A soil and water loss map spot landing system based on weak constraint of land use information

Abstract

The application discloses a soil erosion map spot landing system based on weak constraint of land utilization information, which comprises a land utilization information collection module, a factor evaluation and analysis module, a time sequence analysis and map spot landing module and a data integration and visualization module.The land utilization information collection module is used for collecting land utilization data; the factor evaluation and analysis module is used for calculating soil erosion modulus and evaluating and analyzing influence factors; the time sequence analysis and map spot landing module is used for soil erosion change analysis and generation of landing map spots; and the data integration and visualization module is used for data integration and map spot visualization display.The soil erosion map spot landing system based on weak constraint of land utilization information calculates erosion modulus by using an erosion modulus calculation algorithm based on data fusion and pixel unmixing, and analyzes map spot change by using a soil erosion map spot change analysis algorithm based on time sequence classification.

Description

Technical Field

[0001] This invention relates to the fields of data fusion, pixel demixing, and time series classification technology, specifically to a soil erosion map localization system based on weak constraints of land use information. Background Technology

[0002] Data fusion technology is a technique that improves data quality and accuracy by integrating data from different sources. It aims to solve the problems of insufficient information and poor accuracy from single data sources. This technology improves the spatial resolution and accuracy of soil erosion monitoring by integrating multiple data sources such as remote sensing data, geographic information system data, and land use information. In the soil erosion patch landing system based on weak constraints of land use information, data fusion technology is used to calculate the soil erosion modulus. By fusing different types of factors, including rainfall erosion factors, soil erodibility factors, slope factors, slope length factors, vegetation cover factors, soil and water conservation engineering measures factors, and tillage measures factors, the intensity and distribution of soil erosion can be obtained more comprehensively, thereby ensuring that the generated soil erosion patches can accurately reflect the actual erosion situation.

[0003] Pixel unmixing is a technique for extracting single categories from mixed pixels, aiming to solve the information loss problem caused by pixel mixing in remote sensing data. In the soil erosion patch landing system based on weak constraints of land use information, pixel unmixing is applied to calculate the erosion modulus. By unmixing the pixels in the remote sensing image, independent signals representing various land uses and soil erosion conditions are extracted, thereby accurately obtaining physical parameters related to soil erosion, and thus ensuring the accuracy and reliability of erosion modulus calculation.

[0004] Time series classification technology is a technique for pattern recognition and classification based on time series data. It aims to solve the problem of analyzing the change patterns of soil erosion patches in different time periods. In the soil erosion patch landing system based on weak constraints of land use information, time series classification technology is used to analyze the temporal changes of soil erosion patches. By classifying and analyzing the dynamic monitoring data of soil erosion patches over the years, the system can identify the long-term trend of soil erosion and distinguish between stable and changing areas, thus providing a scientific basis for soil erosion control.

[0005] Existing soil erosion mapping systems based on weak constraints of land use information suffer from several drawbacks. First, the accuracy of soil erosion maps is limited by a single data source, leading to inaccurate calculations of soil erosion moduli and affecting the effectiveness of soil erosion monitoring. Second, due to pixel mixing issues in remote sensing data, it is impossible to effectively extract individual land use and soil erosion information, resulting in blurred map boundaries and misjudgments of erosion intensity. Finally, in the analysis of soil erosion map changes, there is a lack of dynamic analysis based on time series, making it impossible to accurately capture long-term trends and resulting in insufficient assessment of map stability, which affects the guidance for soil erosion control measures. Summary of the Invention

[0006] The purpose of this invention is to provide a soil erosion patch mapping system based on weak constraints of land use information. This addresses the problems mentioned in the background section, such as: firstly, the accuracy of existing soil erosion patch mapping systems based on weak constraints of land use information is limited by a single data source, leading to inaccurate calculation of soil erosion modulus and affecting the effectiveness of soil erosion monitoring; secondly, due to pixel mixing issues in remote sensing data, it is impossible to effectively extract single land use and soil erosion information, resulting in blurred patch boundaries and misjudgments of erosion intensity; and finally, in the analysis of soil erosion patch changes, the lack of time-series-based dynamic analysis makes it impossible to accurately capture long-term trends, leading to insufficient stability assessment of patches and affecting the guidance of soil erosion control measures.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a soil erosion parcel mapping system based on weak constraints of land use information, comprising a land use information collection module, a factor evaluation and analysis module, a time series analysis and parcel mapping module, and a data integration and visualization module. The system is characterized in that: the land use information collection module is used to collect and integrate land use-related data to ensure accurate basic land use information for subsequent analysis; the factor evaluation and analysis module includes an erosion modulus calculation unit and a factor classification and grading unit; the erosion modulus calculation unit proposes an erosion modulus calculation algorithm based on data fusion and pixel unmixing to accurately calculate the soil erosion modulus of each parcel, ensuring that the calculation results of erosion intensity can truly reflect the soil erosion intensity of different plots. The soil erosion severity assessment module includes a factor classification and grading unit for classifying and grading various influencing factors to ensure accurate mapping of soil erosion factors to corresponding landforms. The time-series analysis and landform mapping module comprises a soil erosion change analysis unit and a soil erosion landform generation unit. The soil erosion change analysis unit proposes a time-series classification-based soil erosion landform change analysis algorithm to analyze landform changes, ensuring accurate capture of landform change trends. The soil erosion landform generation unit generates soil erosion landforms based on the analysis results, ensuring accurate landform mapping and clear boundaries. The data integration and visualization module integrates and visualizes the processed data, ensuring users can intuitively obtain soil erosion analysis results and governance suggestions.

[0008] Preferably, the land use information collection module collects and processes land use data, including land type, plot information, and land use data, to ensure that accurate and efficient basic land use data is provided for subsequent generation and analysis of soil erosion patches.

[0009] Preferably, the factor evaluation and analysis module includes an erosion modulus calculation unit. The erosion modulus calculation unit proposes an erosion modulus calculation algorithm based on data fusion and pixel unmixing. By combining remote sensing data and land use information from different sources, pixel unmixing technology is used to improve data resolution, and the calculation accuracy of each factor is optimized through data fusion methods to ensure that the soil erosion modulus of each patch can be accurately evaluated, providing accurate basic data support for the effective management of soil and water loss.

[0010] Preferably, the erosion modulus calculation algorithm based on data fusion and pixel unmixing is as follows: First, data fusion is used to eliminate the uncertainty of a single data source and improve the reliability of the algorithm input parameters. A pixel-level weighted data fusion algorithm is constructed, which spatially aligns and fuses the band reflectance of multispectral remote sensing images with the elevation information of the digital elevation model (DEM). Land use vector data is introduced as a weak constraint condition, and the fusion weight is optimized through spatial overlay analysis. The specific formula is expressed as follows:

[0011]

[0012] Among them, F i,j w is represented as the comprehensive feature value obtained after fusing multi-source data at pixel position (i,j). k Let D represent the contribution of the k-th data source in the fusion process. k,i,j Let represent the specific value at pixel position (i,j) of the k-th data source. Data sources include different band data from remote sensing imagery and DEM elevation information data. 'i' represents the row index of the pixel in space, used to locate the pixel's vertical position in the 2D raster data; 'j' represents the column index of the pixel in space, used to locate the pixel's horizontal position in the 2D raster data; 'n' represents the total number of data sources; 'k' represents the index of the number of data sources, used to distinguish different data sources; 'α' represents a weak constraint adjustment factor used to adjust the contribution of land use information to the fusion result; L i,j This represents the constraint effect of land use type at pixel location (i,j) on the fusion result. By using land use information to weaken unreasonable areas, the fused data can retain spectral and topographic features. Then, pixel demixing technology is used to separate the proportions of vegetation, soil, and water endmembers in the mixed pixels, accurately extracting vegetation cover. Combined with the fused DEM data, slope and slope length are calculated, providing dynamic parameters for the erosion modulus model. The specific formula for vegetation cover demixing is expressed as:

[0013]

[0014] Among them, V i,j The vegetation cover, R, is represented by the pixel (i,j). b,i,j S represents the reflectance of band b at pixel (i,j). b V is expressed as the standard reflectance of the soil end-member in band b. b Let m represent the standard reflectance of the vegetation endmember in band b, and m represent the total number of bands. The specific formula for calculating the improved nine-point slope length is as follows:

[0015]

[0016] Among them, L new The slope length at the center of the pixel, h z Let h represent the elevation value of the z-th pixel surrounding the pixel center, where z represents the index of the number of pixels surrounding the pixel center. center Represented as the elevation value of the center pixel. Let θ represent the horizontal distance between the z-th pixel and the center pixel, and let cos be the cosine function. center The slope angle, represented by the center pixel, is calculated using the following formula:

[0017]

[0018] Among them, S new represents the improved slope factor of the pixel center point, represents the elevation gradient, which is calculated from the elevation difference between the central pixel and the surrounding pixels. Arctan represents the arctangent function. The endmember selection is constrained by land use types to avoid the interference of bare land and building areas on the unmixing results. Secondly, based on the framework of the Universal Soil Loss Equation (USLE) model, combined with the dynamic parameters of data fusion and pixel unmixing, the pixel-by-pixel calculation of the erosion modulus is realized, and the model coefficients are optimized using land use information. The erosion modulus formula is defined, a dynamic adjustment factor is introduced, and the erosion inhibition factor is adjusted according to land use types. The specific calculation formula is expressed as:

[0019] E = β·R·K·L new ·S new (1 - V)·P land

[0020] Among them, E represents the erosion modulus, β represents the dynamic calibration coefficient, R represents the rainfall erosivity factor, K represents the soil erodibility factor, V represents the vegetation coverage, and P land represents the land use inhibition factor, and P land is directly associated with the weak constraint of land use information. Through the dynamic assignment of vector plot attributes, the spatial differential correction of the erosion modulus is realized. Finally, the erosion modulus raster map is classified according to national standards and converted into vector polygons. The spatial continuity of the polygons is optimized in combination with the land use boundary to meet management requirements. The specific formula for the vector polygon optimization rule is expressed as:

[0021] If A patch < T and L land = cultivated land, then it is merged into the adjacent forest land

[0022] Among them, A patch represents the area of the vector polygon, T represents the area threshold, and L land represents the dominant land use type of the polygon. Through the land use boundary constraint vector generalization process, it is ensured that the erosion polygons are consistent with the actual plot boundaries of farmland and forest land, improving the implementation of the results.

[0023] Preferably, the factor evaluation and analysis module includes a factor classification and grading unit. The factor classification and grading unit classifies and grades various factors, including rainfall erosion factors, soil erodibility factors, slope factors, slope length factors, vegetation coverage factors, soil and water conservation engineering measure factors, and tillage measure factors, to ensure that the factors related to soil erosion are accurately mapped onto the plot polygons for subsequent polygon analysis and processing.

[0024] Preferably, the time-series analysis and landfill module includes a soil erosion change analysis unit. This unit proposes a soil erosion landfill change analysis algorithm based on time-series classification. By analyzing the time-series data of soil erosion landfills over the years, it identifies the change patterns and trends of the landfills. Combined with land use change data, it ensures that the dynamic changes and stability of soil erosion landfills can be accurately judged, providing a reliable basis for long-term monitoring and control of soil erosion.

[0025] Preferably, the algorithm for analyzing soil erosion patch changes based on time series classification is as follows: First, by separating the trend term and seasonal term in the time series, the long-term variation pattern of soil erosion is analyzed, providing a basis for the detection of change points. The specific formula is expressed as follows:

[0026]

[0027] Where T(t) represents the global trend of the time series, ε represents the basic intercept, η represents the linear slope, and δ x Let τ represent the trend change magnitude at the x-th change point, t represent the time variable, X represent the total number of change points detected in the time series, and τ represent the total number of change points detected in the time series. x Let t represent the time position of the x-th change point, indicating the point at which the trend abruptly changes. I represents the indicator function, used to introduce a change in trend after the change point. This function is applied when time t is greater than or equal to the change point τ. x When the function is active, its value is 1; otherwise, it is 0. The specific formula for the seasonal model is as follows:

[0028]

[0029] Where S(t) represents the seasonal component of the time series at time t, representing the mathematical expression of the periodic changes in the time series; n represents the order of the Fourier series, used to describe the complexity of the seasonal changes; N represents the highest order of the Fourier series; and a n Let b be the coefficient of the nth cosine term, used to describe the magnitude of the cosine component in seasonal variations. n Represented as the coefficient of the nth-order sine term, used to describe the amplitude of the sinusoidal component in seasonal variations. Represented as the nth cosine term, it is used to capture cosine fluctuations in seasonal variations. Represented as the nth-order sine term, it is used to capture sinusoidal fluctuations in seasonal variations. P represents the seasonal cycle, and the change point τ is adjusted according to land use type using a trend model. x The detection threshold is set to ensure that the change points are related to soil erosion. This is achieved through a seasonal model: adjusting the Fourier coefficient α based on land use type. n and b nTo suppress seasonal noise unrelated to soil erosion, abrupt changes in soil erosion in the time series are identified and segmented into stable subsequences, providing independent analytical units for classification. The specific formula is as follows:

[0030] R(t) = V fused (t)-[T(t)+S(t)]

[0031] Where R(t) represents the residual value at time t, representing the difference between the observed value and the model prediction value, represents the pixel value after multi-source data fusion at time t, T(t) represents the trend term at time t, representing the long-term trend of the time series, and S(t) represents the seasonal term at time t, representing the periodic change in the time series. The specific formula for determining the point of change is as follows:

[0032] BIC = -2ln(L) + ζln(T)

[0033] Where BIC represents the Bayesian Information Criterion, a statistic used to evaluate the goodness of fit of a model; ln(·) represents the logarithmic function; L represents the likelihood function, representing the degree to which the algorithm fits the observed data; ζ represents the number of model parameters; and T represents the total length of the time series.

[0034] The difference between observed values ​​and model predictions at each time point is calculated using residual analysis to identify outliers and points of change. The determination of points of change is then performed based on the results of residual analysis. The rationality of different assumptions regarding points of change is evaluated using the Bayesian Information Criterion (BIC), and the optimal number and location of points of change are selected. ln(·) represents the natural logarithm function. Next, pixel unmixing and patch classification address the problem of mixed pixels, accurately distinguishing soil erosion patches from background features and improving the accuracy of classification boundaries. The specific formula for linear unmixing is expressed as:

[0035]

[0036] Among them, V mixed (t) represents the blended pixel value at time t, f c V represents the abundance of land cover type c, indicating the proportion of land cover type c in the mixed pixels. c (t) represents the pure spectral time series value of the c-th land cover at time t, ∈ represents the noise term, representing the residual part that the algorithm failed to explain, and C represents the total number of land cover categories. The specific formula for dynamic time curvature classification is as follows:

[0037]

[0038] Where DWB(O,R) represents the dynamic time warping distance between time series O and R, O represents the time series to be classified, R represents the reference time series, w represents the alignment path, indicating the optimal alignment between time series O and R, U represents the total length of the alignment path, and u represents the length index of the alignment path. This is represented by the value of O corresponding to the u-th point in the alignment path w. This is represented by the value of R corresponding to the u-th point in the alignment path w. This is expressed as minimizing the total distance of the alignment path w. By decomposing the spectral features of mixed pixels, the abundance of each land cover type is obtained, solving the mixed pixel problem. Through nonlinear alignment of time series, the similarity between the sequence to be classified and the reference sequence is measured to achieve high-precision classification. Finally, prior land use information is fused to optimize patch boundaries and generate high-precision soil erosion maps. The specific formula for spatial probability fusion is expressed as:

[0039] P final (x,y)=λ·P class (x,y)+(1-λ)·P landuse (x,y)

[0040] Among them, P final (x,y) represents the final probability value at position (x,y), indicating the probability that position (x,y) belongs to a soil erosion patch. P class (x,y) represents the probability of soil erosion at location (x,y) obtained through dynamic time warp classification, P landuse (x,y) represents the prior probability at location (x,y) based on historical land use data, λ represents the fusion weight, and the specific formula for level set boundary optimization is as follows:

[0041]

[0042] Where, φ t (x,y) represents the level set function value at position (x,y) at time t, φ t+1 (x,y) represents the level set function value at position (x,y) at time t+1, and η represents the evolution rate parameter, which controls the step size for updating the level set function. The curvature term, expressed as a level set function, represents the change in boundary curvature. The gradient of a level set function represents the rate of change of the function values ​​across space. P represents the magnitude of the gradient of the level set function, κ represents the weight parameter used to control the influence of the probability term on the boundary evolution. finalIt is represented as the final probability value at location (x,y). Data fusion improves the spatiotemporal continuity, pixel demixing solves the problem of mixed pixels, and the weak constraints of land use are used to suppress false detections through probability fusion, ensuring that the map boundary is consistent with the actual situation. Combined with temporal analysis and spatial optimization, the accurate landing of soil erosion maps is achieved.

[0043] Preferably, the time series analysis and land parcel mapping module includes a soil erosion land parcel generation unit. Based on the analysis results of the soil erosion change analysis unit, the soil erosion land parcel generation unit generates accurate soil erosion land parcels by combining different soil erosion factors with land use information, and performs smoothing and merging processing on the land parcels to ensure that the generated land parcels have clear boundaries and conform to the actual distribution of soil erosion.

[0044] Preferably, the data integration and visualization module integrates all generated data and analysis results for visualization, ensuring that users can intuitively view the soil erosion status, erosion intensity, and changes in landforms, thereby providing effective decision support and helping to achieve precise soil erosion control measures.

[0045] Compared with the prior art, the beneficial effects of the present invention are:

[0046] 1. The erosion modulus calculation unit proposes an erosion modulus calculation algorithm based on data fusion and pixel unmixing. First, the algorithm utilizes data fusion technology to effectively integrate band reflectance information from multispectral remote sensing imagery with elevation information from the Digital Elevation Model (DEM). It also incorporates land use vector data as a weak constraint to optimize the fusion weights, eliminating errors and uncertainties from a single data source. Through pixel-level weighted data fusion, it ensures spatial matching and data consistency of various factors, enabling the erosion modulus calculation to accurately reflect the soil and water loss characteristics of different land types. Furthermore, the introduction of land use information as a spatial constraint not only enhances the physical meaning of data fusion but also... Furthermore, this method effectively avoids category confusion caused by spectral mixing effects in remote sensing images, thereby improving the reliability of erosion modulus calculation. The fused data retains both the spectral information of the remote sensing image and the topographic features of the DEM, enabling soil erosion modulus calculation to be optimized based on more complete geographic environmental information, improving the spatial continuity and realism of the calculation results. Secondly, the algorithm uses pixel demixing technology to perform deep analysis on the remote sensing image, accurately separating the proportions of vegetation, soil, and water endmembers in the mixed pixels, extracting high-precision vegetation cover information, and combining it with DEM data to calculate slope and slope length, thereby optimizing the soil erosion modulus calculation model. The pixel unmixing process effectively reduces errors caused by pixel mixing in remote sensing data, making the calculation of vegetation cover factors more accurate. Especially in areas with many mixed pixels, this technology effectively improves the resolution accuracy of vegetation cover information. When calculating slope length and gradient using DEM, an improved nine-point slope length calculation method is adopted to enhance the accuracy of the slope length factor calculation, making the slope information more consistent with actual terrain conditions and providing more refined parameter support for soil erosion intensity assessment. Furthermore, during pixel unmixing, the constraint of land use type effectively eliminates the interference of bare land and built-up areas on vegetation cover calculation, avoiding the misclassification of non-vegetated areas. Misclassified as high vegetation cover areas, this algorithm further improves the reliability of erosion modulus calculation. Finally, based on the dynamic parameters obtained from data fusion and pixel unmixing, the algorithm realizes pixel-by-pixel erosion modulus calculation within the framework of the Universal Soil Loss Equation (USLE) model. It also optimizes model coefficients by combining land use information and introduces dynamic adjustment factors, enabling the erosion modulus to adaptively adjust with changes in spatial environment and land type. To meet management and application needs, the algorithm also optimizes the vectorized erosion patches through land use boundaries, improving the spatial consistency of soil erosion patches and making their boundaries more consistent with the actual land parcel morphology, thus enhancing the feasibility and applicability of the results.

[0047] 2. The Soil and Water Erosion Change Analysis Unit proposes a soil and water erosion patch change analysis algorithm based on time series classification. First, the algorithm analyzes the long-term variation patterns of soil and water erosion by separating the trend and seasonal components in the time series, providing a solid theoretical foundation for change point detection. In soil and water erosion monitoring, time series data is often affected by long-term trends and seasonal changes. Directly using raw data for change analysis is easily subject to noise interference. By constructing a trend model, the algorithm can identify long-term trends and dynamically adjust the threshold for change point detection based on land use information, making the detection results more consistent with the actual soil and water erosion situation. Simultaneously, the seasonal model suppresses seasonal noise unrelated to soil and water erosion through Fourier decomposition, ensuring the reliability of the analysis results. Second, the algorithm identifies abrupt change points in the time series and divides the time series into stable subsequences, providing independent analysis units for subsequent classification. In this process, residual analysis is used to calculate the deviation between the observed value and the model prediction value at each time point, combined with Bayesian decomposition. The BIC (Browser Information Criterion) assesses the rationality of different change point assumptions and ultimately determines the optimal number and location of change points. This method enables the system to more accurately determine the dynamic evolution of soil erosion patches, distinguishing between long-term stable areas and areas susceptible to erosion, thus providing refined data support for soil and water conservation. Secondly, the algorithm further improves the spatial accuracy of soil erosion patches by combining pixel demixing and patch classification techniques. The algorithm utilizes pixel demixing to decompose the spectral features of mixed pixels and extract the abundance information of different land cover categories, making the components of each pixel more clearly distinguishable. Based on this, the Dynamic Time Warp (DTW) classification method is used to measure the similarity between the time series to be classified and the reference time series for accurate classification. DTW classification can non-linearly align time series, effectively identifying soil erosion change patterns even with time shifts and morphological changes. Through this method, the system can accurately distinguish soil erosion patches from background land cover, significantly improving the accuracy of classification boundaries. Furthermore, the algorithm utilizes prior land use information to optimize patch boundaries, thereby improving the spatial consistency of classification results. By constructing a spatial probability fusion model, it weights and fuses the soil erosion probability obtained from dynamic time curvature classification with the prior probability based on historical land use data, thus reducing the false detection rate and improving recognition accuracy. Finally, the level set method is used to optimize patch boundaries. Through curvature constraints and spatial probability fusion, the boundaries of soil erosion patches are made to better reflect the actual erosion area morphology. Overall, the soil erosion patch change analysis algorithm based on time series classification not only improves the accuracy of soil erosion monitoring in the time dimension but also enhances the consistency of patch boundaries by combining spatial optimization techniques, achieving precise mapping of soil erosion patches. Attached Figure Description

[0048] Figure 1 This is a schematic diagram of the structure of the present invention; Detailed Implementation

[0049] 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.

[0050] Please see Figure 1 This invention provides a soil erosion mapping system based on weak constraints of land use information, comprising a land use information collection module, a factor evaluation and analysis module, a time series analysis and mapping module, and a data integration and visualization module. The key features are: the land use information collection module collects and integrates land use-related data to ensure accurate basic land use information for subsequent analysis; the factor evaluation and analysis module includes an erosion modulus calculation unit and a factor classification and grading unit; the erosion modulus calculation unit proposes an erosion modulus calculation algorithm based on data fusion and pixel unmixing to accurately calculate the soil erosion modulus of each mapping patch, ensuring that the calculated erosion intensity truly reflects the soil erosion extent of different plots. The degree, factor classification and grading unit is used to classify and grade various influencing factors to ensure that soil erosion factors are accurately mapped to corresponding patches; the time series analysis and patch mapping module includes a soil erosion change analysis unit and a soil erosion patch generation unit. The soil erosion change analysis unit proposes a soil erosion patch change analysis algorithm based on time series classification to analyze patch changes and ensure accurate capture of patch change trends. The soil erosion patch generation unit is used to generate soil erosion patches based on the analysis results to ensure the accuracy and clear boundaries of patch mapping; the data integration and visualization module is used to integrate and visualize the processed data to ensure that users can intuitively obtain soil erosion analysis results and governance suggestions.

[0051] See Figure 1 Furthermore, the land use information collection module collects and processes land use data, including land type, plot information, and land use data, to ensure that accurate and efficient basic land use data is provided for subsequent generation and analysis of soil erosion patches.

[0052] See Figure 1Furthermore, the factor evaluation and analysis module includes an erosion modulus calculation unit. This unit proposes an erosion modulus calculation algorithm based on data fusion and pixel unmixing. By combining remote sensing data and land use information from different sources, pixel unmixing technology is used to improve data resolution, and data fusion methods are used to optimize the calculation accuracy of each factor. This ensures that the soil erosion modulus of each patch can be accurately evaluated, providing precise basic data support for the effective management of soil and water conservation.

[0053] See Figure 1 Furthermore, the erosion modulus calculation algorithm based on data fusion and pixel unmixing is as follows: First, data fusion is used to eliminate the uncertainty of a single data source and improve the reliability of the algorithm input parameters. A pixel-level weighted data fusion algorithm is constructed, which spatially aligns and fuses the band reflectance of multispectral remote sensing images with the elevation information of the digital elevation model (DEM). Land use vector data is introduced as a weak constraint condition, and the fusion weight is optimized through spatial overlay analysis. The specific formula is expressed as follows:

[0054]

[0055] Among them, F i,j w is represented as the comprehensive feature value obtained after fusing multi-source data at pixel position (i,j). k Let D represent the contribution of the k-th data source in the fusion process. k,i,j Let represent the specific value at pixel position (i,j) of the k-th data source. Data sources include different band data from remote sensing imagery and DEM elevation information data. 'i' represents the row index of the pixel in space, used to locate the pixel's vertical position in the 2D raster data; 'j' represents the column index of the pixel in space, used to locate the pixel's horizontal position in the 2D raster data; 'n' represents the total number of data sources; 'k' represents the index of the number of data sources, used to distinguish different data sources; 'α' represents a weak constraint adjustment factor used to adjust the contribution of land use information to the fusion result; L i,j This represents the constraint effect of land use type at pixel location (i,j) on the fusion result. By using land use information to weaken unreasonable areas, the fused data can retain spectral and topographic features. Then, pixel demixing technology is used to separate the proportions of vegetation, soil, and water endmembers in the mixed pixels, accurately extracting vegetation cover. Combined with the fused DEM data, slope and slope length are calculated, providing dynamic parameters for the erosion modulus model. The specific formula for vegetation cover demixing is expressed as:

[0056]

[0057] Among them, V i,j The vegetation cover, R, is represented by the pixel (i,j). b,i,jS represents the reflectance of band b at pixel (i,j). b V is expressed as the standard reflectance of the soil end-member in band b. b Let m represent the standard reflectance of the vegetation endmember in band b, and m represent the total number of bands. The specific formula for calculating the improved nine-point slope length is as follows:

[0058]

[0059] Among them, L new The slope length at the center of the pixel, h z Let h represent the elevation value of the z-th pixel surrounding the pixel center, where z represents the index of the number of pixels surrounding the pixel center. center Represented as the elevation value of the center pixel. Let θ represent the horizontal distance between the z-th pixel and the center pixel, and let cos be the cosine function. center The slope angle, represented by the center pixel, is calculated using the following formula:

[0060]

[0061] Among them, S new Represented as the improved slope factor at the pixel center point. The elevation gradient is calculated by the elevation difference between the center pixel and surrounding pixels. Arctan represents the arctangent function. Endmember selection is constrained by land use type to avoid interference from bare land and built-up areas in the unmixing results. Secondly, based on the USLE model framework (a general soil loss equation), and combined with dynamic parameters from data fusion and pixel unmixing, the erosion modulus is calculated pixel-by-pixel. Land use information is used to optimize the model coefficients, defining the erosion modulus formula. A dynamic adjustment factor is introduced, and the erosion inhibition factor is adjusted according to land use type. The specific calculation formula is as follows:

[0062] E=β·R·K·L new ·S new ·(1-V)·P land

[0063] Where E represents the erosion modulus, β represents the dynamic calibration coefficient, R represents the rainfall erosivity factor, K represents the soil erodibility factor, V represents the vegetation cover, and P... land Represented as a land use inhibition factor, P land By directly associating with weak constraints on land use information and dynamically assigning vector plot attributes, spatial differentiation correction of erosion modulus is achieved. Finally, the erosion modulus raster map is classified according to national standards and converted into vector plots. The spatial continuity of the plots is optimized in conjunction with land use boundaries to meet management requirements. The specific formula for the vector plot optimization rule is expressed as follows:

[0064] If A patch <T and L land = cultivated land, then merge it into the adjacent forest land

[0065] Among them, A patch represents the vector patch area, T represents the area threshold, and L land represents the dominant land use type of the patch. By constraining the vector generalization process through the land use boundary, ensure that the erosion patches are consistent with the actual plot boundaries of farmland and forest land, and improve the implementation of the results.

[0066] Refer to Figure 1 , further, the factor evaluation and analysis module includes a factor classification and grading unit. The factor classification and grading unit classifies and grades various factors, including rainfall erosion factors, soil erodibility factors, slope factors, slope length factors, vegetation cover factors, soil and water conservation engineering measure factors, and tillage measure factors, to ensure that the factors related to soil erosion are accurately mapped to the plot patches for subsequent patch analysis and processing.

[0067] Refer to Figure 1 , further, the time series analysis and patch implementation module includes a soil and water loss change analysis unit. The soil and water loss change analysis unit proposes a soil and water loss patch change analysis algorithm based on time series classification. By analyzing the time series data of soil and water loss patches over the years, identify the change patterns and trends of the patches, and combine with the land use change situation to ensure that the dynamic changes and stability of soil and water loss patches can be accurately judged, providing a reliable basis for the long-term monitoring and treatment of soil and water loss.

[0068] Refer to Figure 1 , further, the soil and water loss patch change analysis algorithm based on time series classification is specifically as follows: First, by separating the trend term and seasonal term in the time series, analyze the long-term change law of soil and water loss, providing a basis for change point detection. The specific formula is expressed as:

[0069]

[0070] Among them, T(t) represents the global trend of the time series, ε represents the basic intercept, η represents the linear slope, and δ x represents the trend change amplitude of the xth change point, t represents the time variable, X represents the total number of change points detected in the time series, and τ x represents the time position of the xth change point, representing the time point when the trend changes suddenly. I represents the indicator function. The I function is used to introduce a change in the trend after the change point. When the time t is greater than or equal to the change point τ x , the function value is 1, otherwise it is 0. The specific formula of the seasonal model is expressed as:

[0071]

[0072] Where S(t) represents the seasonal component of the time series at time t, representing the mathematical expression of the periodic changes in the time series; n represents the order of the Fourier series, used to describe the complexity of the seasonal changes; N represents the highest order of the Fourier series; and a n Let b be the coefficient of the nth cosine term, used to describe the magnitude of the cosine component in seasonal variations. n Represented as the coefficient of the nth-order sine term, used to describe the amplitude of the sinusoidal component in seasonal variations. Represented as the nth cosine term, it is used to capture cosine fluctuations in seasonal variations. Represented as the nth-order sine term, it is used to capture sinusoidal fluctuations in seasonal variations. P represents the seasonal cycle, and the change point τ is adjusted according to land use type using a trend model. x The detection threshold is set to ensure that the change points are related to soil erosion. This is achieved through a seasonal model: adjusting the Fourier coefficient α based on land use type. n and b n To suppress seasonal noise unrelated to soil erosion, abrupt changes in soil erosion in the time series are identified and segmented into stable subsequences, providing independent analytical units for classification. The specific formula is as follows:

[0073] R(t) = V fused (t)-[T(t)+S(t)]

[0074] Where R(t) represents the residual value at time t, representing the difference between the observed value and the model prediction value, represents the pixel value after multi-source data fusion at time t, T(t) represents the trend term at time t, representing the long-term trend of the time series, and S(t) represents the seasonal term at time t, representing the periodic change in the time series. The specific formula for determining the point of change is as follows:

[0075] BIC = -2ln(L) + ζln(T)

[0076] Where BIC represents the Bayesian Information Criterion, a statistic used to evaluate the goodness of fit of a model; ln(·) represents the logarithmic function; L represents the likelihood function, representing the degree to which the algorithm fits the observed data; ζ represents the number of model parameters; and T represents the total length of the time series.

[0077] The difference between observed values ​​and model predictions at each time point is calculated using residual analysis to identify outliers and points of change. The determination of points of change is then performed based on the results of residual analysis. The rationality of different assumptions regarding points of change is evaluated using the Bayesian Information Criterion (BIC), and the optimal number and location of points of change are selected. ln(·) represents the natural logarithm function. Next, pixel unmixing and patch classification address the problem of mixed pixels, accurately distinguishing soil erosion patches from background features and improving the accuracy of classification boundaries. The specific formula for linear unmixing is expressed as:

[0078]

[0079] Among them, V mixed (t) represents the blended pixel value at time t, f c V represents the abundance of land cover type c, indicating the proportion of land cover type c in the mixed pixels. c (t) represents the pure spectral time series value of the c-th land cover at time t, ε represents the noise term, representing the residual part that the algorithm failed to explain, and C represents the total number of land cover categories. The specific formula for dynamic time curvature classification is as follows:

[0080]

[0081] Where DWB(O,R) represents the dynamic time warping distance between time series O and R, O represents the time series to be classified, R represents the reference time series, w represents the alignment path, indicating the optimal alignment between time series O and R, U represents the total length of the alignment path, and u represents the length index of the alignment path. This is represented by the value of O corresponding to the u-th point in the alignment path w. This is represented by the value of R corresponding to the u-th point in the alignment path w. This is expressed as minimizing the total distance of the alignment path w. By decomposing the spectral features of mixed pixels, the abundance of each land cover type is obtained, solving the mixed pixel problem. Through nonlinear alignment of time series, the similarity between the sequence to be classified and the reference sequence is measured to achieve high-precision classification. Finally, prior land use information is fused to optimize patch boundaries and generate high-precision soil erosion maps. The specific formula for spatial probability fusion is expressed as:

[0082] P final (x,y)=λ·P class (x,y)+(1-λ)·P landuse (x,y)

[0083] Among them, P final (x,y) represents the final probability value at position (x,y), indicating the probability that position (x,y) belongs to a soil erosion patch. P class(x,y) represents the probability of soil erosion at location (x,y) obtained through dynamic time warp classification, P landuse (x,y) represents the prior probability at location (x,y) based on historical land use data, λ represents the fusion weight, and the specific formula for level set boundary optimization is as follows:

[0084]

[0085] Where, φ t (x,y) represents the level set function value at position (x,y) at time t, φ t+1 (x,y) represents the level set function value at position (x,y) at time t+1, and η represents the evolution rate parameter, which controls the step size for updating the level set function. The curvature term, expressed as a level set function, represents the change in boundary curvature. The gradient of a level set function represents the rate of change of the function values ​​across space. P represents the magnitude of the gradient of the level set function, κ represents the weight parameter used to control the influence of the probability term on the boundary evolution. final It is represented as the final probability value at location (x,y). Data fusion improves the spatiotemporal continuity, pixel demixing solves the problem of mixed pixels, and the weak constraints of land use are used to suppress false detections through probability fusion, ensuring that the map boundary is consistent with the actual situation. Combined with temporal analysis and spatial optimization, the accurate landing of soil erosion maps is achieved.

[0086] See Figure 1 Furthermore, the time-series analysis and landfill module includes a soil erosion landfill generation unit. Based on the analysis results of the soil erosion change analysis unit, the soil erosion landfill generation unit generates accurate soil erosion landfills by combining different soil erosion factors with land use information, and performs smoothing and merging processing on the landfills to ensure that the generated landfill boundaries are clear and conform to the actual distribution of soil erosion.

[0087] See Figure 1 Furthermore, the data integration and visualization module integrates all generated data and analysis results for visualization, ensuring that users can intuitively view the soil erosion status, erosion intensity, and changes in landforms, thereby providing effective decision support and helping to achieve precise soil erosion control measures.

[0088] In practical use, firstly, the land use information collection module is used to collect and integrate land use-related data to ensure accurate basic land use information for subsequent analysis. Then, the factor evaluation and analysis module includes an erosion modulus calculation unit and a factor classification and grading unit. The erosion modulus calculation unit proposes an erosion modulus calculation algorithm based on data fusion and pixel unmixing to accurately calculate the soil erosion modulus of each patch, ensuring that the calculation results of erosion intensity can truly reflect the degree of soil and water loss in different plots. The factor classification and grading unit is used to classify and grade various influencing factors, ensuring that soil and water loss factors are accurately reflected. The system first projects the data onto the corresponding map patches. Secondly, the time-series analysis and map patch implementation module includes a soil erosion change analysis unit and a soil erosion map patch generation unit. The soil erosion change analysis unit proposes a time-series classification-based soil erosion map patch change analysis algorithm to analyze map patch changes, ensuring accurate capture of map patch change trends. The soil erosion map patch generation unit generates soil erosion maps based on the analysis results, ensuring accurate map patch implementation and clear boundaries. Finally, the data integration and visualization module integrates and visualizes the processed data, ensuring users can intuitively obtain soil erosion analysis results and governance suggestions.

[0089] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A soil erosion map patch landing system based on weak constraints of land use information, comprising a land use information acquisition module, a factor evaluation and analysis module, a time series analysis and map patch landing module, a data integration and visualization module, characterized in that: The land use information collection module is used for collecting and integrating land use related data, ensuring that accurate land use basic information is provided for subsequent analysis; the factor evaluation and analysis module includes an erosion modulus calculation unit and a factor classification and grading unit, the erosion modulus calculation unit proposes an erosion modulus calculation algorithm based on data fusion and pixel unmixing, which is used for accurately calculating the soil erosion modulus of each polygon, ensuring that the calculation result of the erosion intensity can truly reflect the degree of water and soil loss of different plots, and the factor classification and grading unit is used for classifying and grading each influencing factor, ensuring that the water and soil loss factors are accurately mapped to the corresponding polygons; the time series analysis and polygon landing module includes a water and soil loss change analysis unit and a water and soil loss polygon generation unit, the water and soil loss change analysis unit proposes a water and soil loss polygon change analysis algorithm based on time series classification to analyze the polygon change, ensuring accurate capture of the polygon change trend, and the water and soil loss polygon generation unit is used for generating water and soil loss polygons according to the analysis result, ensuring the accuracy and clear boundary of polygon landing; the data integration and visualization module is used for integrating and visualizing the processed data, ensuring that users can intuitively obtain the water and soil loss analysis result and treatment suggestion.

2. The system according to claim 1, wherein the system is characterized by: The land use information collection module collects and processes land use data, including land type, plot information and land use data, to provide accurate and efficient land use basic data for subsequent water and soil loss polygon generation and analysis.

3. The system according to claim 1, wherein the system is characterized by: The factor evaluation and analysis module includes an erosion modulus calculation unit and a factor classification and grading unit, the erosion modulus calculation unit proposes an erosion modulus calculation algorithm based on data fusion and pixel unmixing, which combines remote sensing data from different sources and land use information, uses pixel unmixing technology to improve data resolution, and uses data fusion method to optimize the calculation accuracy of each factor, ensuring that the soil erosion modulus of each polygon can be accurately evaluated, providing accurate basic data support for effective management of water and soil loss; the factor classification and grading unit classifies and grades various factors, including rainfall erosion factor, soil erodibility factor, slope factor, slope length factor, vegetation cover factor, water and soil conservation engineering measure factor and tillage measure factor, ensuring that soil erosion related factors are accurately mapped to the plot polygons, facilitating subsequent polygon analysis and processing.

4. The system according to claim 3, wherein the system is characterized by: Firstly, data fusion is used to eliminate the uncertainty of a single data source and improve the reliability of algorithm input parameters, a pixel-level weighted data fusion algorithm is constructed, the band reflectance of multispectral remote sensing image and the elevation information of digital elevation model DEM are spatially aligned and fused, land use vector data is introduced as a weak constraint condition, and the fusion weight is optimized through spatial overlay analysis, and the specific formula is represented as: where F i,j represents the integrated feature value at pixel position (i, j) after multi-source data fusion, w k represents the contribution degree of the kth data source in the fusion process, D k,i,j represents the specific value of the kth data source at pixel position (i, j), the data sources include different band data of remote sensing images and DEM elevation information data, i represents the row index of the pixel in space, used to locate the vertical position of the pixel in the two-dimensional grid data, j represents the column index of the pixel in space, used to locate the horizontal position of the pixel in the two-dimensional grid data, n represents the total number of data sources, k represents the number index of the data source, used to distinguish different data sources, a represents a weak constraint adjustment factor for adjusting the contribution degree of land use information to the fusion result, L i,j represents the constraint effect of land use type on the fusion result at pixel position (i, j), which weakens the unreasonable area through land use information, and the fused data can retain spectral features and terrain features, then, pixel unmixing technology is used to separate the vegetation, soil, and water endmember proportion in the mixed pixel, accurately extract the vegetation coverage, combine with the fused DEM data to calculate the slope and slope length, and provide dynamic parameters for the erosion modulus model, the vegetation coverage unmixing specific formula is represented as: where V i,j denotes the vegetation fraction of pixel (i,j), R b,i,j denotes the reflectance of band b at pixel (i,j), S b denotes the standard reflectance of soil endmember at band b, V b denotes the standard reflectance of vegetation endmember at band b, m denotes the total number of bands, and the improved nine-point slope length calculation is specifically represented by the formula: where L new denotes the length of the slope at the center of the pixel, h z denotes the elevation value of the z-th pixel around the center of the pixel, z denotes the index of the number of pixels around the center of the pixel, h center denotes the elevation value of the center pixel, denotes the horizontal distance between the z-th pixel and the center pixel, cos denotes the cosine function, θ center denotes the slope angle of the center pixel, and the slope calculation is specifically represented by: where S new The improved slope factor is represented as the center point of the pixel, The elevation gradient is calculated by the elevation difference between the center pixel and the surrounding pixels, and arctan represents the inverse tangent function. The endmember selection is constrained by land use type to avoid interference with the unmixing results of bare land and building areas. Secondly, based on the Universal Soil Loss Equation (USLE) model framework, combined with data fusion and dynamic parameters of pixel unmixing, the erosion modulus is calculated pixel by pixel. The model coefficients are optimized using land use information, the erosion modulus formula is defined, a dynamic adjustment factor is introduced, and the erosion inhibition factor is adjusted according to the land use type. The specific calculation formula is represented as: E = β - R - K - L new • S new • (1 - V) - P land Wherein, E represents the erosion modulus, β represents the dynamic calibration coefficient, R represents the rainfall erosivity factor, K represents the soil erodibility factor, V represents the vegetation coverage, P land represents the land use inhibition factor, P land Directly associate land use information weak constraint, through the vector block attribute dynamic assignment, realize the spatial differentiation correction of erosion modulus, finally, the erosion modulus grid map is classified according to the national standard, and is converted into a vector graph spot, combine the land use boundary optimization spot spatial continuity, meet the management demand, the specific formula of the vector graph spot optimization rule is represented as: If A patch <T and L land = cultivated land, then merge to adjacent forest land Wherein, A patch is expressed as a vector plot area, T is expressed as an area threshold, L land is expressed as a plot dominant land use type, and the vector generalization process is constrained by a land use boundary to ensure that the erosion plot is consistent with the actual plot boundary of farmland and woodland, thereby improving the landing performance of the results.

5. The system according to claim 1, wherein the system is characterized by: The time series analysis and map spot landing module includes a soil erosion change analysis unit and a soil erosion map spot generation unit, the soil erosion change analysis unit proposes a soil erosion map spot change analysis algorithm based on time series classification, identifies the change mode and trend of the map spot by analyzing the time series data of the soil erosion map spot in previous years, combines the land use change situation, ensures that the dynamic change of the soil erosion map spot and its stability can be accurately judged, and provides a reliable basis for long-term monitoring and management of soil erosion; the soil erosion map spot generation unit generates accurate soil erosion map spots by combining different soil erosion factors and land use information according to the analysis result of the soil erosion change analysis unit, and performs smoothing and merging processing on the map spots, to ensure that the generated map spot boundary is clear and consistent with the actual distribution of soil erosion.

6. The system according to claim 5, wherein the system is characterized by: First, by separating the trend item and the seasonal item in the time series, the long-term change rule of soil erosion is analyzed to provide a basis for change point detection, and the specific formula is expressed as: where T(t) represents the global trend as a time series, ε represents the base intercept, η represents the linear slope, δ x represents the trend change amplitude of the xth change point, t represents the time variable, X represents the total number of change points detected in the time series, τ x represents the time position of the xth change point, represents the time point at which the trend mutates, I represents the indicator function, the I function is used to introduce a change in the trend after a change point, when the time t is greater than or equal to the change point τ x , the function value is 1, otherwise it is 0, and the seasonal model is specifically represented by the formula: where S(t) represents the seasonal component of the time series at time t, represents the mathematical expression of the periodic change in the time series, n represents the order of the Fourier series, which is used to describe the complexity of the seasonal change, N represents the highest order of the Fourier series, a n represents the coefficient of the nth order cosine term, which is used to describe the amplitude of the cosine component in the seasonal change, n represents the coefficient of the nth order sine term, which is used to describe the amplitude of the sine component in the seasonal change, represents the nth order cosine term, which is used to capture the cosine fluctuation in the seasonal change, represents the nth order sine term, which is used to capture the sine fluctuation in the seasonal change, P represents the seasonal period, and τ x represents the detection threshold of the change point, which is adjusted according to the land use type by the trend model, to ensure that the change point is related to soil erosion, and by the seasonal model: adjust the Fourier coefficients a n and b n , suppress the seasonal noise unrelated to soil erosion, then identify the abrupt change points of soil erosion in the time series, and divide them into stable sub-sequences to provide independent analysis units for classification, which is specifically represented by the formula: R(t) = V fused (t) - [T(t) + S(t)] Wherein, R(t) represents the residual value at time t, which represents the difference between the observation value and the model prediction value, represents the pixel value after multi-source data fusion at time t, T(t) represents the trend item at time t, which represents the long-term change trend of the time series, S(t) represents the seasonal item at time t, which represents the periodic change in the time series, and the change point determination formula is expressed as: BIC=-2ln(L)+ζln(T) Wherein, BIC represents the Bayesian information criterion, which is a statistical quantity for evaluating the goodness of fit of the model, ln(·) represents the logarithmic function, L represents the likelihood function, which represents the fitting degree of the algorithm to the observation data, ζ represents the number of model parameters, and T represents the total length of the time series, The difference between the observation value and the model prediction value at each time point is calculated by residual analysis to identify abnormal points and change points, the change point determination calculation is performed based on the results of residual analysis, the rationality of different change point hypotheses is evaluated by Bayesian information criterion BIC, and the optimal number and position of change points are selected, ln(·) represents the natural logarithmic function, secondly, pixel unmixing and map spot classification are performed to solve the mixed pixel problem, accurately distinguish soil erosion map spots and background objects, and improve the classification boundary accuracy, and the linear unmixing formula is expressed as: where V mixed (t) denotes the mixed pixel value at time t, f c denotes the abundance of the cth surface feature, representing the proportion of the cth surface feature in the mixed pixel, V c (t) denotes the pure spectrum time series value of the cth surface feature at time t, ∈ denotes the noise term, representing the residual part that the algorithm fails to explain, C denotes the total number of surface feature categories, and the dynamic time warping classification specific formula is denoted as: wherein, DWB(O, R) represents the dynamic time warping distance between time series O and R, O represents the time series to be classified, R represents the reference time series, w represents the alignment path, represents the optimal alignment between time series O and R, U represents the total length of the alignment path, and u represents the length index of the alignment path, represents the value of O corresponding to the u-th point in the alignment path w, represents the value of R corresponding to the u-th point in the alignment path w, represents the total distance of the alignment path w, the abundance of each land object type is obtained by decomposing the spectral characteristics of mixed pixels, the mixed pixel problem is solved, the similarity between the sequence to be classified and the reference sequence is measured by nonlinear alignment of time series, high-precision classification is realized, finally, the land use prior information is fused to optimize the plot boundary, and high-precision soil erosion plot is generated, and the spatial probability fusion is specifically represented by the following formula: P final (x,y) = λ · P class (x,y) + (1 - λ) · P landuse (x,y) where P final (x, y) is the final probability value at position (x, y), representing the probability of position (x, y) belonging to the soil erosion map spot, P class (x, y) is the soil erosion probability at position (x, y) obtained by dynamic time warping classification, P landuse (x, y) is the prior probability at position (x, y) based on historical land use data, λ is the fusion weight, and the level set boundary optimization is specifically represented by the formula: where φ t (x, y) represents the level set function value at position (x, y) at time t, φ t+1 (x, y) represents the level set function value at position (x, y) at time t+1, η represents the evolution rate parameter, and controls the step size of the level set function update, represents the curvature term of the level set function, and represents the change of the boundary curvature, represents the gradient of the level set function, and represents the rate of change of the function value in space, represents the modulus of the level set function gradient, κ represents the weight parameter, and is used to control the influence of the probability term on the boundary evolution, P final represents the final probability value at position (x, y), improves the spatio-temporal continuity through data fusion, solves the mixed pixel problem by pixel unmixing, suppresses false positives through probability fusion with weak land use constraints, ensures that the plot boundary is consistent with the actual situation, and realizes the precise landing of soil erosion plot by combining time series analysis and spatial optimization.

7. The system according to claim 1, wherein the system is characterized by: The data integration and visualization module integrates all generated data and analysis results for visualization, ensures that users can intuitively view the soil erosion condition, erosion intensity and map spot change, and thus provides effective decision support to help realize accurate soil erosion management measures.

Images