegetation index, green wave wide dynamic vegetation index, improved soil adjustment vegetation index, red edge simple ratio vegetation index, nonlinear vegetation index and structure insensitive pigment index. The texture features of visible light images are 8 types: angular second moment, contrast, homogeneity, correlation, entropy, contrast, cluster prominence and cluster shadow. Figure 6 (a) and Figure 6 (b) shows the normalized vegetation index of wheat at the seedling stage and the tillering stage.

[0052] S4: The date blocked by clouds in the satellite image containing only the farmland area and its elevation is used as the date to be predicted. Based on the precipitation data in the preset time period before the date to be predicted and the elevation data of the farmland area, the farmland that may be affected in the study area is determined and removed to obtain the satellite image of the actual prediction area blocked by clouds and the reference area not blocked by clouds.

[0053] Obtain precipitation data within the study area on the date to be predicted and within the preset time period before it, including precipitation amount, precipitation time and precipitation duration, and at the same time determine the disaster value of each farmland in the study area based on the elevation data of the farmland area. The farmland with a disaster value exceeding the disaster critical value is regarded as the possible disaster-stricken farmland, and the possible disaster-stricken farmland is cropped and removed from the satellite image to be predicted that only contains the farmland area and its elevation. The remaining area blocked by clouds in the study area is used as the actual prediction area, and the area not blocked by clouds is used as the reference area.

[0054] The damage value R of farmland is as follows:

[0055]

[0056] Among them, α1, α2, α3, and α4 are the weights of the first, second, third, and fourth factors, respectively, determined based on historical flooding events; β1 and β2 are the first and second nonlinear exponents, respectively, used to simulate the nonlinear effects of soil permeability and topography on the water accumulation cutoff; P is the cumulative precipitation within the study area during the predicted date and the preset time period before it; T is the duration of precipitation within the study area during the predicted date and the preset time period before it; K is the soil permeability of the farmland within the study area, determined based on the soil type of the farmland; G represents the topography influence coefficient, determined based on the elevation data of the farmland area. The higher the elevation, the lower the topography influence coefficient, and the less susceptible it is to waterlogging. The critical value of the farmland damage value R and the magnitude of each parameter are determined based on historical flooding events. The generated empirical formula is then used to determine the waterlogging status of each plot in the study area.

[0057] S5: Perform the inversion operation in step S3 on the satellite images of the actual prediction area blocked by clouds within a preset time period before the predicted date and the reference area not blocked by clouds within the predicted date and the preset time period before the predicted date, respectively, to obtain their respective crop biomass data, and then compare them to obtain the crop biomass data of the actual prediction area blocked by clouds on the predicted date, thereby obtaining a crop biomass distribution map for the predicted date and realizing crop biomass prediction in the cloud-blocked area.

[0058] For each pixel in the satellite image, the crop biomass data obtained from the satellite image of the reference area not blocked by clouds during the preset time period before the predicted date are arranged in the order of growth period to obtain the first biomass array A. i [a1, a2, …, a n ], where a n , …, a2 and a1 represent the crop biomass data of the i-th pixel in the satellite image on the to-be-predicted date, …, n-1 days before the to-be-predicted date, and n days before the to-be-predicted date, respectively; the crop biomass data obtained from the satellite image of the actual prediction area blocked by clouds within the preset time period before the to-be-predicted date are arranged in order of growth period to obtain the second biomass array B i[ b1, b2, …, b n-1 ], where b n-1 , …, b2 and b1 represent the crop biomass data of the i-th pixel in the satellite image 1 day before the predicted date, …, n-1 days before the predicted date, and n days before the predicted date, respectively; the second biomass array of each pixel in the satellite image is compared with the first n-1 elements in the first biomass array. If the first n-1 elements in one of the first biomass arrays are exactly the same as those in the second biomass array, the n-th element of the first biomass array is used as the n-th element of the second biomass array, that is, as the crop biomass data of the actual predicted area blocked by clouds on the predicted date. The crop biomass data of the actual predicted area blocked by clouds on the predicted date are fused with the actual prediction area to obtain the crop biomass distribution map of the predicted date.

[0059] If cloud-free satellite images are obtained for a later date, the predicted crop biomass value of the actual predicted area for the predicted date should be immediately corrected based on the later data.

[0060] As shown in Table 2, the accuracy evaluation of the first biomass prediction using the Stacking model (the average of 200 rounds of training results) is listed. It can be seen that the performance of the integrated learning model is better than that of a single machine learning algorithm. Figure 8 (a) Figure 8 (b) Figure 8 (c) and Figure 8 As shown in (d), the superior performance of the integrated learning model is more intuitively reflected. The violin shape of the model of the present invention is better than that of a single machine learning model, that is, the model is more stable.

[0061] Table 2

[0062]

[0063] The above description of the embodiments is intended to facilitate understanding and application of the present invention by those skilled in the art. It will be apparent that those skilled in the art can readily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without requiring inventive effort. Therefore, the present invention is not limited to the above embodiments, and improvements and modifications made by those skilled in the art based on the disclosure of the present invention should fall within the scope of protection of the present invention.

Claims

1. A method for predicting crop biomass in cloud-obscured areas using satellite images during rainy seasons, characterized in that: include: S1: Build a farmland image segmentation model, obtain historical satellite images of the key growth period of field crops in the study area that are not blocked by clouds, and train the farmland image segmentation model to obtain a trained image segmentation model; S2: Obtain satellite images, drone images, and ground sampling data for the key growth periods of field crops in the study area. Input the satellite images to be predicted into the trained farmland image segmentation model to process and output a farmland mask. Based on the farmland mask and combined with elevation information, the satellite images to be predicted are processed to obtain a satellite image containing only the farmland area and its elevation. S3: constructing a crop biomass inversion model at different growth stages based on the UAV image and ground sampling data in step S2, thereby obtaining crop biomass data by inverting the crop biomass inversion model at different growth stages, the UAV image, and the satellite image containing only the farmland area and its elevation to be predicted; S4: The date blocked by clouds in the satellite image containing only the farmland area and its elevation is used as the date to be predicted. Based on the precipitation data in the preset time period before the date to be predicted and the elevation data of the farmland area, the potentially affected farmland in the study area is determined and removed to obtain satellite images of the actual predicted area blocked by clouds and the reference area not blocked by clouds. S5: performing the inversion operation in step S3 on satellite images of the actual prediction area obscured by clouds within a preset time period before the predicted date and the reference area not obscured by clouds within the predicted date and the preset time period before the predicted date, respectively, to obtain respective crop biomass data, and then comparing them to obtain the crop biomass data of the actual prediction area obscured by clouds on the predicted date, thereby obtaining a crop biomass distribution map for the predicted date, and realizing crop biomass prediction in the cloud-obscured area; In the step S3, the crop biomass inversion model for different growth periods obtains the biomass distribution of field crops in different growth periods during the key growth period of field crops at the drone scale based on the drone image inversion as the first crop biomass data, and then interpolates the first crop biomass data and upscales it to match the resolution of the satellite image. The upscaled first crop biomass data is used as the true value of the satellite-scale biomass inversion, and the satellite image to be predicted in step S2, which only contains the farmland area and its elevation, and the true value of the satellite-scale biomass inversion are subjected to the same operation as the biomass data of field crops in the drone image and ground sampling data in step S3, thereby inverting the biomass distribution of field crops in different growth periods during the key growth period of field crops at the satellite scale as the final crop biomass data.

2. The method for predicting crop biomass in cloud-blocked areas using satellite images during rainy seasons according to claim 1, wherein: The step S1 is specifically as follows: S11: Replace the ResNet module in the backbone layer of the pyramid attention network PAN with a lightweight MobileNetV1 network to construct a farmland image segmentation model; S12: For each historical satellite image of the key growth period of field crops in the study area that is not obscured by clouds, uniformly crop it into several square cropped satellite images of the same size. Label the farmland coverage area and other areas in each cropped satellite image and divide it into training and validation sets according to the preset ratio; S13: The binary cross entropy loss and the Dice coefficient loss are combined to construct a hybrid loss function for the farmland image segmentation model. The training set and the validation set are input into the farmland image segmentation model for training until the hybrid loss function converges, thereby obtaining a trained image segmentation model.

3. The method for predicting crop biomass in cloud-blocked areas using satellite images during rainy seasons according to claim 1, wherein: In step S2, the satellite images are panchromatic and multispectral images of the Gaofen series satellites; the drone images are visible light RGB and multispectral images; the ground sampling data are the biomass data of field crops at each preset sampling point in the study area; and the key growth period of field crops includes no less than two different growth periods.

4. The method for predicting crop biomass in cloud-blocked areas using satellite images during rainy seasons according to claim 1, wherein: In step S2, farmland extraction processing is performed on the satellite image to be predicted based on the farmland mask, and then the average elevation information of each farmland in the satellite image to be predicted is obtained in combination with the satellite digital elevation model DEM, thereby obtaining a satellite image to be predicted that only contains the farmland area and its elevation.

5. The method for predicting crop biomass in cloud-blocked areas using satellite images during rainy seasons according to claim 3, wherein: In step S3, first, the various vegetation indices and visible light image texture features at each preset sampling point are obtained based on the drone image in step S2. For each preset sampling point, the various vegetation indices at the preset sampling point and the biomass data of field crops in the ground sampling data are subjected to correlation regression analysis, and then several vegetation indices with correlations higher than a preset threshold are screened out. Based on the screened vegetation indices and visible light image texture features, an integrated learning regression algorithm is used to construct an inversion model for the biomass of crops in different growth stages.

6. The method for predicting crop biomass in cloud-blocked areas using satellite images during rainy seasons according to claim 3, wherein: In step S4, precipitation data on the predicted date and the preset time period before it in the study area is obtained, including precipitation amount, precipitation time and precipitation duration, and at the same time, the disaster value of each farmland in the study area is determined based on the elevation data of the farmland area. The farmland with a disaster value exceeding the disaster threshold is regarded as the possible disaster-stricken farmland, and the possible disaster-stricken farmland is cropped and removed from the satellite image to be predicted that only contains the farmland area and its elevation. The remaining area in the study area that is blocked by clouds is used as the actual prediction area, and the area not blocked by clouds is used as the reference area.

7. The method for predicting crop biomass in cloud-blocked areas using satellite images during rainy seasons according to claim 3, wherein: The damage value of the farmland The details are as follows: in, 、 、 and are the first, second, third and fourth factor weights respectively; and are the first and second nonlinear exponents, respectively; The cumulative precipitation in the study area during the predicted date and the preset time period before it; The duration of precipitation in the study area during the predicted date and the preset time period before it; is the soil permeability coefficient of the farmland in the study area; Represents the terrain influence coefficient, which is determined based on the elevation data of the farmland area.

8. The method for predicting crop biomass in cloud-blocked areas using satellite images during rainy seasons according to claim 3, wherein: In step S5, for each pixel in the satellite image, the crop biomass data obtained from the satellite image of the reference area not blocked by clouds during the preset time period before the predicted date are arranged in order of growth period to obtain a first biomass array A. i [a1, a2, …, a n ], where a n , …, a2 and a1 represent the crop biomass data of the i-th pixel in the satellite image on the to-be-predicted date, …, n-1 days before the to-be-predicted date, and n days before the to-be-predicted date, respectively; the crop biomass data obtained from the satellite image of the actual prediction area blocked by clouds within the preset time period before the to-be-predicted date are arranged in order of growth period to obtain the second biomass array B i [b1, b2, …, b n-1 ], where b n-1 , …, b2 and b1 represent the crop biomass data of the i-th pixel in the satellite image 1 day before the predicted date, …, n-1 days before the predicted date, and n days before the predicted date, respectively; the second biomass array of each pixel in the satellite image is compared with the first n-1 elements in the first biomass array. If the first n-1 elements in one of the first biomass arrays are exactly the same as those in the second biomass array, the n-th element of the first biomass array is used as the n-th element of the second biomass array, that is, as the crop biomass data of the actual predicted area blocked by clouds on the predicted date. The crop biomass data of the actual predicted area blocked by clouds on the predicted date are fused with the actual prediction area to obtain the crop biomass distribution map of the predicted date.

Images