Regional soil moisture monitoring method based on airborne spectral reconstruction optical satellite remote sensing

Abstract

A regional soil moisture monitoring method based on airborne spectral reconstruction optical satellite remote sensing, the method comprising: acquiring satellite remote sensing data of a region to be monitored, and collecting unmanned aerial vehicle remote sensing data of a local region in the region to be monitored; on the basis of the unmanned aerial vehicle remote sensing data of the local region, performing spectral reconstruction on the satellite remote sensing data of the region to be monitored, so as to obtain reconstructed satellite spectral data of the satellite remote sensing data; and inputting the reconstructed satellite spectral data into a pre-trained soil moisture inversion model, so as to obtain the soil moisture content of the region to be monitored. Therefore, soil moisture is accurately and quickly monitored.

Description

A regional soil moisture monitoring method of airborne spectrum reconstruction optical satellite remote sensing TECHNICAL FIELD

[0001] The present application relates to the technical field of soil moisture monitoring, in particular to a regional soil moisture monitoring method of airborne spectrum reconstruction optical satellite remote sensing. BACKGROUND

[0002] Soil moisture is usually quantified by soil moisture content, which has important research value in crop growth monitoring, regional drought monitoring and early warning, efficient use of water resources, precision irrigation and intelligent agriculture, etc.

[0003] At present, optical satellite remote sensing can be used to monitor soil moisture in a large irrigation area, but due to the limitations of spatial resolution and the dynamic changes of vegetation coverage (spatial heterogeneity of the ground), the accuracy of monitoring soil moisture is low. Low-altitude unmanned aerial vehicle remote sensing has the characteristics of ultra-high spatial resolution and fast response, which can capture rich ground information and has higher accuracy in monitoring soil moisture than satellite remote sensing. However, for a large irrigation area, it is time-consuming and laborious to collect unmanned aerial vehicle remote sensing data covering the entire irrigation area, and it is impossible to ensure the timeliness of soil moisture monitoring products.

[0004] Therefore, there is an urgent need for a method that can accurately and quickly monitor soil moisture. SUMMARY

[0005] Therefore, it is necessary to provide a regional soil moisture monitoring method of airborne spectrum reconstruction optical satellite remote sensing, which can accurately and quickly monitor soil moisture.

[0006] The present application adopts the following technical scheme:

[0007] The present application provides a regional soil moisture monitoring method of airborne spectrum reconstruction optical satellite remote sensing, comprising:

[0008] Obtaining satellite remote sensing data of the region to be monitored and unmanned aerial vehicle remote sensing data of a local region in the region to be monitored;

[0009] According to the unmanned aerial vehicle remote sensing data of the local region, the satellite remote sensing data of the region to be monitored is spectrum reconstructed to obtain reconstructed satellite spectrum data of the satellite remote sensing data;

[0010] The reconstructed satellite spectrum data is input into a pre-trained soil moisture inversion model to obtain the soil moisture content of the region to be monitored.

[0011] Preferably, according to the unmanned aerial vehicle remote sensing data of the local region, the satellite remote sensing data of the region to be monitored is spectrum reconstructed to obtain reconstructed satellite spectrum data of the satellite remote sensing data, comprising:

[0012] scaling up the unmanned aerial vehicle remote sensing data of the local area to determine unmanned aerial vehicle spectral data of the to-be-monitored area;

[0013] determining satellite spectral data of the to-be-monitored area according to satellite remote sensing data of the to-be-monitored area;

[0014] reconstructing the satellite spectral data by the unmanned aerial vehicle spectral data to obtain reconstructed satellite spectral data.

[0015] Preferably, the spectral data includes band reflectance and spectral index; the satellite spectral data is reconstructed by the unmanned aerial vehicle spectral data to obtain reconstructed satellite spectral data, including:

[0016] reconstructing the satellite band reflectance by the unmanned aerial vehicle spectral data to obtain reconstructed satellite band reflectance;

[0017] determining the reconstructed satellite spectral index according to the reconstructed satellite band reflectance and a preset conversion function.

[0018] Preferably, the calculation method of the reconstructed satellite band reflectance is: B r =B o {1+k·[FVC / (B u ·B o )]}

[0019] wherein, B r is the reconstructed satellite band reflectance; B0 is the satellite band reflectance; FVC is a fractional vegetation cover corresponding to the spatial resolution scale of the optical satellite; k is an adjustment factor, B u is the scaled unmanned aerial vehicle spectral data, including scaled unmanned aerial vehicle band reflectance and spectral index.

[0020] Preferably, the construction process of the soil moisture inversion model includes:

[0021] obtaining original satellite training sample data and unmanned aerial vehicle training sample data, and corresponding soil water content;

[0022] determining original satellite sample spectral data according to the original satellite training sample data;

[0023] reconstructing the original satellite sample spectral data by the unmanned aerial vehicle training sample data to obtain reconstructed satellite sample spectral data;

[0024] training the extreme learning machine model by the reconstructed satellite sample spectral data to obtain the soil moisture inversion model.

[0025] Preferably, the method further includes:

[0026] Before training the soil moisture inversion model, the original satellite sample spectrum data and the reconstructed satellite sample spectrum data are respectively correlated with the soil moisture content.

[0027] Preferably, the method further comprises:

[0028] The extreme learning machine model is trained by the original satellite sample spectrum data to obtain a soil moisture inversion model under the original satellite sample spectrum data;

[0029] The extreme learning machine model is trained by the reconstructed satellite sample spectrum data to obtain a soil moisture inversion model under the reconstructed satellite sample spectrum data;

[0030] The performance of the soil moisture inversion model under the original satellite sample spectrum data and the soil moisture inversion model under the reconstructed satellite sample spectrum data is tested to obtain an original performance test result of the soil moisture inversion model under the original satellite sample spectrum data and a performance test result of the soil moisture inversion model under the reconstructed satellite sample spectrum data.

[0031] The application provides a device for monitoring regional soil moisture by airborne spectrum reconstruction optical satellite remote sensing, comprising:

[0032] An acquisition module is configured to acquire satellite remote sensing data of a region to be monitored and unmanned aerial vehicle remote sensing data of a local region in the region to be monitored;

[0033] A reconstruction module is configured to perform spectrum reconstruction on the satellite remote sensing data of the region to be monitored according to the unmanned aerial vehicle remote sensing data of the local region, and obtain reconstructed satellite spectrum data of the satellite remote sensing data;

[0034] A monitoring module is configured to input the reconstructed satellite spectrum data into a pre-trained soil moisture inversion model, and obtain soil moisture content of the region to be monitored.

[0035] The application provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor to implement the above-mentioned method for monitoring regional soil moisture by airborne spectrum reconstruction optical satellite remote sensing.

[0036] The application provides a computer device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the above-mentioned method for monitoring regional soil moisture by airborne spectrum reconstruction optical satellite remote sensing when executing the program.

[0037] The above-mentioned at least one technical solution adopted by the application can achieve the following beneficial effects:

[0038] The satellite remote sensing data of the to-be-monitored region is spectrally reconstructed by the unmanned aerial vehicle remote sensing data of the local region in the to-be-monitored region, so that the reconstructed satellite spectral data is fused with the unmanned aerial vehicle remote sensing data of the local region, so that the soil water content of the to-be-monitored region determined by the reconstructed satellite spectral data is more accurate, and the timeliness is ensured, so that the accurate and rapid soil moisture monitoring of the to-be-monitored region is realized. BRIEF DESCRIPTION OF DRAWINGS

[0039] The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and serve to explain the principles of the application. In the drawings:

[0040] Fig. 1 is a flowchart of a method for monitoring regional soil moisture by airborne spectral reconstruction of optical satellite remote sensing according to the present application;

[0041] Fig. 2 is a flowchart of another method for monitoring regional soil moisture by airborne spectral reconstruction of optical satellite remote sensing according to the present application;

[0042] Fig. 3 is a flowchart of another method for monitoring regional soil moisture by airborne spectral reconstruction of optical satellite remote sensing according to the present application;

[0043] Fig. 4 is a matrix diagram of correlation analysis of original and reconstructed band reflectance and spectral index and soil water content according to the present application;

[0044] Fig. 5 is a scatter plot of accuracy comparison of soil moisture inversion models under original and reconstructed satellite spectral data according to the present application;

[0045] Fig. 6 is a schematic diagram of a device for monitoring regional soil moisture by airborne spectral reconstruction of optical satellite remote sensing according to the present application;

[0046] Fig. 7 is a schematic diagram of a computer device for implementing a method for monitoring regional soil moisture by airborne spectral reconstruction of optical satellite remote sensing according to the present application. DETAILED DESCRIPTION

[0047] To make the objectives, technical solutions and advantages of the present application clearer, the technical solutions of the present application will be described below in conjunction with specific embodiments of the present application and corresponding drawings. Obviously, the described embodiments are only some of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0048] Multi-scale fusion of satellite and unmanned aerial vehicle (UAV) optical remote sensing data can greatly improve the accuracy of satellite remote sensing monitoring of soil moisture. However, for large-scale irrigation areas, it is time-consuming and laborious to collect full-coverage UAV remote sensing data in the irrigation area, and the timeliness of the soil moisture monitoring product cannot be guaranteed.

[0049] Therefore, the present application provides a method for monitoring regional soil moisture by airborne spectral reconstruction of optical satellite remote sensing, which integrates local area UAV optical remote sensing data to reconstruct satellite remote sensing band reflectivity and spectral index, thereby improving the accuracy of satellite remote sensing monitoring of soil moisture and achieving high spatial and temporal resolution and high precision monitoring of soil moisture in large-scale irrigation areas.

[0050] The technical solutions provided by the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0051] Fig. 1 is a flowchart of a method for monitoring regional soil moisture by airborne spectral reconstruction of optical satellite remote sensing, which specifically includes the following steps:

[0052] S101, acquiring satellite remote sensing data of a region to be monitored and UAV remote sensing data of a local region in the region to be monitored.

[0053] The region to be monitored is a region whose soil moisture is to be monitored, and the local region can be a region whose soil moisture is representative in the region to be monitored.

[0054] The remote sensing data of the region to be monitored can be directly collected by a satellite, and the remote sensing data of the local region in the region to be monitored can be collected by a UAV. After the satellite collects the remote sensing data of the region to be monitored, it is sent to a server to obtain the satellite remote sensing data of the region to be monitored. After the UAV collects the remote sensing data of the local region in the region to be monitored, it can be sent to the server to obtain the UAV remote sensing data of the local region in the region to be monitored.

[0055] The server mentioned in the present application can be a server arranged in a business platform or a device such as a desktop computer or a notebook computer capable of executing the scheme of the present application. For the convenience of description, only the server is described as the execution subject below.

[0056] S102, performing spectral reconstruction on the satellite remote sensing data of the region to be monitored according to the UAV remote sensing data of the local region to obtain reconstructed satellite spectral data of the satellite remote sensing data.

[0057] As shown in Fig. 2, the spectral reconstruction on the satellite remote sensing data of the region to be monitored according to the UAV remote sensing data of the local region to obtain the reconstructed satellite spectral data of the satellite remote sensing data includes the following steps:

[0058] S201, scale up the unmanned aerial vehicle remote sensing data of the local region to determine the unmanned aerial vehicle spectral data of the region to be monitored.

[0059] The unmanned aerial vehicle remote sensing data of the local region can be converted into the unmanned aerial vehicle spectral data of the region to be monitored through scale up. The scale up factor can be determined according to requirements, for example, from 5 centimeter resolution to 10 meter resolution.

[0060] Alternatively, the unmanned aerial vehicle remote sensing data of the local region can also be input into the first learning model to perform scale up conversion through the first learning model to obtain the unmanned aerial vehicle spectral data of the region to be monitored. The unmanned aerial vehicle spectral data includes the spectral data of all regions in the region to be monitored. Specifically, in the first learning model, the unmanned aerial vehicle spectral data of the local region can be determined according to the unmanned aerial vehicle remote sensing data of the local region, and then the other regions of the region to be monitored learn from the unmanned aerial vehicle spectral data of the local region to obtain the unmanned aerial vehicle spectral data of the other regions of the region to be monitored, thereby obtaining the unmanned aerial vehicle spectral data of all regions in the region to be monitored.

[0061] Alternatively, the unmanned aerial vehicle remote sensing data of the local region can also be input into the second learning model to perform scale up conversion through the second learning model to obtain the unmanned aerial vehicle spectral data of the region to be monitored. The unmanned aerial vehicle spectral data includes the spectral data of all regions in the region to be monitored. Specifically, in the second learning model, the other regions of the region to be monitored first learn from the unmanned aerial vehicle remote sensing data of the local region to obtain the unmanned aerial vehicle remote sensing data of the other regions of the region to be monitored, thereby obtaining the unmanned aerial vehicle remote sensing data of all regions in the region to be monitored, and analyzing the unmanned aerial vehicle remote sensing data of all regions in the region to be monitored to obtain the unmanned aerial vehicle spectral data of the region to be monitored.

[0062] S202, determine the satellite spectral data of the region to be monitored according to the satellite remote sensing data of the region to be monitored.

[0063] The satellite remote sensing data of the region to be monitored can include satellite spectral data.

[0064] S203, reconstruct the satellite spectral data through the unmanned aerial vehicle spectral data to obtain reconstructed satellite spectral data.

[0065] The spectral data includes band reflectivity and spectral index. The satellite spectral data is reconstructed through the unmanned aerial vehicle spectral data to obtain reconstructed satellite spectral data, including: reconstructing the satellite band reflectivity through the unmanned aerial vehicle spectral data to obtain reconstructed satellite band reflectivity; and determining the reconstructed satellite spectral index according to the reconstructed satellite band reflectivity and a preset conversion function.

[0066] The calculation method of the reconstructed satellite band reflectivity is: Br = B o {1 + k · [FVC / (B u · B o )]} (1)

[0067] wherein B r is the reconstructed satellite band reflectance; B0 is the satellite band reflectance; FVC is the fractional vegetation cover corresponding to the spatial resolution scale of the optical satellite; k is an adjustment factor, B u is the unmanned aerial vehicle band reflectance or the unmanned aerial vehicle spectral index.

[0068] The reconstructed spectral index is a function of the reconstructed band reflectance, i.e., the conversion function is shown in formula (2). S rec = f(B r ) (2)

[0069] wherein S rec is the reconstructed satellite spectral index, f(·) is a function from B r to S rec , and the reconstructed satellite spectral index can include a vegetation index, a salinity index, and a brightness index.

[0070] For example, as shown in Table 1, Table 1 is a conversion function between band reflectance and spectral index.

[0071] Table 1

[0072] wherein B, R, G, NIR are the band reflectance corresponding to blue (Blue), red (Red), green (Green), and near-infrared (NIR), respectively.

[0073] S103, input the reconstructed satellite spectral data into the pre-trained soil moisture inversion model to obtain the soil water content of the monitoring area.

[0074] The reconstructed satellite band reflectance and the reconstructed satellite spectral index are input into the pre-trained soil moisture inversion model to obtain the soil water content of the monitoring area output by the soil moisture inversion model.

[0075] Preferably, the construction process of the soil moisture inversion model comprises: obtaining original satellite training sample data and unmanned aerial vehicle training sample data, and corresponding soil water content; determining original satellite sample spectral data according to the original satellite training sample data, and performing spectral reconstruction on the original satellite sample spectral data according to the unmanned aerial vehicle training sample data to obtain reconstructed satellite sample spectral data; training the extreme learning machine model through the reconstructed satellite sample spectral data to obtain the soil moisture inversion model.

[0076] Specifically, the unmanned aerial vehicle multispectral remote sensing data, the optical satellite remote sensing image and the soil water content data having local representativeness (different vegetation coverage conditions) can be acquired first and preprocessed. The measured soil water content and the unmanned aerial vehicle multispectral remote sensing data are obtained through field experiments, and the satellite remote sensing data are obtained by downloading from a satellite remote sensing data website. The number of pixels of the unmanned aerial vehicle is scaled up to the satellite remote sensing scale, and the number of ground soil water samples is not less than 120. The unmanned aerial vehicle multispectral remote sensing data are taken as the unmanned aerial vehicle training sample data, and the optical satellite remote sensing image is taken as the satellite training sample data.

[0077] First, the satellite training sample data are spectrally reconstructed according to the unmanned aerial vehicle training sample data to obtain reconstructed satellite sample spectral data. It should be noted that the way of obtaining the reconstructed satellite sample spectral data is the same as the way of obtaining the reconstructed satellite spectral data of the satellite remote sensing data in the above embodiment, which will not be described here again. The reconstructed satellite sample spectral data include reconstructed satellite sample band reflectance and reconstructed satellite sample spectral index.

[0078] The k value in the reconstructed satellite sample band reflectance, that is, the k value in formula (1), is determined by constructing a linear correlation function r(k). The linear correlation function is constructed by formula (3)-formula (6). Formula (3) is the calculation of the linear correlation coefficient of the reconstructed satellite sample band reflectance and the corresponding soil water content.

[0079] Wherein, S m is the soil mass water content.

[0080] According to formula (4), formula (3) can be converted into formula (5).

[0081] In the case of a given sample number n, M0, M1, M2, M3, M4, and M5 are also determined, at this time they all become constants, then equation (5) is expressed as equation (6) using equation (6):

[0082] In the formula, r(k) is the linear correlation function between the reconstructed band reflectance and the soil water content, and r(k) ∈ [-1, 1].

[0083] Differentiate equation (6) with respect to k on both sides to obtain formula (7):

[0084] There is an extreme value (maximum value) of the correlation r(k) in equation (7), and equation (8) is obtained by letting dr / dk = 0 in equation (7):

[0085] For the linear correlation function r(k), there are four special cases (i), (ii), (iii) and (iv) as follows.

[0086] (i) When k = (M1M4-M2M5) / (M2M4-M1M3) in equation (6), there is formula (9):

[0087] where (k e , r e ) is the only extreme point of the function r(k) in the range of real numbers, and r e is the only extreme value of the function r(k).

[0088] (ii) When k = 0 in equation (6), there is formula (10):

[0089] where r o represents the linear correlation coefficient of B o and S m , that is, the correlation coefficient of the original satellite band reflectance and soil moisture.

[0090] (iii) When k = -M1 / M2 in equation (6), r(k) = 0, that is, there is no linear correlation between the reconstructed satellite band reflectance and soil moisture.

[0091] (iv) When k→∞ in equation (6), there is formula (11):

[0092] where r ∞ represents the linear correlation coefficient of B r and S m at positive and negative infinity.

[0093] The reconstructed satellite sample spectral data includes reconstructed satellite sample band reflectance and reconstructed satellite sample spectral index; according to the k m calculated in formula (8), the original satellite band reflectance B0 corresponding to the satellite training sample data, the vegetation coverage FVC, and the unmanned aerial vehicle multi-spectral band reflectance or spectral index B u corresponding to the upscaled unmanned aerial vehicle training sample data are calculated according to formula (1) to obtain the reconstructed satellite sample band reflectance, and the reconstructed satellite sample spectral index is calculated based on the reconstructed satellite sample band reflectance.

[0094] After obtaining the reconstructed satellite sample spectral data, the extreme learning machine model is trained through the reconstructed satellite sample spectral data to obtain a soil moisture inversion model under the reconstructed satellite sample spectral data. Specifically, a machine learning algorithm such as extreme learning machine is used to construct a soil moisture inversion regression model of the reconstructed satellite sample spectral data to obtain the soil moisture inversion model.

[0095] In an exemplary embodiment, the embodiment comprises: before training the soil moisture inversion model, performing correlation analysis on the original satellite sample spectrum data and the reconstructed satellite sample spectrum data and the soil water content, respectively. The correlation analysis adopts Pearson correlation.

[0096] And, the extreme learning machine model is trained by the original satellite sample spectrum data to obtain the soil moisture inversion model under the original satellite sample spectrum data; the extreme learning machine model is trained by the reconstructed satellite sample spectrum data to obtain the soil moisture inversion model under the reconstructed satellite sample spectrum data; the performance test is performed on the soil moisture inversion model under the original satellite sample spectrum data and the soil moisture inversion model under the reconstructed satellite sample spectrum data to obtain the original performance test result of the soil moisture inversion model under the original satellite sample spectrum data and the performance test result of the soil moisture inversion model under the reconstructed satellite sample spectrum data.

[0097] The soil moisture inversion model under the original satellite sample spectrum data and the soil moisture inversion model under the reconstructed satellite sample spectrum data are a kind of inversion regression model.

[0098] Optionally, the soil moisture inversion model constructed by using the original satellite sample spectrum data and the soil moisture inversion model constructed by using the reconstructed satellite sample spectrum data are respectively evaluated in accuracy by coefficient of determination (R 2 ) and root mean square error (RMSE), and the increase and decrease amplitude of R 2 and RMSE is used to evaluate the promotion effect of the soil moisture inversion model under the reconstructed satellite sample spectrum data relative to the soil moisture inversion model under the original satellite sample spectrum data.

[0099] In an exemplary embodiment, as shown in FIG. 3, the present application also provides a kind of regional soil moisture monitoring method of airborne spectrum reconstruction optical satellite remote sensing, specifically, the satellite image of satellite sample area is acquired, then the satellite image is preprocessed, and original spectrum data is obtained, and the unmanned aerial vehicle image of sample area collected by unmanned aerial vehicle is acquired, and the unmanned aerial vehicle image is preprocessed, and the preprocessed unmanned aerial vehicle image data is obtained, the unmanned aerial vehicle image processing is upscaled, and the upscaled unmanned aerial vehicle multispectral data is obtained.

[0100] Then the original satellite spectral data is reconstructed by upscaling the unmanned aerial vehicle multispectral data to obtain reconstructed satellite spectral data. The original satellite spectral data and the reconstructed spectral data are respectively correlated with the measured soil water content. The original satellite spectral data and the reconstructed satellite spectral data are respectively taken as independent variables, and the measured soil water content is taken as a dependent variable. The soil moisture inversion model under the original satellite spectral data and the soil moisture inversion model under the reconstructed satellite spectral data are trained. The soil moisture inversion model under the original satellite spectral data and the soil moisture inversion model under the reconstructed satellite spectral data are evaluated by evaluation indexes (R 2 and RMSE) to determine that the performance of the soil moisture inversion model under the reconstructed satellite spectral data is better.

[0101] The airborne spectral reconstruction optical satellite remote sensing regional soil moisture monitoring method provided by the application is further described below through a specific embodiment.

[0102] Firstly, the local (four different vegetation coverage fields are selected) unmanned aerial vehicle multispectral remote sensing data of Jiefangzha Shahao Canal Irrigation Area in A River Basin, the measured soil moisture data of four typical fields collected synchronously, and the cloud-free Landsat 8 satellite multispectral remote sensing data corresponding to the research area are downloaded. The irrigation area soil water content inversion model based on the original satellite spectral data and the irrigation area soil water content inversion model based on the airborne spectral reconstruction satellite spectral data are respectively constructed. The respective inversion accuracy and improvement effect are compared by evaluation indexes.

[0103] Step 1, soil sample collection and unmanned aerial vehicle flight test are carried out on the selected four typical fields, and the time is August 12-15, 2019. The visible light bands of the unmanned aerial vehicle multispectral data, namely blue (Blue), green (Green) and red (Red), are used for upscaling conversion. Five-point sampling method is used to collect 0-10 cm depth soil samples. The soil mass water content S m is calculated by the drying method (105℃, 24h constant temperature treatment), and the calculation formula is as follows:

[0104] Wherein, m1 is the weight of wet soil plus empty aluminum box, m2 is the weight of dry soil plus empty aluminum box, and m3 is the weight of empty aluminum box.

[0105] The Landsat 8 multispectral satellite remote sensing data is downloaded from USGS Earth Explorer, and the obtained Landsat 8 satellite image is a Collection2 Level-1 product after geometric correction processing, and the imaging time of the image is August 15, 2019, which is close to the field sample collection time. Further processing is performed on the Landsat 8 satellite image data through ENVI software, including radiation calibration, atmospheric correction, cutting, etc., and then the reflectivity data of each band at the sampling point is extracted. In the present application, the blue (Blue), green (Green), red (Red), near-infrared (NIR), short-wave infrared-1 (SWIR1) and short-wave infrared-2 (SWIR2) of the Landsat 8 optical satellite data are selected for reconstruction.

[0106] Step 2, FVC is obtained by extracting the proportion of vegetation elements in high-resolution unmanned aerial multi-spectral images at a given scale (30m), and the calculation formula is as follows.

[0107] In the formula: N v and N no-v are the number of vegetation and non-vegetation pixels corresponding to the spatial resolution scale of the Landsat 8 satellite, respectively.

[0108] B u In the present application, the S3 salinity index calculated by upscaled unmanned aerial remote sensing data is selected, that is, S3 u , and the calculation formula is as follows:

[0109] S3 u =Green u ×Red u / Blue u (14)

[0110] Wherein, Blue u , Green u and Red u are the blue, green and red band reflectivity of the unmanned aerial remote sensing data upscaled to 30m spatial resolution, and the upscaled conversion method is the pixel aggregation method.

[0111] The k m value of each reconstructed band reflectivity is calculated by the k value formula at the extreme point of the linear r(k) function, that is, k m-Blue , k m-Green , k m-Red , k m-NIR , k m-SWIR1 and k m-SWIR2, and finally the reconstructed band reflectance of each band is obtained, as shown in equations (15)-(20). re = Blue {1 + k m-Blue · [FVC / (S3 u · Blue)]} (15) re = Green {1 + k m-Green · [FVC / (S3 u · Green)]} (16) re = Red {1 + k m-Red · [FVC / (S3 u · Red)]} (17) re = NIR {1 + k m-NIR · [FVC / (S3 u · NIR)]} (18) re = SWIR1 {1 + k m-SWIR1 · [FVC / (S3 u · SWIR1)]} (19) re = SWIR2 {1 + k m-SWIR2 · [FVC / (S3 u · SWIR2)]} (20)

[0112] Step 3, construct 8 original spectral indices (including 4 salt indices, 3 vegetation indices and 1 brightness index) based on the band reflectance of the original Landsat 8 optical satellite data; construct 8 reconstructed spectral indices (including 4 reconstructed salt indices, 3 reconstructed vegetation indices and 1 reconstructed brightness index) based on the band reflectance of the reconstructed Landsat 8 optical satellite data, and the specific calculation formula is shown in Table 2:

[0113] Table 2

[0114] Wherein, B, G, B and NIR represent the original blue, green, red and near-infrared band reflectance, that is, B = Blue, G = Green, R = Red; B re , G re , R re and NIR re represent the reconstructed blue, green, red and near-infrared band reflectance, that is, B re = Blue re , G re = Green re , R re = Redre .

[0115] Step 4, the measured soil moisture content is respectively correlated with the original and reconstructed satellite band reflectance and spectral index, and the results are shown in the Pearson correlation matrix diagram, see Figure 4, wherein the larger the circle represents the stronger the correlation. As can be seen from Figure 4, compared with the original satellite band reflectance and spectral index, the correlation of the reconstructed satellite band reflectance and spectral index with the soil moisture content is significantly enhanced.

[0116] Step 5, the extreme learning machine is a kind of artificial neural network model training algorithm, and its composition usually includes input layer, hidden layer and output layer. In the model training process, the connection weight of the input layer and the hidden layer is random and does not need to be adjusted, and only the number of hidden layer nodes of the network needs to be set, that is, a global optimal solution can be generated. In the present application, the number of hidden layer nodes is uniformly set to 20, and the construction of the extreme learning machine model is completed through MATLAB software. The sample number ratio of the modeling set and the validation set is set to 2:1, that is, the sample number of the modeling set is 79, and the sample number of the validation set is 40. The extreme learning machine regression algorithm is used to construct the soil moisture inversion model under the original satellite data, taking all original band reflectance and spectral index as independent variables and soil moisture content as dependent variable. The extreme learning machine regression algorithm is used to construct the soil moisture inversion model under the reconstructed satellite data, taking all reconstructed band reflectance and spectral index as independent variables and soil moisture content as dependent variable, and the results are shown in Figure 5.

[0117] Step 6, the coefficient of determination can reflect the fitting effect of the model. R 2 The closer to 1, the smaller the RMSE, indicating that the prediction effect of the model is better, and the error between the predicted value and the measured value is smaller. The R 2 and RMSE between the measured value and the predicted value of the soil moisture content are calculated, as shown in Table 3.

[0118] Wherein, y i represents the measured value, represents the predicted value, represents the average value of the measured value, and n represents the sample number.

[0119] Table 3

[0120] It should be noted that the results in Table 3 are the R 2 and RMSE increase and decrease amplitude of the extreme learning machine model constructed by using the reconstructed satellite spectral data compared with the extreme learning machine model constructed by using the original satellite spectral data; + for rising, - for reducing.

[0121] As can be seen from Figure 5 and Table 3, the R 2The value is 0.441, and the RMSE is 0.034 cm. 3 / cm 3 In contrast, the validation set R of the soil moisture retrieval model based on reconstructed satellite data, constructed using the extreme learning machine regression algorithm, is... 2 The value is 0.663, and the RMSE is 0.025cm. 3 / cm 3 Validation set R 2 The index increased by 0.222 (an increase of approximately 50.34%), while the RMSE decreased by 0.009 cm. 3 / cm 3 (The decrease was approximately 26.47%). This demonstrates that incorporating localized UAV multispectral data can effectively improve the accuracy of soil moisture models retrieved from single satellite optical remote sensing data, enhance the accuracy of satellite remote sensing monitoring of soil moisture, and achieve high spatiotemporal resolution and high-precision monitoring of soil moisture in large-scale irrigation areas. This invention provides a theoretical basis and technical support for improving the accuracy of regional soil moisture content monitoring based on satellite remote sensing spectral data.

[0122] When applying the regional soil moisture monitoring method of airborne spectral reconstruction optical satellite remote sensing provided by the present invention, the steps shown in Figure 1 may not be executed in the same order. The specific execution order of each step can be determined as needed, and the present invention does not impose any restrictions on this.

[0123] The above describes one or more embodiments of the regional soil moisture monitoring method using airborne spectral reconstruction optical satellite remote sensing. Based on the same idea, the present invention also provides a corresponding regional soil moisture monitoring device using airborne spectral reconstruction optical satellite remote sensing, as shown in Figure 6.

[0124] Figure 6 is a schematic diagram of a regional soil moisture monitoring device for airborne spectral reconstruction optical satellite remote sensing provided by the present invention. The device 600 includes:

[0125] The acquisition module 601 is used to acquire satellite remote sensing data of the area to be monitored and to collect UAV remote sensing data of local areas within the area to be monitored.

[0126] The reconstruction module 602 is used to perform spectral reconstruction of satellite remote sensing data of the area to be monitored based on UAV remote sensing data of a local area, so as to obtain reconstructed satellite spectral data of the satellite remote sensing data.

[0127] The monitoring module 603 is used to input the reconstructed satellite spectral data into a pre-trained soil moisture inversion model to obtain the soil moisture content of the area to be monitored.

[0128] The specific limitations of the device for monitoring regional soil moisture by airborne spectral reconstruction optical satellite remote sensing can refer to the limitations of the method for monitoring regional soil moisture by airborne spectral reconstruction optical satellite remote sensing, which will not be repeated here. Each module in the device for monitoring regional soil moisture by airborne spectral reconstruction optical satellite remote sensing can be realized by software, hardware, and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to each module.

[0129] The application also provides a computer readable storage medium storing a computer program, which can be used to execute the method for monitoring regional soil moisture by airborne spectral reconstruction optical satellite remote sensing provided in FIG. 1.

[0130] The application also provides a structural diagram of the computer device shown in FIG. 7. As shown in FIG. 7, at the hardware level, the computer device includes a processor, an internal bus, a network interface, a memory, and a non-volatile memory, and can also include other hardware required by the business. The processor reads the corresponding computer program from the non-volatile memory into the memory and then runs to implement the method for monitoring regional soil moisture by airborne spectral reconstruction optical satellite remote sensing provided in FIG. 1.

[0131] Those skilled in the art can understand that all or part of the above-mentioned embodiment methods can be completed by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer readable storage medium, and when executed, can include the flow of the above-mentioned embodiment methods. In the embodiments of the present application, any reference to the memory, storage, database, or other medium can include at least one of the non-volatile and volatile memories. The non-volatile memory can include a read-only memory (ROM), a magnetic tape, a floppy disk, a flash memory, or an optical memory. The volatile memory can include a random access memory (RAM) or an external cache memory. As an illustration but not limitation, the RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0132] The technical features of the above embodiments can be combined in any manner. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described, however, as long as the combinations of the technical features do not contradict each other, they should be considered to be within the scope of the present application.

Claims

1. A method for regional soil moisture monitoring of airborne spectral reconstruction optical satellite remote sensing, characterized in that, The method comprises: acquiring satellite remote sensing data of a to-be-monitored region and collecting unmanned aerial vehicle remote sensing data of a local region in the to-be-monitored region; reconstructing the satellite remote sensing data of the to-be-monitored region according to the unmanned aerial vehicle remote sensing data of the local region to obtain reconstructed satellite spectral data of the satellite remote sensing data; the spectral data comprises band reflectivity; the reconstructed satellite band reflectivity is calculated in the following manner: B r = B o {1 + k · [FVC / (B u · B o )]} wherein B r is the reconstructed satellite band reflectance; B0 is the satellite band reflectance; FVC is the fractional vegetation cover corresponding to the spatial resolution scale of the optical satellite; k is an adjustment factor, B u For UAV band reflectance or UAV spectral index; inputting the reconstructed satellite spectral data into a pre-trained soil moisture inversion model to obtain soil water content of the to-be-monitored region.

2. The method of claim 1, wherein, The method comprises: scaling up the unmanned aerial vehicle remote sensing data of the local region to determine unmanned aerial vehicle spectral data of the to-be-monitored region; determining satellite spectral data of the to-be-monitored region according to the satellite remote sensing data of the to-be-monitored region; reconstructing the satellite spectral data according to the unmanned aerial vehicle spectral data to obtain the reconstructed satellite spectral data.

3. The method of claim 2, wherein, The spectral data further comprises spectral indices; the method comprises: reconstructing the satellite band reflectivity according to the unmanned aerial vehicle spectral data to obtain reconstructed satellite band reflectivity; determining reconstructed satellite spectral indices according to the reconstructed satellite band reflectivity and a preset conversion function.

4. The method of claim 1, wherein, The method further comprises: acquiring original satellite training sample data and unmanned aerial vehicle training sample data, and corresponding soil water content; determining original satellite sample spectral data according to the original satellite training sample data; reconstructing the original satellite sample spectral data according to the unmanned aerial vehicle training sample data to obtain reconstructed satellite sample spectral data; training an extreme learning machine model according to the reconstructed satellite sample spectral data to obtain the soil moisture inversion model.

5. The method of claim 4, wherein, The method further comprises: performing correlation analysis on the original satellite sample spectral data and the reconstructed satellite sample spectral data and the soil water content respectively before training the soil moisture inversion model.

6. The method of claim 4, wherein, The method further comprises: training an extreme learning machine model according to the original satellite sample spectral data to obtain a soil moisture inversion model under original satellite sample spectral data; training an extreme learning machine model according to the reconstructed satellite sample spectral data to obtain a soil moisture inversion model under reconstructed satellite sample spectral data; performing performance testing on the soil moisture inversion model under the original satellite sample spectral data and the soil moisture inversion model under the reconstructed satellite sample spectral data to obtain original performance testing results of the soil moisture inversion model under the original satellite sample spectral data and performance testing results of the soil moisture inversion model under the reconstructed satellite sample spectral data.

7. An airborne optical spectral reconstruction optical satellite remote sensing device for regional soil moisture monitoring, characterized in that, The method further comprises: an acquisition module configured to acquire satellite remote sensing data of a to-be-monitored region and collect unmanned aerial vehicle remote sensing data of a local region in the to-be-monitored region; The reconstruction module is configured to perform spectral reconstruction on the satellite remote sensing data of the region to be monitored according to the UAV remote sensing data of the local region, to obtain reconstructed satellite spectral data of the satellite remote sensing data; the spectral data includes band reflectivity; and the calculation manner of the reconstructed satellite band reflectivity is as follows: B r = B o {1 + k · [FVC / (B u · B o )]} wherein B r is the reconstructed satellite band reflectance; B0 is the satellite band reflectance; FVC is the fractional vegetation cover corresponding to the spatial resolution scale of the optical satellite; k is an adjustment factor, B u For UAV band reflectance or UAV spectral index; The monitoring module is configured to input the reconstructed satellite spectral data into a pre-trained soil moisture inversion model, to obtain the soil water content of the region to be monitored.

8. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, and the computer program is executed by the processor to implement the method in any one of claims 1-6.

9. A computer device, comprising: The computer program is stored in the memory and executable on the processor, and the processor implements the method in any one of claims 1-6 when executing the program.

Images