Method, system, device and medium for analyzing rapid restoration strategy of alpine meadow

Abstract

The application relates to the technical field of policy analysis, and discloses a high-cold grassland rapid restoration policy analysis method, system, device and medium. The method comprises the following steps: performing quality evaluation and restoration on a multispectral image of a high-cold grassland to obtain a standard image set; extracting spectral texture features, performing degradation classification, obtaining a feature set and a score table; combining multi-dimensional index to evaluate the degradation degree, generating a distribution matrix and an index set; forming a restoration scheme library and a parameter list through case analysis and parameter optimization; analyzing the restoration effect in combination with environmental data to obtain an index matrix and an evaluation table; and optimizing monitoring data to obtain a strategy scheme and an adjustment parameter set. The application realizes accurate identification of the degradation conditions of high-cold grasslands of different types and in different regions, and can dynamically adjust the restoration policy according to real-time monitoring data, thereby improving the restoration effect.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of policy analysis, and particularly relates to a high-cold grassland rapid restoration strategy analysis method, system, device and medium. BACKGROUND

[0002] The current degradation problem of high-cold grassland is increasingly serious, and various restoration methods and technical means have been developed. The traditional restoration method is mainly based on field investigation and experience judgment, and the information of grassland degradation condition is obtained by setting sample plots, collecting samples and other ways, and then the restoration measures are selected according to historical experience. At the same time, the development of remote sensing technology provides a new means for large-scale grassland monitoring. Through multi-spectral remote sensing image, vegetation index, coverage and other information can be obtained, and spatial analysis can be carried out combined with geographic information system. The formulation of restoration measures usually adopts single index evaluation method or comprehensive evaluation method, and the evaluation index system is set to evaluate the degradation degree of grassland, and then the corresponding restoration scheme is selected.

[0003] However, the existing high-cold grassland restoration technology has obvious deficiencies. First, the traditional experience-based restoration scheme lacks data support and is difficult to adapt to the complex and changeable high-cold environment. Second, although remote sensing technology can obtain large-scale data, the analysis and processing of data still stay at the simple statistical level, and the deep correlation between data cannot be fully tapped. Third, the existing evaluation method often uniformly processes different regions and different types of degraded grassland, ignoring the regional difference and time sequence characteristics of high-cold grassland restoration. Finally, the formulation of restoration scheme lacks dynamic adjustment mechanism, and cannot timely optimize the restoration strategy according to real-time monitoring data, resulting in unsatisfactory restoration effect. SUMMARY

[0004] The present application provides a high-cold grassland rapid restoration strategy analysis method, system, device and medium, which is used to realize accurate identification of different types and different regions of high-cold grassland degradation, and can dynamically adjust the restoration strategy according to real-time monitoring data, and improve the restoration effect.

[0005] In a first aspect, the application provides a high-cold grassland rapid restoration strategy analysis method, which comprises: performing quality scoring and image restoration processing on collected high-cold grassland multispectral images, and obtaining a standardized image dataset through terrain correction; extracting spectral and texture features according to the standardized image dataset, performing degradation type classification processing on the features, and obtaining a degradation type distribution feature set and a grassland quality scoring table; performing multi-dimensional evaluation based on the degradation type distribution feature set and the grassland quality scoring table, combining remote sensing indexes, terrain indexes, and soil sampling data, generating a degradation degree spatial distribution matrix and a grassland state index set; forming a regional restoration scheme library and a technical parameter list according to the degradation degree spatial distribution matrix and the grassland state index set, through historical restoration case analysis and parameter optimization; performing spatio-temporal response analysis according to the regional restoration scheme library and the technical parameter list, combining meteorological data and terrain data, obtaining a restoration expected index matrix and an effect evaluation table; and obtaining a restoration strategy optimization scheme and a dynamic adjustment parameter set through multi-objective optimization and online monitoring data analysis based on the restoration expected index matrix and the effect evaluation table.

[0006] In a second aspect, the application provides a high-cold grassland rapid restoration strategy analysis system, which comprises:

[0007] A collection module for performing quality scoring and image restoration processing on collected high-cold grassland multispectral images, and obtaining a standardized image dataset through terrain correction;

[0008] A classification module for extracting spectral and texture features according to the standardized image dataset, performing degradation type classification processing on the features, and obtaining a degradation type distribution feature set and a grassland quality scoring table;

[0009] An evaluation module for performing multi-dimensional evaluation based on the degradation type distribution feature set and the grassland quality scoring table, combining remote sensing indexes, terrain indexes, and soil sampling data, generating a degradation degree spatial distribution matrix and a grassland state index set;

[0010] An optimization module for forming a regional restoration scheme library and a technical parameter list according to the degradation degree spatial distribution matrix and the grassland state index set, through historical restoration case analysis and parameter optimization;

[0011] An analysis module for performing spatio-temporal response analysis according to the regional restoration scheme library and the technical parameter list, combining meteorological data and terrain data, obtaining a restoration expected index matrix and an effect evaluation table;

[0012] A monitoring module for obtaining a restoration strategy optimization scheme and a dynamic adjustment parameter set through multi-objective optimization and online monitoring data analysis based on the restoration expected index matrix and the effect evaluation table.

[0013] The third aspect of the present application provides a computer device, wherein the memory stores machine readable instructions executable by the processor, and when the computer device is running, the processor communicates with the memory through a bus, and the machine readable instructions are executed by the processor to perform the steps of the alpine meadow rapid restoration strategy analysis method.

[0014] The fourth aspect of the present application provides a computer readable storage medium, wherein the computer readable storage medium stores instructions, and when the instructions are run on a computer, the computer executes the alpine meadow rapid restoration strategy analysis method.

[0015] In the technical solution provided by the present application, the quality score and image restoration processing are performed on the multispectral image of the alpine meadow, the standardized image dataset is obtained by combining the terrain correction, and the problem of unstable image quality in the traditional method is overcome; the degradation type distribution feature set and the meadow quality score table are obtained by extracting the spectral and texture features for degradation type classification processing, and the accurate identification and scoring of the degradation type are realized; based on the degradation type distribution feature set and the meadow quality score table, the multi-dimensional evaluation is performed in combination with the remote sensing index, the terrain index and the soil sampling data, the degradation degree spatial distribution matrix and the meadow state index set are generated, and the comprehensive degradation condition evaluation result is provided; the regional restoration scheme library and the technical parameter list are formed by the historical restoration case analysis and the parameter optimization, and the targeted restoration strategy is formulated for different regions; the spatiotemporal response analysis is performed in combination with the meteorological data and the terrain data, the restoration expected index matrix and the effect evaluation table are obtained, and the scientific prediction of the restoration effect is realized; finally, the restoration strategy optimization scheme and the dynamic adjustment parameter set are obtained by the multi-objective optimization and the online monitoring data analysis, and the dynamic optimization mechanism of the restoration strategy is established, which not only improves the accuracy of the alpine meadow degradation identification and the scientificity of the restoration scheme formulation, but also ensures the adaptability and effectiveness of the restoration strategy through the dynamic adjustment mechanism. BRIEF DESCRIPTION OF DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor based on these drawings.

[0017] Figure 1 FIG. 1 is a schematic diagram of an embodiment of the alpine meadow rapid restoration strategy analysis method in the embodiment of the present application;

[0018] Figure 2 FIG. 2 is a schematic diagram of an embodiment of the alpine meadow rapid restoration strategy analysis system in the embodiment of the present application;

[0019] Figure 3 FIG. 1 is a structural schematic diagram of a computer device in an embodiment of the present application. DETAILED DESCRIPTION

[0020] The present application provides a high-cold grassland rapid restoration strategy analysis method, system, device and medium. The terms "first", "second", "third", "fourth" and the like (if any) in the specification and claims of the present application and the above-described drawings are used to distinguish similar objects, and do not necessarily have to be used to describe a specific order or sequence. It should be understood that the data thus used can be interchanged under appropriate circumstances, so that the embodiments described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "include" or "have" and any variations thereof are intended to cover non-exclusive inclusion, for example, a process, method, system, product or device including a series of steps or units does not have to be limited to only those steps or units clearly listed, but can include other steps or units not clearly listed or inherent to these processes, methods, products or devices.

[0021] For ease of understanding, the specific process of the embodiments of the present application is described below. Please refer to FIG. 1 Figure 1 One embodiment of the high-cold grassland rapid restoration strategy analysis method in the present application includes:

[0022] Step S101, quality scoring and image restoration processing are performed on the collected high-cold grassland multispectral images, and a standardized image dataset is obtained through terrain correction;

[0023] Step S102, according to the standardized image dataset, spectral and texture features are extracted, and the features are classified and processed according to the degradation type, to obtain a degradation type distribution feature set and a grassland quality scoring table;

[0024] Step S103, based on the degradation type distribution feature set and the grassland quality scoring table, multi-dimensional evaluation is performed in combination with remote sensing indicators, terrain indicators and soil sampling data, to generate a degradation degree spatial distribution matrix and a grassland state index set;

[0025] Step S104, according to the degradation degree spatial distribution matrix and the grassland state index set, historical restoration case analysis and parameter optimization are performed to form a regional restoration scheme library and a technical parameter list;

[0026] Step S105, according to the regional restoration scheme library and the technical parameter list, spatio-temporal response analysis is performed in combination with meteorological data and terrain data to obtain a restoration expected index matrix and an effect evaluation table;

[0027] Step S106, based on the repair expected index matrix and the effect evaluation table, through multi-objective optimization and online monitoring data analysis, the repair strategy optimization scheme and the dynamic adjustment parameter set are obtained.

[0028] It can be understood that the execution subject of the present application can be a high-cold grassland rapid repair strategy analysis system, and can also be a terminal or a server, and the specific place is not limited. The server is taken as an example for description in the embodiments of the present application.

[0029] Specifically, the collected high-cold grassland multispectral image is processed. In the process of processing the high-cold grassland multispectral image, the image data is divided into fixed-size image blocks, and the definition index and the signal-to-noise ratio of each image block are calculated. The definition index is calculated by combining the image gradient and the Laplace operator, and the signal-to-noise ratio is based on the ratio of the pixel value variance to the mean value of the image block. When the definition index is lower than 0.6 or the signal-to-noise ratio is lower than 10 dB, the corresponding image block is marked as a repair area. For the marked repair area, the interference type is further identified. Through the wave band combination analysis method, the reflectivity difference of different wave bands is calculated to identify the cloud layer and the shadow area. Then, the identified interference area is repaired, the spectral information of the adjacent non-interference area is used as a reference to reconstruct the pixel value of the interference area. After obtaining the repaired image, the terrain correction is performed in combination with the digital elevation model data to eliminate the radiation difference caused by the terrain fluctuation, so as to obtain a standardized image data set.

[0030] Based on the obtained standardized image data set, the spectral and texture features are extracted. In the process of spectral feature extraction, the data of each wave band is normalized, and the normalized vegetation index, soil-adjusted vegetation index and other wave band combination indexes are calculated. The texture feature extraction is achieved by calculating the eigenvalues of the gray level co-occurrence matrix to obtain the contrast, entropy, correlation and other statistical quantities. The extracted spectral and texture features are combined to form a feature matrix, which is processed by clustering analysis method to identify the feature patterns of different degradation types and obtain the degradation type distribution feature set. At the same time, a scoring standard is established based on historical sample data to construct a grassland quality scoring table. After obtaining the degradation type distribution feature set and the grassland quality scoring table, multi-dimensional evaluation is performed in combination with the remote sensing index, terrain index and soil sampling data. The evaluation process establishes an evaluation unit grid, and the grassland quality score is mapped into the grid to form a basic score. The vegetation index and coverage data in the remote sensing index are spatially interpolated, combined with the slope and aspect values calculated by the terrain index, and the organic matter content and nutrient index in the soil sampling data, and a degradation degree spatial distribution matrix is generated by weighted superposition method. Based on the matrix, regional statistics and grading are performed to obtain the grassland state index set.

[0031] According to the spatial distribution matrix of degradation degree and the grassland state index set, the repair strategy is formulated by analyzing the historical repair cases. The spatial clustering is performed on the spatial distribution matrix of degradation degree to divide the repair area unit. According to the grassland state index set, the repair priority is determined, and the relevant repair measure parameters and effect records are extracted from the historical repair case data. Through parameter matching and optimization, the specific technical parameters such as fence height, reseeding density and fertilizer amount are determined to form a regional repair scheme library and a technical parameter list. According to the regional repair scheme library and the technical parameter list, the spatio-temporal response analysis is performed combined with the meteorological data and the terrain data. During the analysis process, the spatio-temporal response analysis grid is constructed, and the repair parameters are mapped into the grid. The time series decomposition is performed on the precipitation and temperature data in the meteorological data to obtain the change trend. Combined with the terrain data, the water accumulation index and the radiation amount are calculated, the repair expected index matrix is obtained through spatial superposition analysis, and the effect evaluation table is generated.

[0032] The strategy optimization is performed based on the repair expected index matrix and the effect evaluation table. The target value is extracted from the repair expected index matrix to construct a target parameter space, and the index weight calculation is performed on the effect evaluation table. The time series comparison is performed on the online monitoring vegetation coverage and soil moisture data to obtain the monitoring deviation. According to the deviation, the repair parameters are adjusted in the interval to form the parameter constraint condition. According to the target parameter space and the index weight, the analysis is performed to generate the repair strategy optimization scheme, and the dynamic adjustment parameter set is output.

[0033] For example, in the analysis of a high-cold grassland repair area with an area of 100 square kilometers. In the image processing stage, the multispectral image is divided into 100 image blocks with a size of 1 square kilometer. By calculating the definition index and signal-to-noise ratio of each image block, it is found that 15 image blocks are disturbed by clouds and 8 image blocks are disturbed by shadows. The disturbed areas are repaired by using the spectral information of the adjacent 8 non-disturbed areas as reference values to calculate the average spectral features for reconstruction. In the terrain correction process, combined with the digital elevation model data, the slope and slope direction of each pixel point are calculated, the reflectivity value is adjusted according to the solar incident angle, and the standardized image data set is generated. In the feature extraction stage, the normalized vegetation index and soil-adjusted vegetation index of each pixel point in the standardized image data set are calculated. The texture feature calculation adopts a 7x7 pixel window to calculate the contrast, entropy and correlation of the gray level co-occurrence matrix. The clustering analysis divides the grassland degradation types into quantity degradation type and quality degradation type, and each type is further divided into three levels of mild, moderate and severe. The grassland quality score adopts a 0-100 point system, and the score interval of different degradation types and degrees is determined based on historical sample data.

[0034] In the multi-dimensional assessment stage, the study area is divided into a 10x10 grid of assessment units. Vegetation index is spatially interpolated using inverse distance weighting, and slope and aspect are calculated from digital elevation models. Soil sampling points are located at the center of each assessment unit, and soil organic matter content and nutrient index data are collected. The weights of each index are determined by the analytic hierarchy process, and the weighted superposition method is used to generate the degradation score. According to the score, the study area is divided into 20 restoration region units based on the spatial distribution matrix of degradation degree. By analyzing historical restoration case data, the degradation characteristics, restoration measure parameters, and restoration effects of each case are extracted. For different types and degrees of degradation, the fence height is determined to be between 1.2-1.8 meters, the supplemental seeding density is between 15-30 kg / ha, and the fertilizer amount is between 150-300 kg / ha. In the spatiotemporal response analysis stage, the restoration region unit is used as the basic analysis unit, and monthly precipitation and temperature data for the past five years are collected. The water accumulation index and radiation of each unit are calculated, and the response analysis is conducted in combination with the restoration measure parameters. The expected vegetation coverage improvement value, biomass increase, and species diversity change index of each unit are calculated using spatial statistical methods, and the restoration expected index matrix is generated. In the strategy optimization stage, online monitoring data show that the actual restoration effect deviates from the expected effect, such as the vegetation coverage improvement in some areas not meeting the expected value. Through comparative analysis, the value range of the restoration parameters is adjusted, such as increasing the supplemental seeding density to 20-35 kg / ha. According to the weights and constraints of each index, the optimized restoration strategy is generated, and a dynamic adjustment parameter set is formed.

[0035] In the embodiments of the present application, the quality score and image repair processing are performed on the multispectral image of alpine grassland, the standardized image dataset is obtained by combining terrain correction, and the problem of unstable image quality in the traditional method is overcome; the degradation type distribution feature set and the grassland quality score table are obtained by extracting the spectral and texture features for degradation type classification processing, and the accurate recognition and scoring of the degradation type are realized; based on the degradation type distribution feature set and the grassland quality score table, the multi-dimensional evaluation is performed in combination with the remote sensing index, the terrain index and the soil sampling data, the degradation degree spatial distribution matrix and the grassland state index set are generated, and the comprehensive degradation condition evaluation result is provided; through the historical repair case analysis and parameter optimization, the regional repair scheme library and the technical parameter list are formed, and the targeted repair strategy is formulated for different regions; the spatio-temporal response analysis is performed in combination with the meteorological data and the terrain data, the repair expected index matrix and the effect evaluation table are obtained, and the scientific prediction of the repair effect is realized; finally, the repair strategy optimization scheme and the dynamic adjustment parameter set are obtained through the multi-objective optimization and the online monitoring data analysis, and the dynamic optimization mechanism of the repair strategy is established, which not only improves the accuracy of the alpine grassland degradation recognition and the scientificity of the repair scheme formulation, but also ensures the adaptability and effectiveness of the repair strategy through the dynamic adjustment mechanism.

[0036] In a specific embodiment, the process of performing step S101 can specifically include the following steps:

[0037] (1) The multispectral image of alpine grassland is divided into blocks to obtain an image block sequence;

[0038] (2) The sharpness coefficient and the signal-to-noise ratio parameter of each image block are calculated according to the image block sequence to generate a quality score result;

[0039] (3) The quality score result is compared with a quality threshold value, and the area below the threshold value is marked as a repair area;

[0040] (4) The cloud layer and the shadow are identified in the repair area through band combination analysis to obtain an interference area bitmap;

[0041] (5) The repair image is generated by performing image reconstruction according to the interference area bitmap and the spectral information of the adjacent area;

[0042] (6) The repair image is registered with the terrain elevation data, and the terrain relief coefficient is extracted;

[0043] (7) The repair image is radiometrically corrected based on the terrain relief coefficient and the incident angle parameter to generate a standardized image dataset.

[0044] Specifically, the acquired original image is block cut according to a fixed size. The block cutting adopts a non-overlapping sliding window method, each window size is 1024x1024 pixels, the sliding step is 1024 pixels, starting from the top left corner of the image, scanning row by row to generate an image block sequence. The image block sequence is numbered by a unique identifier, recording the spatial position information of each image block.

[0045] For each image block, the sharpness coefficient and the signal-to-noise ratio parameter are calculated. The calculation of the sharpness coefficient adopts the following formula:

[0046]

[0047] wherein, represents the sharpness coefficient, represents the gray value of the pixel point (x, y), M and N are the row and column numbers of the image block, is the alpine meadow terrain complexity parameter.

[0048] The cloud and shadow recognition process uses the following band combination analysis formula:

[0049]

[0050] wherein, is the cloud recognition index, , , are the reflectivities of the near-infrared, short-wave infrared and visible light bands respectively, , , are the corresponding weight coefficients, is the altitude correction coefficient.

[0051] The extraction of the terrain relief coefficient adopts the following calculation formula:

[0052]

[0053] wherein, is the terrain relief coefficient, is the relative height difference at the grid point (i, j), P and Q are the row and column numbers of the elevation data, is the slope factor, is the vegetation coverage correction coefficient.

[0054] For example, taking an alpine meadow as an example, the original multispectral image has a size of 10240x10240 pixels. After block segmentation, 100 image blocks are obtained, each with a size of 1024x1024 pixels. The sharpness coefficient of each image block is calculated, and the result is between 0 and 1. When the sharpness coefficient is less than 0.6, the image block is marked as a to-be-repaired area. In the band combination analysis, the weight coefficients of the near-infrared, short-wave infrared and visible light bands are 0.4, 0.35 and 0.25 respectively. For the identified cloud layer area, the spectral information of the adjacent 8 cloud-free areas is used for spatial interpolation reconstruction. In the terrain relief coefficient calculation, the elevation data uses a 10-meter resolution digital elevation model, which is registered with the repaired image through bilinear interpolation. Finally, according to the terrain relief coefficient and the solar incident angle, the repaired image is radiometrically corrected to generate a standardized image dataset.

[0055] In a specific embodiment, the process of performing step S102 can specifically include the following steps:

[0056] (1) Separate the standardized image dataset by band, and calculate the reflectance value of each band;

[0057] (2) Normalize the separated band data to generate a band feature matrix;

[0058] (3) Calculate the vegetation index and moisture index based on the band feature matrix to form a spectral feature vector;

[0059] (4) Calculate the gray level co-occurrence matrix parameters and local image gradient values of the standardized image dataset to obtain a texture parameter set;

[0060] (5) Fuse the spectral feature vector and the texture parameter set to construct a degradation feature table;

[0061] (6) Analyze the degradation feature table through classification discrimination to generate a degradation type distribution feature set;

[0062] (7) Score the degradation type distribution feature set in combination with historical sample data to obtain a grassland quality score table.

[0063] Specifically, the standardized image dataset is processed using band separation technology. The band separation technology divides the multispectral image according to the spectral range of different bands, including the blue band (450-520 nm), the green band (520-600 nm), the red band (630-690 nm), and the near-infrared band (760-900 nm). The reflectance value of each band is calculated by radiometric calibration, converting the digital quantitative value into the actual reflectance value. The separated band data is normalized, and the maximum and minimum value normalization method is used to map the reflectance values of each band to the [0, 1] interval. The normalized data forms a band feature matrix, with each row representing a pixel and each column representing the normalized reflectance value of different bands.

[0064] Based on the band feature matrix, the vegetation index and the water index are calculated to form a spectral feature vector. The calculation formula is:

[0065]

[0066] wherein, is the comprehensive vegetation index, is the alpine vegetation type weight coefficient, and are the near-infrared and red band reflectance of the kth vegetation type, is the soil background adjustment factor, is the atmospheric attenuation coefficient, is the altitude.

[0067] Texture analysis is performed on the standardized image dataset to calculate the gray level co-occurrence matrix parameters and local image gradient values. The texture feature extraction formula is:

[0068]

[0069] wherein, is the texture feature value, d is the pixel spacing, θ is the direction angle, and G is the gray level, is the gray level co-occurrence matrix element, is the distance weight, is the vegetation coverage correction factor, is the soil background variance.

[0070] The spectral feature vector and the texture parameter set are fused to construct a degradation feature table using the feature concatenation method. The degradation feature table contains multi-dimensional information such as spectral features, texture features, and terrain features. Through classification discriminant analysis, the degradation feature table is processed, the Mahalanobis distance method is used to calculate the distance between the feature vector and the feature center of each type, the degradation type attribution is determined, and the degradation type distribution feature set is generated.

[0071] In combination with historical sample data, a scoring standard system is established. Feature values of typical sample points in the historical data are extracted as reference standards for each degradation level. Then, for each region in the current degradation type distribution feature set, the similarity of the feature values to the reference standards is calculated, and the scoring value is determined according to the similarity size to form a grassland quality score table.

[0072] For example, a high-cold grassland region is analyzed to obtain multispectral remote sensing images of the region. Four single-band images are obtained through band separation, and each band image has a size of 2048*2048 pixels. The separated band data is normalized to generate a band feature matrix. Based on the matrix, vegetation indices are calculated, and different weight coefficients are set considering the special vegetation types and environmental conditions in the high-cold region. The texture feature calculation uses a 7*7 pixel window to calculate the gray level co-occurrence matrix in four directions (0°, 45°, 90°, 135°). The obtained spectral features and texture features are fused to construct a degradation feature table containing multi-dimensional features. Finally, the degradation degree score is completed by comparing with the historical sample data, and a grassland quality score table is generated.

[0073] In a specific embodiment, the process of performing step S103 can specifically include the following steps:

[0074] (1) Extract spatial position information from the degradation type distribution feature set to establish an evaluation unit grid;

[0075] (2) Map the scoring values in the grassland quality score table to the evaluation unit grid to form a basic scoring matrix;

[0076] (3) Spatially interpolate the vegetation index and coverage data in the remote sensing index to obtain a vegetation state matrix;

[0077] (4) Calculate the slope and slope direction combination value from the terrain index to construct a terrain influence matrix;

[0078] (5) Calculate the organic matter content and nutrient index from the soil sampling data to generate a soil property matrix;

[0079] (6) Weighted superimpose the basic scoring matrix, vegetation state matrix, terrain influence matrix, and soil property matrix to generate a degradation degree spatial distribution matrix;

[0080] (7) Regionally statistically and classify the degradation degree spatial distribution matrix to obtain a grassland state index set.

[0081] Specifically, the degradation assessment module extracts spatial location information from the degradation type distribution feature set, including geographic coordinates, elevation, block boundaries, and other data. Based on the extracted spatial information, an evaluation unit grid is established according to a 1 square kilometer grid size, each grid cell contains a unique identifier and spatial attribute information. When mapping the score values in the grassland quality score table to the evaluation unit grid, a spatial position matching method is used to map each score value to the corresponding grid position to form a basic score matrix. The mapping process considers the consistency of spatial position to ensure that the score values are accurately mapped to the actual geographic location.

[0082] Processing of remote sensing index data mainly includes spatial interpolation of vegetation index and coverage data. Discrete observation point data is interpolated by inverse distance weighting method to generate a continuous vegetation state matrix. The spatial distribution characteristics of the sampling points are considered in the interpolation process to ensure the spatial continuity and reasonableness of the interpolation results. In the processing of terrain indicators, the slope and aspect of each grid cell are calculated using digital elevation model data. The slope calculation is based on the elevation difference of adjacent grid points, and the aspect represents the orientation angle of the slope surface. The slope and aspect data are combined to construct a terrain influence matrix.

[0083] The processing of soil sampling data involves the calculation of organic matter content and nutrient index. The original data of the sampling points are standardized, and a continuous distribution soil property matrix is generated by Kriging interpolation method. The spatial autocorrelation of the sampling points is considered in the interpolation process to improve the accuracy of the interpolation results.

[0084] The generation of the degradation degree spatial distribution matrix uses the following calculation method:

[0085]

[0086] wherein, is the degradation degree spatial distribution matrix, is the basic score matrix element, is the vegetation state matrix element, is the terrain influence matrix element, is the soil property matrix element, , , , are the corresponding weight coefficients, m and n are the row and column numbers of the matrix.

[0087] The degradation degree spatial distribution matrix is subjected to regional statistics and classification processing. Statistical analysis includes calculating the mean, standard deviation, spatial distribution characteristics, and other indicators of each region, and classifying the degradation degree based on the statistical results to generate a grassland state index set. The classification standard is determined based on expert knowledge and historical data to ensure the scientificity and practicality of the classification results.

[0088] For example, a certain alpine meadow research area is 100 square kilometers, and a 10x10 evaluation unit grid is established. Each grid cell is 1 square kilometer in size, and the grassland quality score value is mapped into the grid through spatial matching. The vegetation index data is derived from multispectral remote sensing images, and the coverage data is obtained through field sampling points, using inverse distance weighting method for spatial interpolation. The terrain data uses 10-meter resolution digital elevation model to calculate the slope and slope direction values. Soil sampling is set up at 50 sampling points in the study area, and continuous distribution of soil property data is generated by Kriging interpolation. Through the weighted superposition method, the spatial distribution matrix of degradation degree is generated, and the hierarchical evaluation is completed.

[0089] In a specific embodiment, the process of performing step S104 can specifically include the following steps:

[0090] (1) Spatial clustering of the degradation degree spatial distribution matrix is performed to divide the restoration area unit;

[0091] (2) The restoration area unit is classified according to the grassland state index set to obtain a restoration priority ranking table;

[0092] (3) The restoration measure parameters and effect records are extracted from the historical restoration case data to establish a restoration experience database;

[0093] (4) The parameters in the restoration experience database are matched with the characteristics of the restoration area unit to generate a preliminary restoration measure set;

[0094] (5) The fence height, reseeding density, and fertilizer amount parameters in the preliminary restoration measure set are numerically optimized to form a regional restoration scheme library;

[0095] (6) The restoration measure parameters in the regional restoration scheme library are summarized and arranged to generate a technical parameter list.

[0096] Specifically, the spatial clustering of the degradation degree spatial distribution matrix is performed. The spatial clustering adopts the K-means clustering algorithm, considering the data characteristics of two dimensions of spatial position and degradation degree. In the specific operation, each grid cell in the degradation degree spatial distribution matrix is taken as the clustering object, and the degradation degree value in the matrix and the spatial coordinates of the grid cell are taken as the clustering features. The Euclidean distance between the class center and the sample point is calculated by iteration until the intra-class difference is minimized and the inter-class difference is maximized, and the division of the repair area unit is completed. For the divided repair area unit, combined with the grassland condition index set, a grading process is performed. The grading process is based on the index values in the grassland condition index set, including vegetation coverage, biomass, species diversity, and other indexes. The comprehensive index value of each repair area unit is calculated, and the repair area unit is divided into different priority levels according to the order of the index value. The priority division considers the severity of the degradation degree, the repair difficulty, and the ecological importance, etc., to form a repair priority ranking table.

[0097] The processing process of historical repair case data includes data extraction, cleaning and standardization. The repair measure parameters are extracted from the historical cases, such as the specific parameter values of fence height, reseeding density, fertilizer amount, etc., and the corresponding repair effect records are extracted, including vegetation recovery rate, biomass increase, etc. The extracted data is cleaned to eliminate abnormal values and incomplete records, and standardized to establish a unified format repair experience database. When matching the parameters in the repair experience database with the characteristics of the repair area unit, a similarity calculation method is adopted. The similarity between the characteristics of each repair area unit and the characteristics in the historical cases is calculated, and the repair measure parameters in the case with the highest similarity are selected as the preliminary repair scheme for the region. The similarity calculation considers multiple feature dimensions, including terrain conditions, soil properties, climate factors, etc. The specific parameters in the preliminary repair measure set are optimized. The optimization process considers three key parameters: fence height, reseeding density and fertilizer amount. The value range of these parameters is set, combined with the characteristics of the region and historical experience, and the best parameter combination is determined through numerical optimization method. In the numerical optimization process, multiple objectives such as repair effect, resource input and ecological impact are considered to generate a regional repair scheme library.

[0098] Finally, the repair measure parameters in the regional repair scheme library are summarized and arranged. The summary process includes statistical analysis of the parameter distribution characteristics of different regions, analysis of the relationship between parameters, and arrangement to form a standard format technical parameter list. The technical parameter list contains the specific implementation parameters of each repair region and the corresponding implementation requirements.

[0099] For example, a high-cold grassland restoration analysis is performed. The study area is divided into 20 restoration area units through spatial clustering, with an area of about 5 square kilometers per unit. According to the vegetation coverage index (obtained through remote sensing data), biomass index (obtained through field sampling), and species diversity index (obtained through quadrat investigation) in the grassland state index, a priority ranking is performed. 50 successful cases are selected from the historical restoration case library, and the restoration parameters and effect records are extracted. In the similarity matching process, three characteristic dimensions are mainly considered: altitude (obtained through digital elevation model), annual precipitation (obtained through meteorological station data), and soil type (obtained through soil sampling). Parameter optimization determines the fence height, reseeding density, and fertilizer amount in different areas, and generates a technical list containing specific implementation parameters.

[0100] In a specific embodiment, the process of performing step S105 can specifically include the following steps:

[0101] (1) Extract the spatial distribution information of restoration measures from the sub-regional restoration scheme library, and construct a spatio-temporal response analysis grid;

[0102] (2) Map the parameter values in the technical parameter list to the spatio-temporal response analysis grid to obtain a parameter spatial distribution map;

[0103] (3) Perform time series decomposition on the precipitation and temperature data in the meteorological data to generate a meteorological change trend chart;

[0104] (4) Calculate the water accumulation index and radiation from the terrain data to form an environmental condition matrix;

[0105] (5) Perform spatial overlay analysis on the parameter spatial distribution map and the environmental condition matrix to obtain a restoration expected index matrix;

[0106] (6) Perform spatio-temporal statistics and index calculation on the restoration expected index matrix to generate an effect evaluation table.

[0107] Specifically, the spatial distribution information of the restoration measures in the sub-regional restoration scheme library is extracted. The extraction process includes obtaining the geographic range, restoration measure type, and specific parameters of each restoration area. Based on the extracted spatial information, a spatio-temporal response analysis grid is constructed, with a grid size of 250 meters x 250 meters to ensure that the analysis accuracy meets the restoration effect evaluation requirements. The parameter values in the technical parameter list need to be mapped to the spatio-temporal response analysis grid to establish a spatial correspondence relationship. Each grid cell obtains the corresponding restoration measure parameters, including fence height, reseeding density, and fertilizer amount, according to its geographic location. Through spatial interpolation methods, discrete parameter values are converted into continuous parameter spatial distribution maps.

[0108] The processing of meteorological data includes time series decomposition of precipitation and temperature data. The monthly average temperature and monthly precipitation data are seasonally decomposed with 12 months as the cycle unit to extract long-term trend items, seasonal items and random items. The trend items obtained by decomposition reflect the long-term change characteristics of meteorological elements, and the seasonal items represent the periodic change law, and a meteorological change trend chart is generated.

[0109] The processing of terrain data uses the following calculation formula:

[0110]

[0111] wherein, is an environmental condition matrix, is a slope confluence coefficient, is a runoff parameter, is a slope aspect coefficient, is solar radiation intensity, is an altitude correction factor, g and h are grid row and column numbers, and G and H are matrix dimensions.

[0112] When the parameter spatial distribution map is spatially superimposed with the environmental condition matrix, a weighted superimposition method is used. The restoration expected index value of each grid cell is determined by the restoration measure parameter and the environmental condition, considering the mutual influence relationship between the parameters, and a restoration expected index matrix is generated. Finally, the restoration expected index matrix is subjected to spatio-temporal statistical analysis to calculate the index mean, variance, spatial autocorrelation and other statistical quantities of each region. Based on the statistical analysis results, the restoration expected effect of each region is quantitatively evaluated, and an effect evaluation table containing multiple evaluation indexes is generated.

[0113] For example, a certain alpine meadow research area has an area of 100 square kilometers and is divided into 1600 time-space response analysis grid cells. The fence height parameter in the technical parameter list is spatially interpolated by the inverse distance weighting method to generate a continuously distributed parameter spatial distribution map. Meteorological data is collected from automatic weather stations in the research area, and nearly 5 years of monthly observation data are collected. The terrain data uses a digital elevation model with a resolution of 30 meters, and the water accumulation index and solar radiation of each grid cell are calculated. In the superposition analysis process, the environmental condition weight is 0.4, and the restoration measure parameter weight is 0.6. The spatio-temporal statistical analysis includes calculating the restoration expected index in the short term (within 1 year) and the long term (3 years) to generate time-period and region-specific effect evaluation results.

[0114] In a specific embodiment, the process of performing step S106 can specifically include the following steps:

[0115] (1) Extract the restoration target value from the restoration expected index matrix to construct a target parameter space;

[0116] (2) Calculate the weights of the evaluation indicators in the effect evaluation table to generate an indicator weight vector;

[0117] (3) Time-series comparison of vegetation coverage and soil moisture data in online monitoring data to obtain a monitoring deviation table;

[0118] (4) Interval adjustment of repair parameters based on the monitoring deviation table to form a parameter constraint condition set;

[0119] (5) Repair effect analysis based on the target parameter space and the indicator weight vector to generate a repair strategy optimization scheme;

[0120] (6) Parameter adjustment of the repair strategy combined with the parameter constraint condition set to output a dynamic adjustment parameter set.

[0121] Specifically, the key repair target values are extracted from the repair expected index matrix. The extraction process analyzes the data in each grid cell in the matrix and identifies the expected values of indicators such as vegetation coverage, biomass, and species diversity. For each indicator, a threshold range is set and numerical normalization is performed to convert the values of different dimensions to a unified scale. The extracted target values are reorganized according to the spatial location information to construct the target parameter space. When calculating the weights of the evaluation indicators in the effect evaluation table, the analytic hierarchy process is used to determine the indicator weights. The hierarchical structure of the index system is constructed, including ecological benefit indicators (vegetation coverage, biomass), environmental benefit indicators (soil and water conservation, biodiversity), and economic benefit indicators (input-output ratio). A judgment matrix is established through expert scoring, the characteristic vector is calculated, and consistency check is performed to finally obtain the weight values of each indicator and form the indicator weight vector.

[0122] The processing of online monitoring data involves time-series comparison of vegetation coverage and soil moisture. The monitoring data are arranged in time series, and the data of each monitoring point include monitoring time, spatial location, monitoring value, and other information. The actual monitoring value of each monitoring point is compared with the repair expected value to calculate the deviation value. The deviation calculation considers the influence of time factors and assigns different weights to deviations at different periods to finally generate a monitoring deviation table. The repair parameters are adjusted based on the monitoring deviation table. The adjustment process analyzes the spatial distribution characteristics and time variation trend of the deviation, and identifies areas and periods with large deviations. For different types of deviations, parameter adjustment rules are set, such as increasing the reseeding density for low vegetation coverage and adjusting the irrigation parameters for low soil moisture. The adjustment interval of the parameters is generated through rule operation to form a parameter constraint condition set.

[0123] When the target parameter space and the index weight vector are used for repair effect analysis, a multi-objective optimization method is adopted. The repair target value is taken as the optimization objective, and the index weight is taken as the optimization weight to construct an optimization objective function. In the optimization process, the mutual influence and constraint relationship between each index are considered, and the optimal parameter combination is found through iterative calculation to generate the repair strategy optimization scheme. Finally, the repair strategy is adjusted in combination with the parameter constraint condition set. In the adjustment process, the adjustment range of the constraint condition is strictly followed to ensure that the adjusted parameters meet the actual demand. The parameters of each repair area are adjusted one by one, the correlation between the parameters is considered, and parameter conflicts are avoided. The adjustment result is verified for feasibility, and finally the dynamic adjustment parameter set is output.

[0124] For example, a certain alpine grassland repair project extracts the target values of vegetation coverage, biomass, and species diversity as three key indicators from the repair expected index matrix. The weights of the three indicators are calculated by the analytic hierarchy process to be 0.4, 0.35, and 0.25, respectively. The online monitoring data are derived from 30 automatic monitoring stations arranged in the study area, and each monitoring station records vegetation coverage and soil moisture data every day. Through time series comparison, it is found that the vegetation coverage in some areas is lower than the expected value, and the soil moisture fluctuates greatly. Based on the monitoring deviation, the reseeding density and irrigation frequency are adjusted to form a new parameter constraint range. Finally, through multi-objective optimization calculation, the optimal parameter combination that meets the requirements of each index is obtained, and a dynamic adjustment scheme containing specific implementation parameters is generated.

[0125] The above describes the alpine grassland rapid repair strategy analysis method in the embodiments of the present application, and the alpine grassland rapid repair strategy analysis system in the embodiments of the present application is described below. Please refer to Figure 2 An embodiment of the alpine grassland rapid repair strategy analysis system in the embodiments of the present application includes:

[0126] The acquisition module 201 is configured to perform quality scoring and image repair processing on the collected alpine grassland multispectral image, and obtain a standardized image dataset through terrain correction;

[0127] The classification module 202 is configured to extract spectral and texture features from the standardized image dataset, perform degradation type classification processing on the features, and obtain a degradation type distribution feature set and a grassland quality scoring table;

[0128] The evaluation module 203 is configured to perform multi-dimensional evaluation based on the degradation type distribution feature set and the grassland quality scoring table in combination with remote sensing indexes, terrain indexes, and soil sampling data, and generate a degradation degree spatial distribution matrix and a grassland state index set;

[0129] The optimization module 204 is used for forming a regional repair scheme library and a technical parameter list through historical repair case analysis and parameter optimization according to the degradation degree spatial distribution matrix and the grassland state index set;

[0130] The analysis module 205 is used for obtaining a repair expected index matrix and an effect evaluation table through time-space response analysis combined with meteorological data and terrain data according to the regional repair scheme library and the technical parameter list;

[0131] The monitoring module 206 is used for obtaining a repair strategy optimization scheme and a dynamic adjustment parameter set through multi-objective optimization and online monitoring data analysis based on the repair expected index matrix and the effect evaluation table.

[0132] Through the cooperation of the above components, the quality score and image repair processing are performed on the alpine grassland multispectral image, the standardized image data set is obtained by combining terrain correction, and the problem of unstable image quality in the traditional method is overcome; the degradation type classification processing is performed by extracting spectral and texture features, the degradation type distribution feature set and the grassland quality score table are obtained, and the accurate recognition and scoring of the degradation type are realized; the multi-dimensional evaluation is performed based on the degradation type distribution feature set and the grassland quality score table, combined with remote sensing indexes, terrain indexes and soil sampling data, the degradation degree spatial distribution matrix and the grassland state index set are generated, and comprehensive degradation condition evaluation results are provided; the regional repair scheme library and the technical parameter list are formed through historical repair case analysis and parameter optimization, and the targeted repair strategy is formulated for different regions; the time-space response analysis is performed combined with meteorological data and terrain data, the repair expected index matrix and the effect evaluation table are obtained, and the scientific prediction of the repair effect is realized; finally, the repair strategy optimization scheme and the dynamic adjustment parameter set are obtained through multi-objective optimization and online monitoring data analysis, and the dynamic optimization mechanism of the repair strategy is established, which not only improves the accuracy of the alpine grassland degradation recognition and the scientificity of the repair scheme formulation, but also ensures the adaptability and effectiveness of the repair strategy through the dynamic adjustment mechanism.

[0133] Based on the same technical concept, the embodiment of the present application also provides a computer device. Referring to FIG. 13, Figure 3 The computer device 300 provided by the embodiment of the present application includes a processor 301, a memory 302, and a bus 303. The memory 302 is used for storing execution instructions, including an internal memory 3021 and an external memory 3022; the internal memory 3021 is also called an internal storage, and is used for temporarily storing operation data in the processor 301 and data exchanged with the external memory 3022 such as a hard disk, the processor 301 exchanges data with the external memory 3022 through the internal memory 3021, and the processor 301 and the memory 302 communicate through the bus 303 when the computer device 300 is running.

[0134] The application also provides a computer readable storage medium, which can be a non-volatile computer readable storage medium, or a volatile computer readable storage medium, and the computer readable storage medium stores instructions, and the instructions make a computer execute the steps of the alpine meadow rapid restoration strategy analysis method when the instructions are run on the computer.

[0135] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described system, system and unit can refer to the corresponding processes in the foregoing method embodiments, which will not be described here.

[0136] The integrated unit, if implemented in the form of a software function unit and sold or used as an independent product, can be stored in a computer readable storage medium. Based on such understanding, the technical solutions of the present application, essentially or say the part that makes contributions to the prior art, or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a number of instructions for making a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the methods described in the various embodiments of the present application. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk, and various program code storage media.

[0137] The above-described and above-embodied examples are only used to illustrate the technical solutions of the present application, rather than limit them; although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that: they can still modify the technical solutions recorded in the foregoing examples, or make equivalent replacement for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for analyzing rapid restoration strategies for alpine grasslands, characterized in that, The analytical method for the rapid restoration strategy of alpine grasslands includes: Quality scoring and image restoration processing were performed on the collected multispectral images of alpine grasslands, and a standardized image dataset was obtained through terrain correction. Based on the standardized image dataset, spectral and texture features are extracted, and degradation type classification processing is performed on the features to obtain a degradation type distribution feature set and a grassland quality scoring table; Based on the degradation type distribution feature set and grassland quality scoring table, a multi-dimensional assessment is conducted by combining remote sensing indicators, topographic indicators and soil sampling data to generate a spatial distribution matrix of degradation degree and a grassland state index set. Based on the spatial distribution matrix of degradation degree and the grassland state index set, a regional restoration scheme library and technical parameter list are formed through historical restoration case analysis and parameter optimization. Based on the regional restoration scheme library and technical parameter list, spatiotemporal response analysis is performed using meteorological and topographic data to obtain a restoration expectation index matrix and an effect evaluation table. This includes: extracting spatial distribution information of restoration measures from the regional restoration scheme library to construct a spatiotemporal response analysis grid; mapping parameter values ​​from the technical parameter list to the spatiotemporal response analysis grid to obtain a parameter spatial distribution map; performing time-series decomposition on precipitation and temperature data from the meteorological data to generate a meteorological change trend map; and calculating the moisture accumulation index and radiation from the topographic data to form an environmental condition matrix. The topographic data processing uses the following calculation formula: in, This is the environmental condition matrix. This is the slope runoff coefficient. For runoff parameters, This is the aspect coefficient. Solar radiation intensity, The elevation correction factor is denoted by g and h, which are the grid row and column numbers, and G and H are the matrix dimensions. The spatial distribution map of the parameters is spatially overlaid with the environmental condition matrix to obtain the expected restoration index matrix. Spatiotemporal statistics and index calculations are performed on the expected restoration index matrix to generate an effect evaluation table. Based on the aforementioned repair expectation index matrix and effect evaluation table, through multi-objective optimization and online monitoring data analysis, an optimized repair strategy and a set of dynamically adjusted parameters are obtained.

2. The method for analyzing rapid restoration strategies for alpine grasslands according to claim 1, characterized in that, The process involves quality scoring and image inpainting of the acquired alpine grassland multispectral images, followed by terrain correction to obtain a standardized image dataset, including: Multispectral images of alpine grasslands are segmented into blocks to obtain image block sequences; The sharpness coefficient and signal-to-noise ratio parameter of each image patch are calculated based on the image patch sequence to generate a quality score result; The quality score results are compared with the quality threshold, and areas below the threshold are marked as areas to be repaired. By using band combination analysis, cloud and shadow identification is performed on the area to be repaired to obtain an interference location map; Image reconstruction is performed based on the interference location map and the spectral information of adjacent regions to generate a repaired image; The repaired image is registered with the terrain elevation data, and the terrain undulation coefficient is extracted. Radiometric correction is performed on the restored image based on the terrain undulation coefficient and incident angle parameters to generate a standardized image dataset.

3. The method for analyzing rapid restoration strategies for alpine grasslands according to claim 1, characterized in that, The process involves extracting spectral and texture features from the standardized image dataset, classifying the features according to their degradation types, and obtaining a degradation type distribution feature set and a grassland quality scoring table, including: The standardized image dataset is separated by band, and the reflectance value of each band is calculated. The separated band data are normalized to generate a band feature matrix; Based on the band feature matrix, vegetation index and moisture index are calculated to form spectral feature vector; The gray-level co-occurrence matrix parameters and local image gradient values ​​are calculated on the standardized image dataset to obtain the texture parameter set; The spectral feature vectors are fused with the texture parameter set to construct a degradation feature table; The degradation feature table is analyzed by classification and discrimination to generate a degradation type distribution feature set; By combining historical sample data, a score is calculated on the distribution feature set of the degradation type to obtain a grassland quality score table.

4. The method for analyzing rapid restoration strategies for alpine grasslands according to claim 1, characterized in that, The process involves a multi-dimensional assessment based on the degradation type distribution feature set and grassland quality scoring table, combined with remote sensing indicators, topographic indicators, and soil sampling data, to generate a spatial distribution matrix of degradation degree and a set of grassland state indices, including: Spatial location information is extracted from the degradation type distribution feature set to establish an evaluation unit grid; The score values ​​in the grassland quality scoring table are mapped to the evaluation unit grid to form a basic scoring matrix; Spatial interpolation is performed on the vegetation index and cover data in the remote sensing indicators to obtain the vegetation state matrix; The slope and aspect combination values ​​are calculated from the aforementioned topographic indicators to construct a topographic influence matrix; Based on the soil sampling data, the organic matter content and nutrient index are calculated to generate a soil characteristic matrix; The basic scoring matrix, vegetation state matrix, topographic influence matrix and soil property matrix are weighted and superimposed to generate a spatial distribution matrix of degradation degree. Regional statistics and classification are performed on the spatial distribution matrix of the degradation degree to obtain a set of grassland state indices.

5. The method for analyzing rapid restoration strategies for alpine grasslands according to claim 1, characterized in that, Based on the spatial distribution matrix of degradation degree and the grassland state index set, and through historical restoration case analysis and parameter optimization, a regional restoration scheme library and a list of technical parameters are formed, including: Spatial clustering is performed on the spatial distribution matrix of the degradation degree to divide it into repair region units; The restoration area units are classified according to the grassland state index set to obtain a restoration priority ranking table. Extract restoration measure parameters and effect records from historical restoration case data to establish a restoration experience database; The parameters in the repair experience database are matched with the features of the repair area units to generate a preliminary set of repair measures; Numerical optimization was performed on the parameters of fence height, reseeding density, and fertilizer application rate in the initial remediation measures set to form a regional remediation scheme library; The repair measures parameters in the regional repair scheme library are summarized and organized to generate a list of technical parameters.

6. The method for analyzing rapid restoration strategies for alpine grasslands according to claim 1, characterized in that, Based on the repair expectation index matrix and effect evaluation table, the repair strategy optimization scheme and dynamic adjustment parameter set are obtained through multi-objective optimization and online monitoring data analysis, including: Extract the repair target values ​​from the repair expectation index matrix to construct the target parameter space; The evaluation indicators in the effect evaluation table are weighted to generate an indicator weight vector. By comparing the vegetation cover and soil moisture data in the online monitoring data over time, a monitoring deviation table is obtained. Based on the monitoring deviation table, the repair parameters are adjusted within a range to form a set of parameter constraints. Based on the target parameter space and indicator weight vector, the repair effect is analyzed, and an optimized repair strategy is generated. The repair strategy is adjusted based on the set of parameter constraints, and a dynamically adjusted parameter set is output.

7. A rapid restoration strategy analysis system for alpine grasslands, used to implement the rapid restoration strategy analysis method for alpine grasslands as described in any one of claims 1 to 6, characterized in that, The rapid restoration strategy analysis system for alpine grasslands includes: The acquisition module is used to perform quality scoring and image restoration processing on the acquired alpine grassland multispectral images, and obtain a standardized image dataset through terrain correction. The classification module is used to extract spectral and texture features based on the standardized image dataset, perform degradation type classification processing on the features, and obtain a degradation type distribution feature set and a grassland quality score table. The evaluation module is used to perform multi-dimensional evaluation based on the degradation type distribution feature set and grassland quality scoring table, combined with remote sensing indicators, topographic indicators and soil sampling data, to generate a spatial distribution matrix of degradation degree and a grassland state index set. The optimization module is used to generate a regional restoration scheme library and a list of technical parameters based on the spatial distribution matrix of degradation degree and grassland state index set, through historical restoration case analysis and parameter optimization. The analysis module is used to perform spatiotemporal response analysis based on the regional restoration scheme library and technical parameter list, combined with meteorological and topographic data, to obtain a restoration expectation index matrix and an effect evaluation table. This includes: extracting spatial distribution information of restoration measures from the regional restoration scheme library to construct a spatiotemporal response analysis grid; mapping parameter values ​​from the technical parameter list to the spatiotemporal response analysis grid to obtain a parameter spatial distribution map; performing time-series decomposition on precipitation and temperature data from the meteorological data to generate a meteorological change trend map; and calculating the moisture accumulation index and radiation from the topographic data to form an environmental condition matrix. The topographic data processing uses the following calculation formula: in, This is the environmental condition matrix. This is the slope runoff coefficient. For runoff parameters, This is the aspect coefficient. Solar radiation intensity, The elevation correction factor is denoted by g and h, which are the grid row and column numbers, and G and H are the matrix dimensions. The spatial distribution map of the parameters is spatially overlaid with the environmental condition matrix to obtain the expected restoration index matrix. Spatiotemporal statistics and index calculations are performed on the expected restoration index matrix to generate an effect evaluation table. The monitoring module is used to obtain an optimized repair strategy and a set of dynamically adjusted parameters based on the repair expected indicator matrix and the effect evaluation table through multi-objective optimization and online monitoring data analysis.

8. A computer device, characterized in that, include: The computer device includes a processor, a memory, and a bus. The memory stores machine-readable instructions executable by the processor. When the computer device is running, the processor communicates with the memory via the bus. When the machine-readable instructions are executed by the processor, they perform the steps of the rapid restoration strategy analysis method for alpine grassland as described in any one of claims 1 to 6.

9. A computer-readable storage medium storing instructions thereon, characterized in that, When the instruction is executed by the processor, it implements the rapid restoration strategy analysis method for alpine grasslands as described in any one of claims 1 to 6.

Images