Soil water salt content prediction method based on interaction effect and optical remote sensing image
By extracting spectral reflectance from optical remote sensing images and constructing mapping relationships, combining quadratic equations and training a machine learning model, the accuracy problem of predicting soil water and salt content in irrigation areas using optical remote sensing images was solved, achieving higher accuracy prediction.
Patent Information
- Application Number
- CN202511731953.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-24
- Publication Date
- 2026-02-24
AI Technical Summary
In irrigated areas affected by salt, optical remote sensing images are difficult to accurately estimate soil moisture and salinity in farmland, especially when water and salt are mixed, resulting in poor remote sensing resolution.
By extracting the spectral reflectance of optical remote sensing images, calculating the spectral index, constructing a mapping relationship between soil water and salt content and the spectral index, combining two quadratic equations, and using machine learning methods to train a model for prediction, the model is calibrated to improve accuracy.
It improves the prediction accuracy of soil water and salt content, adapts to nonlinear relationships in complex environments, and reduces the error accumulation of a single model.
Smart Images
Figure CN121564569A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of agricultural remote sensing monitoring technology, and in particular to a method for predicting soil water and salt content based on interaction effects and optical remote sensing images. Background Technology
[0002] In irrigated areas affected by salinity, timely and accurate monitoring of soil water and salinity is crucial for water resource management and soil salinization control. Complex soil and groundwater conditions, coupled with irrigation management practices, lead to significant spatiotemporal variations in soil water and salinity. These factors exacerbate the complexity of optical remote sensing information, making it difficult to accurately estimate soil moisture and salinity from mixed data.
[0003] Optical remote sensing images have become an indispensable tool for understanding soil moisture processes in agricultural areas. Spectral indices based on broadband bands are often used to assess soil water and salt conditions in farmland. Artificial intelligence algorithms can also be used to address the performance limitations of single spectral indices.
[0004] However, when soil moisture deficit and salt accumulation occur simultaneously in farmland, the mixing of water and salt obscures the spectral behavior of crops, resulting in poor accuracy of remote sensing analysis of farmland soil water and salt information. Summary of the Invention
[0005] Therefore, it is necessary to provide a method for predicting soil water and salt content based on interaction effects and optical remote sensing images to address the aforementioned technical problems. This method improves the prediction accuracy of soil water and salt content.
[0006] The present invention adopts the following technical solution: This invention provides a method for predicting soil water and salt content based on interaction effects and optical remote sensing images, comprising: The spectral reflectance of all bands is extracted from sample optical remote sensing images of the target area over a historical period, and the corresponding spectral index is calculated based on the spectral reflectance of each band; the types of spectral indices include soil moisture index, soil salinity index and vegetation index. For each spectral index, a mapping relationship between soil water and salt content and the spectral index is constructed. Choose two or three from multiple mapping relationships to form a combination of two quadratic equations in two variables, and solve the combination of two quadratic equations in two variables to obtain the soil water and salt content; the soil water and salt content includes soil moisture content and soil salt content. With the goal of minimizing the difference between the calculated water and salt content and the actual water and salt content of the target area within a preset time period, a soil water and salt content estimation model based on machine learning is trained to obtain the soil water and salt content estimation model. The soil water and salt content estimation model is used to predict the water and salt content of the target area at the current time by using the optical remote sensing image of the target area.
[0007] Preferably, the mapping relationship is as follows: ; in, SI For spectral index, Soil moisture content, Soil salinity content, , , , and All are coefficients.
[0008] Preferably, the method for determining the coefficients of the mapping relationship specifically includes: The soil water and salt content in the target area was measured, and spectral remote sensing data of the target area was obtained. Using measured soil water and salt content as the independent variable and spectral remote sensing data as the dependent variable, the coefficients of the mapping relationship were determined by partial least squares method.
[0009] Preferably, the method further includes: Based on the predicted water and salt content of the target area at the current moment, draw a water and salt content distribution map of the target area.
[0010] Preferably, the method further includes: After acquiring the optical remote sensing image to be predicted, the image is preprocessed; the preprocessing includes atmospheric correction, image stitching, and cropping.
[0011] This invention provides a soil water and salt content prediction device based on interaction effects and optical remote sensing images, comprising: The calculation module is used to extract the spectral reflectance of all bands from sample optical remote sensing images of the target area within a historical time period, and calculate the corresponding spectral index based on the spectral reflectance of each band; the types of spectral indices include soil moisture index, soil salinity index and vegetation index. The mapping module is used to construct a mapping relationship between soil water and salt content and spectral index for each spectral index. The module is used to select two or three from multiple mapping relationships to form a combination of two quadratic equations in two variables, and solve the combination of two quadratic equations in two variables to obtain the soil water and salt content; the soil water and salt content includes soil moisture content and soil salt content. The training module is used to train the soil water and salt content estimation model based on machine learning methods with the goal of minimizing the difference between the calculated water and salt content and the actual water and salt content of the target area within a preset time period, so as to obtain the soil water and salt content estimation model. The prediction module is used to predict the water and salt content of the target area at the current time by using a soil water and salt content estimation model.
[0012] The present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for predicting soil water and salt content based on interaction effects and optical remote sensing images.
[0013] The present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, it implements the above-mentioned method for predicting soil water and salt content based on interaction effects and optical remote sensing images.
[0014] The above-mentioned at least one technical solution adopted in this invention can achieve the following beneficial effects: by extracting the spectral reflectance of multi-band remote sensing images, the spectral feature differences of surface materials can be captured, providing basic data for subsequent water and salt content analysis; for each spectral index, a mapping relationship between soil water and salt content and the spectral index is constructed respectively. By establishing the relationship between a single spectral index and soil water and salt content, remote sensing data and field measurement data are associated, providing independent variables for subsequent simultaneous equations; two or three mapping relationships are selected from multiple relationships to form a combination of two quadratic equations, and the combination of two quadratic equations is solved to obtain the soil water and salt content. Through the constraint effect of the equation set, the error accumulation of a single model is reduced; with the goal of minimizing the difference between the solved water and salt content and the actual water and salt content of the target area in a preset time period, the soil water and salt content estimation model based on machine learning is trained to obtain the soil water and salt content estimation model. The model is calibrated through training to adapt to nonlinear relationships in complex environments; the soil water and salt content estimation model is used to predict the optical remote sensing image of the target area to obtain the predicted value of water and salt content of the target area. This method improves the accuracy of soil water and salt content prediction. Attached Figure Description
[0015] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:
[0016] Figure 1 A schematic diagram of the process for predicting soil water and salt content based on interaction effects and optical remote sensing images provided by the present invention; Figure 2 The flowchart for predicting water salt content provided by this invention; Figure 3 A schematic diagram of a soil water and salt content prediction device based on interaction effects and optical remote sensing images provided by the present invention; Figure 4 This is a schematic diagram of a computer device for implementing a method for predicting soil water and salt content based on interaction effects and optical remote sensing images, as provided by the present invention. Detailed Implementation
[0017] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.
[0018] Devices such as desktop computers, servers, and laptops are capable of executing the present invention. For ease of explanation, the following description will focus on servers as the executing entity.
[0019] The technical solutions provided by the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
[0020] Figure 1 This is a schematic diagram of a method for predicting soil water and salt content based on interaction effects and optical remote sensing images according to the present invention, which specifically includes the following steps: S101: Extract the spectral reflectance of all bands from sample optical remote sensing images of the target area within a historical time period, and calculate the corresponding spectral index based on the spectral reflectance of each band; the types of spectral indices include soil moisture index, soil salinity index and vegetation index.
[0021] In one exemplary embodiment, after acquiring the optical remote sensing image to be predicted, the image is preprocessed; the preprocessing includes atmospheric correction, image stitching, and cropping.
[0022] Specifically, optical remote sensing images of a target area within a preset time period are acquired using a spectral camera mounted on a satellite or UAV. The original optical remote sensing images are then imported into the satellite or UAV's accompanying software to complete preprocessing steps such as atmospheric correction, image stitching, and cropping. The spectral reflectance of each band is then extracted from the preprocessed optical remote sensing images and calculated into various spectral indices according to the formulas provided in Table 1.
[0023] Table 1 R Blue R Green R Red R RedEdge R NIR R SWIR These represent the spectral reflectance in the blue, green, red, red-edged, near-infrared, and short-wave infrared bands, respectively.
[0024] S102: For each spectral index, construct the mapping relationship between soil water and salt content and spectral index.
[0025] In an exemplary embodiment, the mapping relationship is formula (1): (1); Where SI stands for spectral index. Soil moisture content, Soil salinity content, , , , and All are coefficients.
[0026] For each spectral index, the mapping relationship between soil water and salt content and the spectrum will be defined using a formula in the form of a binary quadratic regression.
[0027] S103: Select two or three from multiple mapping relationships to form a combination of two quadratic equations in two variables, and solve the combination of two quadratic equations in two variables to obtain the soil water and salt content; the soil water and salt content includes soil moisture content and soil salt content.
[0028] According to formula (1), any two or three spectral indices are mapped together to form a combination of two quadratic equations, thereby analyzing the soil water and salt content.
[0029] The constructed combination of two quadratic equations in two variables is formula (2) or formula (3): (2); (3); All spectral indices are combined in pairs or groups of three, and the water salt content is analyzed according to the above formulas (2) to (3) as the analysis result corresponding to the index combination.
[0030] S104: With the goal of minimizing the difference between the calculated water and salt content and the actual water and salt content of the target area within a preset time period, the soil water and salt content estimation model based on machine learning is trained to obtain the soil water and salt content estimation model.
[0031] The analytical results of all multidimensional index combinations were compared with the measured soil water and salt content. The analytical results of sensitive multidimensional index combinations were used as input variables, and a soil water and salt content estimation model was established using machine learning methods.
[0032] S105: The soil water and salt content estimation model is used to predict the optical remote sensing image of the target area at the current time, and the predicted value of water and salt content of the target area at the current time is obtained.
[0033] By running a trained soil water and salt content estimation model and inputting the optical remote sensing image to be predicted, the estimated results of farmland soil water and salt content are output.
[0034] In one exemplary embodiment, the method further includes: drawing a water and salt content distribution map of the target area based on the predicted water and salt content of the target area at the current moment.
[0035] In one exemplary embodiment, the present invention provides as follows Figure 2 The flowchart for predicting water and salt content is shown below. Figure 2 As shown, Step 1: Acquire an optical remote sensing image of the specified area (e.g., satellite or UAV), where the specified area is the target area. Step 2: Preprocess the acquired optical remote sensing image, extracting band spectral reflectance and calculating various spectral indices. Step 3: Establish a mapping relationship between the calculated spectral indices and the soil water and salt content of farmland. Step 4: Based on the established mapping relationship expression and multiple index formulas, solve the system to analyze the soil water and salt content. Step 5: Use machine learning methods, taking the analysis results of various index combinations as input variables, to jointly predict the soil water and salt content.
[0036] When applying the soil water and salt content prediction method based on interaction effects and optical remote sensing images provided by this invention, it is not necessary to rely on... Figure 1 The steps shown are executed in sequence. The specific execution order of each step can be determined as needed, and this invention does not impose any restrictions on it.
[0037] The above describes one or more embodiments of the present invention for predicting soil water and salt content based on interaction effects and optical remote sensing images. Based on the same idea, the present invention also provides a corresponding device for predicting soil water and salt content based on interaction effects and optical remote sensing images, such as... Figure 3 As shown.
[0038] Figure 3 A schematic diagram of a soil water and salt content prediction device based on interaction effects and optical remote sensing images provided by the present invention includes: The calculation module 301 is used to extract the spectral reflectance of all bands from the sample optical remote sensing images of the target area within a historical time period, and to calculate the corresponding spectral index based on the spectral reflectance of each band; the types of spectral indices include soil moisture index, soil salinity index and vegetation index.
[0039] The mapping module 302 is used to construct a mapping relationship between soil water and salt content and spectral index for each spectral index.
[0040] The construction module 303 is used to select two or three from multiple mapping relations to form a combination of two quadratic equations in two variables, and to solve the combination of two quadratic equations in two variables to obtain the soil water and salt content; the soil water and salt content includes soil moisture content and soil salt content.
[0041] Training module 304 is used to train a soil water and salt content estimation model based on machine learning methods with the goal of minimizing the difference between the calculated water and salt content and the actual water and salt content of the target area within a preset time period, so as to obtain a soil water and salt content estimation model.
[0042] The prediction module 305 is used to predict the optical remote sensing image of the target area at the current time using the soil water and salt content estimation model, and obtain the predicted value of water and salt content of the target area at the current time.
[0043] Specific limitations regarding the soil water and salt content prediction device based on interaction effects and optical remote sensing images can be found in the limitations of the soil water and salt content prediction method based on interaction effects and optical remote sensing images mentioned above, and will not be repeated here. Each module in the aforementioned soil water and salt content prediction device based on interaction effects and optical remote sensing images can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device in hardware form, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.
[0044] The present invention also provides a computer-readable storage medium storing a computer program that can be used to execute the above-described... Figure 1 A method for predicting soil water and salt content based on interaction effects and optical remote sensing images is provided.
[0045] The present invention also provides Figure 4 The schematic diagram of the computer device shown is as follows: Figure 4As shown, at the hardware level, this computer device includes a processor, internal bus, network interface, memory, and non-volatile memory, and may also include other hardware required for business operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then executes it to achieve the above. Figure 1 A method for predicting soil water and salt content based on interaction effects and optical remote sensing images is provided.
[0046] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the methods described above. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, or optical storage, etc. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM), etc.
[0047] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this invention.
Claims
1. A method for predicting soil water and salt content based on interaction effects and optical remote sensing images, characterized in that, include: The spectral reflectance of all bands is extracted from sample optical remote sensing images of the target area within a historical time period, and the corresponding spectral index is calculated based on the spectral reflectance of each band; the types of spectral indices include soil moisture index, soil salinity index and vegetation index. For each spectral index, a mapping relationship between soil water and salt content and the spectral index is constructed. Choose two or three from multiple mapping relationships to form a combination of two quadratic equations in two variables, and solve the combination of two quadratic equations in two variables to obtain the soil water and salt content; the soil water and salt content includes soil moisture content and soil salt content. With the goal of minimizing the difference between the calculated water and salt content and the actual water and salt content of the target area within a preset time period, a soil water and salt content estimation model based on machine learning is trained to obtain the soil water and salt content estimation model. The soil water and salt content estimation model is used to predict the water and salt content of the target area at the current time by using the optical remote sensing image of the target area at the current time.
2. The method as described in claim 1, characterized in that, The mapping relationship is as follows: ； in, SI For spectral index, Soil moisture content, Soil salinity content, , , , and All are coefficients.
3. The method as described in claim 2, characterized in that, The methods for determining the coefficients of the mapping relationship include: The soil water and salt content in the target area was measured, and spectral remote sensing data of the target area was obtained. Using measured soil water and salt content as the independent variable and spectral remote sensing data as the dependent variable, the coefficients of the mapping relationship were determined by partial least squares method.
4. The method as described in claim 1, characterized in that, The method further includes: Based on the predicted water and salt content of the target area at the current moment, draw a water and salt content distribution map of the target area.
5. The method as described in claim 1, characterized in that, The method further includes: After acquiring the optical remote sensing image to be predicted, the image is preprocessed; the preprocessing includes atmospheric correction, image stitching and cropping.
6. A soil water and salt content prediction device based on interaction effects and optical remote sensing images, characterized in that, include: The calculation module is used to extract the spectral reflectance of all bands from sample optical remote sensing images of the target area within a historical time period, and to calculate the corresponding spectral index based on the spectral reflectance of each band; the types of spectral indices include soil moisture index, soil salinity index and vegetation index. The mapping module is used to construct a mapping relationship between soil water and salt content and spectral index for each spectral index. The module is used to select two or three from multiple mapping relationships to form a combination of two quadratic equations in two variables, and to solve the combination of two quadratic equations in two variables to obtain the soil water and salt content; the soil water and salt content includes soil moisture content and soil salt content. The training module is used to train the soil water and salt content estimation model based on machine learning methods with the goal of minimizing the difference between the calculated water and salt content and the actual water and salt content of the target area within a preset time period, so as to obtain the soil water and salt content estimation model. The prediction module is used to predict the water and salt content of the target area at the current time using the soil water and salt content estimation model.
7. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method as described in any one of claims 1 to 5.
8. A computer device, characterized in that, It includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the program, implements the method as described in any one of claims 1 to 5.