A method for monitoring rice irrigation patterns based on satellite remote sensing technology

Abstract

This invention discloses a method for monitoring rice irrigation patterns based on satellite remote sensing technology. It uses multispectral remote sensing image data to estimate the soil moisture content of rice planting areas and uses high-resolution remote sensing satellite images to identify and divide high-standard farmland areas in rice planting areas. This achieves the goal of global monitoring of each high-standard farmland in rice planting areas during the rice planting stage, thereby quickly, accurately and unbiasedly calculating the irrigation pattern of the entire rice planting area during the entire rice planting stage.

Description

Technical Field

[0001] This invention relates to the field of agricultural monitoring technology, specifically a method for monitoring rice irrigation patterns based on satellite remote sensing technology. Background Technology

[0002] Rice is a vital food crop for humankind, consumed by half the world's population. China, as a major rice-producing country, ranks first globally in rice production. Rice paddy methane is a significant component of carbon emissions from agriculture, accounting for 10% of global methane emissions. Rice methane emissions are crucial for achieving carbon peaking and carbon neutrality goals. Appropriate irrigation patterns can effectively reduce rice methane emissions. Currently, research on rice irrigation patterns relies on field sampling, which cannot provide comprehensive monitoring of rice-growing areas. This approach only allows for point-to-area analysis, making it difficult to guarantee the accuracy and unbiasedness of the samples. Furthermore, the time and cost of field sampling are high, and it's impossible to obtain an overall picture of the entire rice growing stage; statistical analysis is primarily conducted through interviews. Summary of the Invention

[0003] The technical problem to be solved by this invention is to provide a method for monitoring rice irrigation patterns based on satellite remote sensing technology. By using multispectral remote sensing image data and high-resolution remote sensing satellite image data, the soil moisture content of rice planting areas is calculated and analyzed regionally, so as to achieve the purpose of global monitoring of rice planting areas during the rice planting stage, thereby quickly, accurately and unbiasedly statistically analyzing the irrigation patterns of rice planting areas throughout the entire rice planting stage.

[0004] The technical solution of this invention is as follows: A method for monitoring rice irrigation patterns based on satellite remote sensing technology includes the following steps: (1) Multispectral remote sensing image data and high-resolution remote sensing satellite image data of rice planting areas are obtained through multispectral remote sensing images and high-resolution remote sensing satellite images of rice planting areas; (2) Construct a time series of multispectral remote sensing image data of rice planting area, and calculate the soil moisture content and rice growth status of rice planting area during the rice planting stage according to the SMI and EVI index calculation formulas respectively; the calculation formula of SMI index is shown in formula (1), and the calculation formula of EVI index is shown in formula (2): (1); In formula (1), reflection 1600nm The reflectance is the satellite reflectance at 1600 nm obtained from multispectral remote sensing image data. 2100nm The value of the SMI index is the satellite reflectance at 2100 nm, obtained from multispectral remote sensing image data; the smaller the calculated value of the SMI index, the higher the soil moisture content. EVI = 2.5 × (reflectance) 840nm - reflectance 650nm ) / ( reflectance 840nm +6×reflectance 650nm - 7.5×reflectance 480nm +1) (2); In formula (2), reflection 840nm The reflectance at 840 nm is the satellite reflectance obtained from multispectral remote sensing image data. 650nm The reflectance at 650 nm is the satellite reflectance obtained from multispectral remote sensing image data. 480nm The value of EVI is the satellite reflectance at 480nm obtained from multispectral remote sensing image data; the higher the calculated value of the EVI index, the more vigorous the rice growth. (3) High-standard farmland boundaries are identified using a deep residual network with U-Net architecture for high-resolution remote sensing satellite image data of rice planting areas, thereby dividing high-standard farmland areas in high-resolution remote sensing satellite images. (4) Based on the SMI index of the rice planting area during the rice planting stage and the division of high-standard farmland in high-resolution remote sensing satellite images, the irrigation patterns of high-standard farmland in the rice planting area were analyzed. The specific analysis process is as follows: a. When the SMI index during the rice planting stage is less than 0.14, the high-standard farmland is judged to be in a state of flooding. b. When the SMI index is greater than 0.16 during the rice planting stage, the high-standard farmland is judged to be in a drying state. c. When the SMI index during the rice planting stage is between 0.14 and 0.16, it is determined that the high-standard farmland is in a state of drainage or irrigation. d. Based on the SMI index of all high-standard farmland in the rice planting area during the rice planting stage, establish the probability distribution of the SMI index during the rice planting stage. Compare the probability distribution of the SMI index of all high-standard farmland during the rice planting stage with the probability distribution of the SMI index under the standard intermittent irrigation mode scenario to obtain the probability that the rice planting stage of high-standard farmland meets the standard intermittent irrigation mode scenario. e. Based on the SMI index value of each high-standard farmland in the rice planting area during the rice planting stage, determine the irrigation mode of each high-standard farmland during the rice planting stage. Based on the EVI index value of each high-standard farmland in the rice planting area during the rice planting stage, determine the rice growth status of each high-standard farmland during the rice planting stage. Combine the analysis of the irrigation mode and rice growth status of each high-standard farmland in the rice planting area to determine the rice growth status of each high-standard farmland under different irrigation modes during the rice planting stage.

[0005] The multispectral remote sensing images and high-resolution remote sensing satellite images mentioned above are all subjected to radiometric correction, geometric correction and atmospheric correction using remote sensing image processing tools, and clouds and snow are removed through red, green and blue channels to obtain corrected multispectral remote sensing images and high-resolution remote sensing satellite images. Then, multispectral remote sensing image data and high-resolution remote sensing satellite image data of rice planting areas are obtained from the corrected multispectral remote sensing images and high-resolution remote sensing satellite images.

[0006] The residual formula for the deep residual network of the U-Net architecture is formula (3): Y1=f resnet (f pool (f conv (X inp )))=f m1 (X1), Y2=f resnet (f deconv (f resnet (X inp )))=f m2 (X2), Output = sigmoid(f m2 (f m2 (f m2 (f m1 (f m1 (f m1 (Input))))))) ; (3) In formula (3), Input is the RGB three-channel data of high-resolution remote sensing satellite imagery, and Output is the model output, X inp For input data, f conv For convolutional layers, f pool For pooling layers, f resnet For the residual layer, f deconv For a deconvolutional layer, X1 is the residual formula f m1 The input data, i.e., X1, is equivalent to the residual formula f. m1 X in inp Y1 is the residual formula f m1 The output data, X2 is the residual formula fm2 The input data, i.e., X2, is equivalent to the residual formula f. m2 X in inp Y2 is the residual formula f m2 The output data is sigmoid, which is the activation function layer.

[0007] The activation function of the sigmoid is f(x) = 1 / (1+e^x). -x The loss function of the deep residual network in the U-Net architecture is Loss, which is Loss = 1 / 2[1 - 2 × |X|Y| / (|X| + |Y|)] + 1 / 2[1 - 2 × |X|Y|Y] outline ∣ / (∣X∩Y outline ∣+∣Y∩Y outline ∣)], where ∣X∩Y∣ is the intersection between the model output X and the label Y, where label Y is the high-standard farmland area identified visually, and Y outline The bounded area is the boundary region of the high-standard farmland in label Y.

[0008] The soil moisture content data of the rice planting area under natural conditions was obtained by simulation using the WRF-CHEM model. Specifically, meteorological satellite data of the rice planting area was obtained through meteorological satellite cloud images. Based on the meteorological satellite data of the rice planting area, precipitation data and soil moisture content data under natural conditions were simulated using the WRF-CHEM model. When the soil moisture content corresponding to the SMI index during the rice planting stage exceeds 10% of the soil moisture content under natural conditions simulated by the WRF-CHEM model, it is determined that the high-standard farmland has been irrigated. When the SMI index is less than 0.14, it is determined that the high-standard farmland is in a state of irrigation flooding.

[0009] The specific method for obtaining precipitation data and soil moisture content data under natural conditions using the WRF-CHEM model is as follows: Meteorological data from the U.S. National Center for Environmental Prediction (NCEPA) at 0.25°×0.25° is used, and a three-layer nested subgrid high-resolution simulation is performed to obtain soil moisture content data at a resolution of 1 km. The parameters for the three-layer nested subgrid high-resolution simulation are set as follows: the initial boundary field uses NCEPA meteorological data at 0.25°×0.25°; the resolution of the three-layer nesting is 0.15°×0.15°, 0.05°×0.05°, and 0.01°×0.01°; the WRF parameters are set as follows: microphysical process scheme: WSM 6class; longwave radiation scheme: RRTM; shortwave radiation scheme: Dudhia; land surface process scheme: Noah; boundary layer scheme: YSU; cumulus parameterization scheme: Kain-Fritsch.

[0010] The process of establishing the probability distribution of the SMI index of all high-standard farmland in the rice planting area during the rice planting stage is as follows: the SMI index of all high-standard farmland in the rice planting area during the rice planting stage is calculated with a time resolution of 3 days and a spatial resolution of 10m, thereby obtaining the SMI index of each high-standard farmland in the rice planting area during the rice planting stage, and plotting the probability distribution of the SMI index of each high-standard farmland.

[0011] The probability distribution of the SMI index under the standard intermittent irrigation mode scenario is obtained by analyzing historical soil moisture content data of the current rice planting area under the intermittent irrigation mode scenario.

[0012] The probability distribution of intermittent irrigation under the standard intermittent irrigation mode scenario is as follows: P=[N(0.13,0.02)+ N(0.16,0.02)] / 2, where N is a normal distribution, 0.13 is the mean of the normal distribution of irrigation flooding under the standard intermittent irrigation mode scenario, 0.16 is the mean of the normal distribution of field drying under the standard intermittent irrigation mode scenario, and 0.02 is the standard deviation of the two normal distributions.

[0013] Advantages of this invention: (1) This invention uses multispectral remote sensing images to obtain satellite reflectance data of rice planting areas during the rice planting stage, and then estimates the soil moisture content and rice growth status of rice planting areas during the rice planting stage through SMI and EVI index calculation formulas. The irrigation mode of rice planting areas is derived through soil moisture content. The estimation and derivation results are highly accurate, and the purpose of comprehensive monitoring of irrigation mode of rice planting areas during the rice planting stage is achieved. (2) This invention utilizes high-resolution remote sensing satellite images of rice planting areas and adopts a deep residual network with U-Net architecture to identify the boundaries of high-standard farmland. This allows for the statistical analysis of irrigation patterns for each identified high-standard farmland, and the division of areas for statistical analysis, which greatly improves the authenticity and accuracy of rice irrigation pattern monitoring. (3) Based on meteorological satellite data of rice planting areas, the present invention uses the WRF-CHEM model to simulate soil moisture content data under natural conditions, and then compares the real-time monitored and estimated soil moisture content data with the soil moisture content data under natural conditions to confirm whether the soil has been irrigated, and confirms whether the soil is in an irrigated flooded state based on the estimated SMI index, eliminating the flooded situation under natural conditions, and further improving the accuracy of rice irrigation mode monitoring. Attached Figure Description

[0014] Figure 1 This is a flowchart of the present invention.

[0015] Figure 2 This is a probability distribution diagram of the SMI index during the rice planting stage in an embodiment of the present invention. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] See Figure 1 A method for monitoring rice irrigation patterns based on satellite remote sensing technology, specifically including the following steps: (1) Multispectral remote sensing image data (using Sentinel-2 data, with a spatial resolution resampling of 10 meters), high-resolution remote sensing satellite image data (spatial resolution less than 2 meters), and meteorological satellite data of rice planting areas were obtained through multispectral remote sensing images, high-resolution remote sensing satellite images, and meteorological satellite cloud images of rice planting areas. Among them, the multispectral remote sensing images and high-resolution remote sensing satellite images were all radiometrically corrected, geometrically corrected, and atmospherically corrected using remote sensing image processing tools (ENVI5.4 software). After removing clouds and snow through the red, green, and blue channels, the multispectral remote sensing image data and high-resolution remote sensing satellite image data were obtained. The specific method for removing clouds and snow through the red, green, and blue channels was to normalize the red, green, and blue channels to the 0-1 range and remove pixel values ​​that were greater than 0.75 in both red, green, and blue channels. (2) Construct a time series of multispectral remote sensing image data of rice planting area, and calculate the soil moisture content and rice growth status of rice planting area during the rice planting stage according to the SMI and EVI index calculation formulas respectively; the calculation formula of SMI index is shown in formula (1), and the calculation formula of EVI index is shown in formula (2): (1); In formula (1), reflection 1600nm The reflectance is the satellite reflectance at 1600 nm obtained from multispectral remote sensing image data. 2100nm The value of the SMI index is the satellite reflectance at 2100 nm, obtained from multispectral remote sensing image data; the smaller the calculated value of the SMI index, the higher the soil moisture content. EVI = 2.5 × (reflectance) 840nm - reflectance 650nm ) / ( reflectance 840nm +6×reflectance650nm - 7.5×reflectance 480nm +1) (2); In formula (2), reflection 840nm The reflectance at 840 nm is the satellite reflectance obtained from multispectral remote sensing image data. 650nm The reflectance at 650 nm is the satellite reflectance obtained from multispectral remote sensing image data. 480nm The satellite reflectance at 480nm is obtained from multispectral remote sensing image data; an EVI index of 0.25 is used as the starting threshold for rice growth. The higher the calculated EVI index value, the more vigorous the rice growth. (3) High-standard farmland boundaries are identified using a deep residual network with U-Net architecture for high-resolution remote sensing satellite image data of rice planting areas, thereby dividing high-standard farmland areas in high-resolution remote sensing satellite images. Among them, the residual formula of the deep residual network of the U-Net architecture is formula (3): Y1=f resnet (f pool (f conv (X inp )))=f m1 (X1), Y2=f resnet (f deconv (f resnet (X inp )))=f m2 (X2), Output = sigmoid(f m2 (f m2 (f m2 (f m1 (f m1 (f m1 (Input))))))); (3) In formula (3), Input is the RGB three-channel data of high-resolution remote sensing satellite imagery, and Output is the model output, X inp For input data, f conv For convolutional layers, f pool For pooling layers, f resnet For the residual layer, f deconv For a deconvolutional layer, X1 is the residual formula f m1 The input data, i.e., X1, is equivalent to the residual formula f. m1 X in inp Y1 is the residual formula f m1 The output data, X2 is the residual formula fm2 The input data, i.e., X2, is equivalent to the residual formula f. m2 X in inp Y2 is the residual formula f m2 The output data, sigmoid is the activation function layer, and the activation function of sigmoid is f(x) = 1 / (1+e^(-x)). -x The loss function of the deep residual network in the U-Net architecture is Loss, which is Loss = 1 / 2[1 - 2 × |X|Y| / (|X| + |Y|)] + 1 / 2[1 - 2 × |X|Y|Y] outline ∣ / (∣X∩Y outline ∣+∣Y∩Y outline ∣)], where ∣X∩Y∣ is the intersection between the model output X and the label Y, where label Y is the high-standard farmland area identified visually, and Y outline The bounded area representing the boundary of high-standard farmland in label Y; (4) Based on meteorological satellite data of the rice planting area, precipitation data and soil moisture content data under natural conditions were obtained by simulation using the WRF-CHEM model. The specific method is as follows: meteorological data of 0.25°×0.25° from the National Center for Environmental Prediction (NCEP) of the United States were used, and a three-layer nested subgrid high-resolution simulation was performed to obtain soil moisture content data with a resolution of 1 km. The parameters for the three-layer nested subgrid high-resolution simulation were set as follows: the initial boundary field used meteorological data of 0.25°×0.25° from the NCEP of the United States, the resolution of the three-layer nesting was 0.15°×0.15°, 0.05°×0.05°, and 0.01°×0.01°, and the WRF parameters were set as follows: microphysical process scheme: WSM 6class; longwave radiation scheme: RRTM; shortwave radiation scheme: Dudhia; land surface process scheme: Noah; boundary layer scheme: YSU; cumulus parameterization scheme: Kain-Fritsch. (5) Based on soil moisture content data under natural conditions, the SMI index of all high-standard farmland in the rice planting area during the rice planting stage, and the division of high-standard farmland in high-resolution remote sensing satellite images, the irrigation patterns of high-standard farmland in the rice planting area were analyzed; the specific analysis process is as follows: a. When the soil moisture content corresponding to the SMI index during the rice planting stage exceeds 10% of the soil moisture content under natural conditions (values ​​range from 0 to 100%) simulated by the WRF-CHEM model, it is determined that the high-standard farmland has been irrigated. When the SMI index is less than 0.14, it is determined that the high-standard farmland is in a state of irrigation flooding. b. When the SMI index is greater than 0.16 during the rice planting stage, the high-standard farmland is judged to be in a drying state. c. When the SMI index during the rice planting stage is between 0.14 and 0.16, it is determined that the high-standard farmland is in a state of drainage or irrigation. d. Irrigation does not occur simultaneously in all high-standard farmland areas within the rice-growing region. Irrigation of different high-standard farmland plots varies over time. Therefore, it is necessary to obtain the SMI index data for all high-standard farmland during the rice-growing stage and plot its probability distribution. The SMI index of all high-standard farmland in the rice-growing region during the rice-growing stage is calculated with a temporal resolution of every 3 days (i.e., the SMI index is calculated every 3 days based on real-time satellite reflectivity) and a spatial resolution of 10m. This yields the SMI index for each high-standard farmland during the rice-growing stage, and the probability distribution of the SMI index for each high-standard farmland is plotted (see...). Figure 2 The frequency (number of times the SMI index is calculated during the rice planting stage) is used to compare the probability distribution of the SMI index of each high-standard farmland during the rice planting stage with the probability distribution of the SMI index under the standard intermittent irrigation mode scenario, so as to obtain the probability (confidence interval) that each high-standard farmland during the rice planting stage meets the standard intermittent irrigation mode scenario. Among them, the probability distribution of the SMI index under the standard intermittent irrigation mode scenario is obtained by analyzing the historical soil moisture content data of the current rice planting area under the intermittent irrigation mode scenario. The probability distribution of intermittent irrigation under the standard intermittent irrigation mode scenario is as follows: P=[N(0.13,0.02)+ N(0.16,0.02)] / 2, where N is a normal distribution, 0.13 is the mean of the normal distribution of the irrigation flooding state under the standard intermittent irrigation mode scenario, 0.16 is the mean of the normal distribution of the field drying state under the standard intermittent irrigation mode scenario, and 0.02 is the standard deviation of the two normal distributions. e. Based on the SMI index value of each high-standard farmland in the rice planting area during the rice planting stage, determine the irrigation mode of each high-standard farmland during the rice planting stage. Based on the EVI index value of each high-standard farmland in the rice planting area during the rice planting stage, determine the rice growth status of each high-standard farmland during the rice planting stage. Combine the analysis of the irrigation mode and rice growth status of each high-standard farmland in the rice planting area to determine the rice growth status of each high-standard farmland under different irrigation modes during the rice planting stage.

[0018] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring rice irrigation patterns based on satellite remote sensing technology, characterized in that: Specifically, it includes the following steps: (1) Multispectral remote sensing image data and high-resolution remote sensing satellite image data of rice planting areas are obtained through multispectral remote sensing images and high-resolution remote sensing satellite images of rice planting areas; (2) Construct a time series of multispectral remote sensing image data of rice planting area, and calculate the soil moisture content and rice growth status of rice planting area during the rice planting stage according to the SMI and EVI index calculation formulas respectively; the calculation formula of SMI index is shown in formula (1), and the calculation formula of EVI index is shown in formula (2): （1）； In formula (1), reflectance1600nm is the satellite reflectance at 1600nm obtained from multispectral remote sensing image data, and reflectance2100nm is the satellite reflectance at 2100nm obtained from multispectral remote sensing image data; the smaller the calculated value of the SMI index, the higher the soil moisture content. （2）； In formula (2), The satellite reflectance at 840 nm is obtained from multispectral remote sensing image data. The satellite reflectance at 650 nm is obtained from multispectral remote sensing image data. The value of EVI is the satellite reflectance at 480nm obtained from multispectral remote sensing image data; the higher the calculated value of the EVI index, the more vigorous the rice growth. (3) High-standard farmland boundaries are identified using a deep residual network with U-Net architecture for high-resolution remote sensing satellite image data of rice planting areas, thereby dividing high-standard farmland areas in high-resolution remote sensing satellite images. (4) Based on the SMI index of the rice planting area during the rice planting stage and the division of high-standard farmland in high-resolution remote sensing satellite images, the irrigation patterns of high-standard farmland in the rice planting area were analyzed. The specific analysis process is as follows: a. When the SMI index during the rice planting stage is less than 0.14, the high-standard farmland is judged to be in a state of flooding. b. When the SMI index is greater than 0.16 during the rice planting stage, the high-standard farmland is judged to be in a drying state. c. When the SMI index during the rice planting stage is between 0.14 and 0.16, it is determined that the high-standard farmland is in a state of drainage or irrigation. d. Based on the SMI index of all high-standard farmland in the rice planting area during the rice planting stage, establish the probability distribution of the SMI index during the rice planting stage. Compare the probability distribution of the SMI index of all high-standard farmland during the rice planting stage with the probability distribution of the SMI index under the standard intermittent irrigation mode scenario to obtain the probability that the rice planting stage of high-standard farmland meets the standard intermittent irrigation mode scenario. e. Based on the SMI index value of each high-standard farmland in the rice planting area during the rice planting stage, determine the irrigation mode of each high-standard farmland during the rice planting stage. Based on the EVI index value of each high-standard farmland in the rice planting area during the rice planting stage, determine the rice growth status of each high-standard farmland during the rice planting stage. Combine the analysis of the irrigation mode and rice growth status of each high-standard farmland in the rice planting area to determine the rice growth status of each high-standard farmland under different irrigation modes during the rice planting stage.

2. The method for monitoring rice irrigation patterns based on satellite remote sensing technology according to claim 1, characterized in that: The multispectral remote sensing images and high-resolution remote sensing satellite images mentioned above are all subjected to radiometric correction, geometric correction and atmospheric correction using remote sensing image processing tools, and clouds and snow are removed through red, green and blue channels to obtain corrected multispectral remote sensing images and high-resolution remote sensing satellite images. Then, multispectral remote sensing image data and high-resolution remote sensing satellite image data of rice planting areas are obtained from the corrected multispectral remote sensing images and high-resolution remote sensing satellite images.

3. The method for monitoring rice irrigation patterns based on satellite remote sensing technology according to claim 1, characterized in that: The soil moisture content data of the rice planting area under natural conditions was obtained by simulation using the WRF-CHEM model. Specifically, meteorological satellite data of the rice planting area was obtained through meteorological satellite cloud images. Based on the meteorological satellite data of the rice planting area, precipitation data and soil moisture content data under natural conditions were simulated using the WRF-CHEM model. When the soil moisture content corresponding to the SMI index during the rice planting stage exceeds 10% of the soil moisture content under natural conditions simulated by the WRF-CHEM model, it is determined that the high-standard farmland has been irrigated. When the SMI index is less than 0.14, it is determined that the high-standard farmland is in a state of irrigation flooding.

4. The method for monitoring rice irrigation patterns based on satellite remote sensing technology according to claim 3, characterized in that: The specific method for obtaining precipitation data and soil moisture content data under natural conditions using the WRF-CHEM model is as follows: Meteorological data from the U.S. National Center for Environmental Prediction (NCEPA) at 0.25°×0.25° is used, and a three-layer nested subgrid high-resolution simulation is performed to obtain soil moisture content data at a resolution of 1 km. The parameters for the three-layer nested subgrid high-resolution simulation are set as follows: the initial boundary field uses NCEPA meteorological data at 0.25°×0.25°; the resolution of the three-layer nesting is 0.15°×0.15°, 0.05°×0.05°, and 0.01°×0.01°; the WRF parameters are set as follows: microphysical process scheme: WSM 6class; longwave radiation scheme: RRTM; shortwave radiation scheme: Dudhia; land surface process scheme: Noah; boundary layer scheme: YSU; cumulus parameterization scheme: Kain-Fritsch.

5. The method for monitoring rice irrigation patterns based on satellite remote sensing technology according to claim 1, characterized in that: The process of establishing the probability distribution of the SMI index of all high-standard farmland in the rice planting area during the rice planting stage is as follows: the SMI index of all high-standard farmland in the rice planting area during the rice planting stage is calculated with a time resolution of 3 days and a spatial resolution of 10m, thereby obtaining the SMI index of each high-standard farmland in the rice planting area during the rice planting stage, and plotting the probability distribution of the SMI index of each high-standard farmland.

6. The method for monitoring rice irrigation patterns based on satellite remote sensing technology according to claim 1, characterized in that: The probability distribution of the SMI index under the standard intermittent irrigation mode scenario is obtained by analyzing historical soil moisture content data of the current rice planting area under the intermittent irrigation mode scenario.

7. The method for monitoring rice irrigation patterns based on satellite remote sensing technology according to claim 1, characterized in that: The probability distribution of intermittent irrigation under the standard intermittent irrigation mode scenario is as follows: P=[N(0.13,0.02)+ N(0.16,0.02)] / 2, where N is a normal distribution, 0.13 is the mean of the normal distribution of irrigation flooding under the standard intermittent irrigation mode scenario, 0.16 is the mean of the normal distribution of field drying under the standard intermittent irrigation mode scenario, and 0.02 is the standard deviation of the two normal distributions.

Images