Adjustable grassland remote sensing monitoring system and method

Abstract

The invention relates to the technical field of grassland monitoring, in particular to an adjustable grassland remote sensing monitoring system and method, and the system comprises a data preprocessing module, a variation trend analysis module, a self-adaptive remote sensing parameter configuration module, a grassland monitoring network layout optimization module and a remote sensing observation and intelligent estimation module. The data preprocessing module integrates historical remote sensing and ground actual measurement data and provides a data basis for subsequent analysis; the change trend analysis module calculates the change rate and the standard deviation of the key features and identifies the change trend of the grassland ecological environment; the adaptive remote sensing parameter configuration module adaptively adjusts remote sensing monitoring parameters, and the monitoring efficiency is improved; the grassland monitoring network layout optimization module optimizes the layout of the remote sensing detection unit and improves the efficiency of the monitoring network; and remote sensing observation and intelligent estimation use a spatial interpolation method to estimate the value of a future observation point, so that the workload of field measurement is reduced. The method is used for solving the technical problem that the current grassland monitoring management effect is poor.

Description

Technical Field

[0001] The present invention relates to the technical field of grassland monitoring, and in particular to an adjustable grassland remote sensing monitoring system and method. Background Art

[0002] The existing grassland monitoring technology has the following problems: the grassland monitoring and management effect is not good. The current technology may lead to poor grassland monitoring and management effect due to incomplete data processing, fixed monitoring parameters or unreasonable monitoring network layout; lack of ability to adjust remote sensing monitoring parameters. The existing technology may not be able to adjust remote sensing monitoring parameters according to the changing trend of grassland ecological environment, such as spectral range, resolution and acquisition time; unreasonable grassland monitoring network layout. The current grassland monitoring network layout may not be able to fully cover the monitoring area, or the unreasonable layout may lead to uneven data quality; the workload of field measurement is large. Traditional grassland monitoring methods may require a lot of field measurement work, which is time-consuming, labor-intensive and costly. To this end, the present invention proposes an adjustable grassland remote sensing monitoring system and method. Summary of the invention

[0003] The purpose of the present invention is to solve the problems in the background technology and to propose an adjustable grassland remote sensing monitoring system and method.

[0004] In order to achieve the above object, the present invention adopts the following technical solutions: An adjustable grassland remote sensing monitoring system includes: a data preprocessing module, a change trend analysis module, an adaptive remote sensing parameter configuration module, a grassland monitoring network layout optimization module, and a remote sensing observation and intelligent estimation module; Data preprocessing module: obtain historical remote sensing data and ground measured data, preprocess the acquired historical remote sensing data and ground measured data, and obtain data preprocessing results; Change trend analysis module: Combine the data preprocessing results to extract key feature vectors, use the formula to calculate the average change rate and standard deviation for each key feature, and analyze the change trend of each key feature; Adaptive remote sensing parameter configuration module: According to the trend analysis results of key features, the remote sensing monitoring parameters are adjusted by establishing the objective function; Grassland monitoring network layout optimization module: Based on the output of the adaptive remote sensing parameter configuration module, the grassland monitoring area is divided into grid units and equipped with remote sensing detection units, and the layout of the remote sensing monitoring network is adjusted by calculating and analyzing the ecological environment data quality indicators; Remote sensing observation and intelligent estimation module: The semivariogram function is used to analyze grassland observation points, and the spatial interpolation method is used to estimate the values ​​of future observation points.

[0005] Furthermore, the data preprocessing module obtains historical remote sensing data and ground measured data, and preprocesses the obtained historical remote sensing data and ground measured data. The process of obtaining the data preprocessing result includes: Get a collection of historical remote sensing data and mark it as ;in, Indicates Historical remote sensing data samples, represents the number of historical remote sensing data samples, Represents the number of features of each historical remote sensing data sample (for example, reflectance in different bands, different vegetation indices, etc. can be used as a feature); historical remote sensing data can be obtained from images of the earth's surface taken at past time points by satellites or other remote sensing platforms; Get the ground truth data set ;in, Indicates Ground truth data samples, is the number of ground truth data samples, The number of features for each ground-measured data sample (e.g., the value of soil moisture at different depths, the actual height of ground vegetation, etc. can be used as features); ground-measured data is data obtained through measurement activities conducted in the field, such as using a drone equipped with a remote sensing detection unit to perform low-altitude remote sensing photography or laser measurement to obtain detailed terrain data; The original data matrix is ​​composed of historical remote sensing data and ground-based measured data ,Right now ;in, Represents the total number of data samples, which is the sum of the number of historical remote sensing data samples and the number of ground measured data samples; It represents the total number of features of the data sample, which is the sum of the number of features of the historical remote sensing data and the number of features of the ground measured data; represents the real number threshold; By calculating the covariance matrix Measuring the correlation between different features of the data: , where Represents the transpose of a matrix; Calculate the covariance matrix The characteristic equation of , and get the eigenvalue and the corresponding eigenvector ,in, It represents the eigenvalue, which is used to reflect the variance of the principal component represented by the corresponding eigenvector. The larger the variance, the more information the principal component contains. Represents the identity matrix, which is used to maintain the regularity of matrix operations when calculating the characteristic equation; Calculate the cumulative contribution rate using the formula When the cumulative contribution rate reaches the preset threshold (the preset threshold is 85%), the previous The eigenvectors form the projection matrix ;in, represents the feature vector index, It indicates the number of eigenvectors determined according to the cumulative contribution rate, which is used to indicate how many principal components are selected to represent the main information of the original data; Project the original matrix into a lower dimensional space , and obtain a set of features in the low-dimensional space; among them, Represents the data matrix after dimensionality reduction. The data matrix after dimensionality reduction is the projection result of the historical remote sensing data and the ground measured data in the selected feature vector direction. These projection results retain the information of the original high-dimensional data (i.e., the historical remote sensing data and the ground measured data) in the main feature direction, thereby realizing data dimensionality reduction and feature extraction; Key features are selected from this set of features and used as the result of data preprocessing for subsequent change trend analysis.

[0006] Furthermore, the change trend analysis module extracts key feature vectors based on the data preprocessing results, and uses the formula to calculate the average change rate and standard deviation for each key feature. The process of analyzing the change trend of each key feature includes: Extract key feature vectors and mark them as ; For each key feature , whose time series is ;in, represents the index of the key feature, Indicates time, and , Indicates the total number of time points; calculate key features using the formula In different time windows The average rate of change within: , where Respectively represent the start and end time points of the time window; Calculate key characteristics using formula The standard deviation of , where Representing key features In the time window The average value within , standard deviation To reflect key features In the time window The degree of fluctuation within According to the average rate of change and standard deviation , identify key features In the time window The changing trends within.

[0007] Furthermore, the adaptive remote sensing parameter configuration module adjusts the remote sensing monitoring parameters by establishing an objective function according to the change trend analysis results of the key features, including: Define the remote sensing monitoring parameter vector as ;in, represent the spectral range, resolution, and acquisition time, respectively; Establish the objective function: , where Indicates The weight of the key feature is used to reflect the The importance of the key features in the objective function; Indicates The expected rate of change of each key characteristic; The weight representing the degree of fluctuation is used to measure the importance of the degree of fluctuation of key features in the objective function; Indicates The expected standard deviation of the key features; According to the objective function, the remote sensing monitoring parameters are adjusted, including the spectral range, resolution and acquisition time; among them, for the adjustment of the spectral range, , where is the initial spectral range, respectively representing the lower and upper limits of the initial spectral range, Represents the learning rate of the spectral range, which determines the step size each time the spectral range is adjusted. is the adjusted spectral range; for resolution adjustment, , where Indicates the adjusted resolution. Indicates the initial resolution, that is, the resolution set at the beginning of remote sensing monitoring. Represents the learning rate of resolution, which is used to control the step size of resolution adjustment; for acquisition time adjustment, , where represents the adjusted acquisition time, represents the initial acquisition time, is the learning rate of the acquisition time, which is used to control the step size of the acquisition time adjustment; Output adjusted remote sensing monitoring parameters .

[0008] Furthermore, the grassland monitoring network layout optimization module divides the grassland monitoring area into grid units and configures remote sensing detection units based on the output of the adaptive remote sensing parameter configuration module, and adjusts the layout of the remote sensing monitoring network by calculating and analyzing the ecological environment data quality indicators. The process includes: The grassland monitoring area is defined as , and discretized into grid cells, and the coordinates of each grid cell are ;in, Respectively represent the grid cell indexes of the grassland monitoring area in the north-south and east-west directions, and , They represent the number of grid cells divided into the grassland monitoring area in the north-south and east-west directions respectively; For each grid unit, each grid unit is equipped with a remote sensing detection unit for obtaining ecological environment parameters; the remote sensing detection unit includes a sensor component, a platform, a processor and an output device. The sensor component is the core component of the remote sensing detection unit, and includes different types of sensors for simultaneously capturing the electromagnetic wave energy emitted or reflected by multiple objects such as vegetation and ground, and converting it into a signal that can be recorded or transmitted; the platform is the carrier of the remote sensing detection unit, such as a satellite, an aircraft or a drone, etc., responsible for selecting the best observation position to ensure high-quality observation data; the processor performs necessary processing on the signal output by the sensor component, including signal amplification, filtering, transformation, etc., to extract useful information and convert it into a data format that can be used for further analysis or display; the output device is responsible for converting the signal processed by the processor into an observable or storable form, namely, ecological environment parameters (vegetation coverage, soil moisture and surface temperature); Get the remote sensing detection unit location collection , where Represents the number of remote sensing detection units; The position of a remote sensing detection unit is expressed as , the location of each remote sensing detection unit is located within the constructed grid unit, that is, , Represents the index of the remote sensing detection unit; Simultaneous remote sensing unit ,satisfy The constraint condition is, Indicates the actual size of the monitoring area of ​​the remote sensing detection unit. It represents the sensing radius of the remote sensing detection unit, that is, the farthest distance at which the remote sensing detection unit can obtain data; According to the ecological environment parameters obtained by the remote sensing detection unit, calculate the ecological environment data quality index of the remote sensing detection unit : , where are preset different proportional coefficients, which respectively represent the relative importance of vegetation coverage, soil moisture and surface temperature in calculating the ecological environment data quality index; among them, the ecological environment data quality index The vegetation coverage in the ecological environment parameters is used , soil moisture and surface temperature As a measure; Collect the ecological environment data quality indicators within the coverage area of ​​all remote sensing detection units and compare them with the preset standard values ​​in turn; if the ecological environment data quality indicators are lower than the preset standard values, the remote sensing detection unit will be marked for layout optimization; otherwise, it will not be marked; When optimizing the layout of the grassland monitoring network, by adjusting the coverage of the marked remote sensing detection units, including the location, number or perception radius of the remote sensing detection units, the aim is to optimize the ecological environmental data quality indicators of the entire grassland monitoring area, thereby outputting the optimized grassland remote sensing monitoring network layout; it is understandable that the adjustment of the number of remote sensing detection units can increase the number of remote sensing detection units in the grassland monitoring area, which can improve the coverage and accuracy of the data; for the adjustment of the position of the remote sensing detection units, according to the monitoring objectives and geographical features, the position of the remote sensing detection units can be reasonably adjusted to ensure that the entire grassland monitoring area can be effectively covered; for the adjustment of the perception radius of the remote sensing detection unit, its monitoring range and the precision of data collection can be further controlled, and increasing the perception radius can expand the monitoring area of ​​a single remote sensing detection unit, thereby improving the data coverage when the number of remote sensing detection units is limited.

[0009] Furthermore, the remote sensing observation and intelligent estimation module uses the semivariogram to analyze the grassland observation points and uses the spatial interpolation method to estimate the values ​​of future observation points. The process includes: Combined with the optimized grassland remote sensing monitoring network layout information, determine Grassland observation point ; Among them, the observation points represent the observation data at different locations within the grassland monitoring area; Building a semivariogram , where Represents the grassland observation point index, which is used to traverse all grassland observation point pairs involved in the calculation, and ; represents the number of grassland observation points, that is, the number of grassland observation point pairs involved in calculating the semivariogram; Indicates The location of the grassland observation points; Indicates at location The observation value at a certain location, where the observation value can be any variable related to grassland monitoring, such as vegetation coverage, soil moisture or surface temperature data measured at a certain geographical coordinate; is the spatial lag distance, which represents the spatial interval between two observation locations. For example, if the grassland observation point The coordinates of , grassland observation point The coordinates of ,but , reflects the distance relationship between two observation points in space; By fitting the model to the semivariogram, the parameters of the semivariogram are obtained: , In the formula, Indicates the value of the membership fee. represents the partial sill value, represents the range; the purpose of fitting the model is to find the parameters of the semivariogram The value of makes the model curve close to the semivariogram value; it is understandable that the semivariogram is an important tool in geostatistics, which describes the statistical law of the difference between the values ​​of a variable (such as vegetation coverage, soil moisture, etc.) between two points in space. By solving the parameters of the semivariogram (such as the base value, nugget value, range, etc.), we can deeply understand the characteristics of spatial variability in the grassland monitoring area, including the intensity and range of spatial correlation; After obtaining the parameters of the semivariogram model, the spatial interpolation method is used to estimate the values ​​of future observation points: the coordinates of the future observation points to be estimated are marked as , get known grassland observation points ; According to the semivariogram model, the interpolation equations are: , where represents the weight coefficient, represents the Lagrange multiplier; solve the interpolation equations to obtain the weight coefficients , where the weight coefficients and Lagrange multipliers can be used as unknowns to construct a linear equation system, and the weight coefficients can be extracted from the solved linear equation system; using the formula Estimating future observation points The value of; It is understandable that the obtained semivariogram function parameters can be used for spatial interpolation. Spatial interpolation is a method of inferring unknown point data based on known point data. It plays an important role in spatial analysis. According to the characteristics of the semivariogram function, the data of future observation points can be inferred. These interpolation results provide more comprehensive data distribution information in the grassland monitoring area. The estimated values ​​of future observation points are stored in the background data management center as the basis for adjusting the frequency of grassland remote sensing monitoring. Specifically, the frequency of grassland remote sensing monitoring is adjusted as follows: a monitoring frequency adjustment logic based on remote sensing data is formulated. For example, when remote sensing monitoring shows that the ecological status of the grassland may change significantly (such as a sharp drop in vegetation coverage, a sharp change in soil moisture, and other obvious changes in remote sensing image features), the frequency of remote sensing monitoring should be increased to obtain more detailed and accurate grassland ecological data. At the same time, the adjustment logic should also comprehensively consider other factors, such as differences in vegetation growth cycles caused by seasonal changes, the impact of weather conditions on the quality of remote sensing data, and the potential impact of human activities (such as grazing, farming, etc.) on grassland ecosystems, so as to ensure the rationality and pertinence of the remote sensing monitoring frequency and provide a scientific basis for the management and protection of grassland resources.

[0010] An adjustable grassland remote sensing monitoring method, comprising: Step 1: Acquire historical remote sensing data and ground measured data, preprocess the acquired historical remote sensing data and ground measured data to obtain data preprocessing results; Step 2: Combine the data preprocessing results to extract key feature vectors, use the formula to calculate the average change rate and standard deviation for each key feature, and analyze the change trend of each key feature; Step 3: According to the trend analysis results of key features, adjust the remote sensing monitoring parameters by establishing the objective function; Step 4: Based on the adjusted remote sensing monitoring parameters, the grassland monitoring area is divided into grid units and equipped with remote sensing detection units, and the layout of the remote sensing monitoring network is adjusted by calculating and analyzing the ecological environment data quality indicators; Step 5: Use the semivariogram function to analyze the grassland observation points and use spatial interpolation to estimate the values ​​of future observation points.

[0011] Compared with the existing technology, the advantages of the adjustable grassland remote sensing monitoring system and method provided by the present invention are: 1. The present invention can retain the main information in the original data and reduce the complexity of data processing by acquiring historical remote sensing data and ground measured data and performing preprocessing; by extracting key feature vectors and analyzing their average change rate and standard deviation, the changing trend of the grassland ecological environment is revealed, which helps to understand the dynamic characteristics of the grassland ecosystem; 2. The present invention can improve the acquisition efficiency and accuracy of remote sensing data by adaptively adjusting remote sensing monitoring parameters such as spectral range, resolution and acquisition time according to the change trend analysis results of key features, thereby better meeting the needs of grassland monitoring; 3. The present invention ensures comprehensive coverage of the monitoring network by dividing the grassland monitoring area into grid units and equipping remote sensing detection units. By calculating and analyzing ecological environmental data quality indicators, it can identify monitoring areas with low data quality and optimize the layout, so that the optimized monitoring network layout can improve data quality and coverage. By analyzing grassland observation points using semivariogram functions and using spatial interpolation methods to estimate the values ​​of future observation points, it helps to reduce the workload of field measurements and improve monitoring efficiency.

[0012] In summary, the present invention can, according to actual conditions, constitute a comprehensive and reliable grassland remote sensing monitoring system through the mutual collaboration of five links: data preprocessing, change trend analysis, remote sensing parameter configuration, monitoring network layout optimization, and remote sensing observation and intelligent estimation, thereby ensuring the normal operation of a subsequent adjustable grassland remote sensing monitoring system and method. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 This is a module diagram of an adjustable grassland remote sensing monitoring system proposed by the present invention.

[0014] Figure 2 The present invention is a flowchart of an adjustable grassland remote sensing monitoring method. DETAILED DESCRIPTION

[0015] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the implementation regulations described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0016] Reference Figure 1 , an adjustable grassland remote sensing monitoring system, the system includes a data preprocessing module, a change trend analysis module, an adaptive remote sensing parameter configuration module, a grassland monitoring network layout optimization module and a remote sensing observation and intelligent estimation module; Data preprocessing module: obtain historical remote sensing data and ground measured data, preprocess the acquired historical remote sensing data and ground measured data, and obtain data preprocessing results; Change trend analysis module: Combine the data preprocessing results to extract key feature vectors, use the formula to calculate the average change rate and standard deviation for each key feature, and analyze the change trend of each key feature; Adaptive remote sensing parameter configuration module: According to the trend analysis results of key features, the remote sensing monitoring parameters are adjusted by establishing the objective function; Grassland monitoring network layout optimization module: Based on the output of the adaptive remote sensing parameter configuration module, the grassland monitoring area is divided into grid units and equipped with remote sensing detection units, and the layout of the remote sensing monitoring network is adjusted by calculating and analyzing the ecological environment data quality indicators; Remote sensing observation and intelligent estimation module: The semivariogram function is used to analyze grassland observation points, and the spatial interpolation method is used to estimate the values ​​of future observation points.

[0017] The data preprocessing module obtains historical remote sensing data and ground measured data, and preprocesses the obtained historical remote sensing data and ground measured data. The steps of obtaining the data preprocessing results include: Step 101: Obtain a historical remote sensing data set and mark it as ;in, Indicates Historical remote sensing data samples, represents the number of historical remote sensing data samples, Indicates the number of features of each historical remote sensing data sample (for example, reflectance of different bands, different vegetation indices, etc. can be used as a feature); In step 101, historical remote sensing data may be obtained from images of the earth's surface taken at past time points by satellites or other remote sensing platforms; Step 102: Obtaining a set of ground-based measured data ;in, Indicates Ground truth data samples, is the number of ground truth data samples, The number of features for each ground-truth data sample (e.g., the values ​​of soil moisture at different depths, the actual height of ground vegetation, etc. can be used as features); In step 102, the ground measured data is data obtained through measurement activities conducted in the field, such as using a drone equipped with a remote sensing detection unit to perform low-altitude remote sensing photography or laser measurement, which can obtain fine terrain data; Step 103: Combining historical remote sensing data and ground-measured data into an original data matrix ,Right now ;in, Represents the total number of data samples, which is the sum of the number of historical remote sensing data samples and the number of ground measured data samples; It represents the total number of features of the data sample, which is the sum of the number of features of the historical remote sensing data and the number of features of the ground measured data; represents the real number threshold; Step 104: Calculate the covariance matrix Measuring the correlation between different features of the data: , where Represents the transpose of a matrix; Step 105: Calculate the covariance matrix The characteristic equation of , and get the eigenvalue and the corresponding eigenvector ,in, It represents the eigenvalue, which is used to reflect the variance of the principal component represented by the corresponding eigenvector. The larger the variance, the more information the principal component contains. Represents the identity matrix, which is used to maintain the regularity of matrix operations when calculating the characteristic equation; Step 106: Calculate the cumulative contribution rate using the formula When the cumulative contribution rate reaches the preset threshold (the preset threshold is 85%), the previous The eigenvectors form the projection matrix ;in, represents the feature vector index, It indicates the number of eigenvectors determined according to the cumulative contribution rate, which is used to indicate how many principal components are selected to represent the main information of the original data; Step 107: Project the original matrix into a low-dimensional space , and obtain a set of features in the low-dimensional space; among them, Represents the data matrix after dimensionality reduction. The data matrix after dimensionality reduction is the projection result of the historical remote sensing data and the ground measured data in the selected feature vector direction. These projection results retain the information of the original high-dimensional data (i.e., the historical remote sensing data and the ground measured data) in the main feature direction, thereby realizing data dimensionality reduction and feature extraction; Step 108: Filter out key features from the set of features and use them as the result of data preprocessing for subsequent change trend analysis.

[0018] The change trend analysis module extracts key feature vectors based on the data preprocessing results, and uses the formula to calculate the average change rate and standard deviation for each key feature. The steps for analyzing the change trend of each key feature include: Step 201: Extract key feature vectors and mark them as ; Step 202: For each key feature , whose time series is ;in, represents the index of the key feature, Indicates time, and , Indicates the total number of time points; calculate key features using the formula In different time windows The average rate of change within: , where Respectively represent the start and end time points of the time window; Step 203: Calculate key features using formula The standard deviation of , where Representing key features In the time window The average value within , standard deviation To reflect key features In the time window The degree of fluctuation within Step 204: According to the average rate of change and standard deviation , identify key features In the time window The changing trends within.

[0019] The adaptive remote sensing parameter configuration module adjusts the remote sensing monitoring parameters by establishing an objective function based on the trend analysis results of key features. The steps include: Step 301: define the remote sensing monitoring parameter vector as ;in, represent the spectral range, resolution, and acquisition time, respectively; Step 302: Establish the objective function: , where Indicates The weight of the key feature is used to reflect the The importance of the key features in the objective function; Indicates The expected rate of change of each key characteristic; The weight representing the degree of fluctuation is used to measure the importance of the degree of fluctuation of key features in the objective function; Indicates The expected standard deviation of the key features; Step 303: adjust the remote sensing monitoring parameters according to the objective function, including the spectral range, resolution and acquisition time; wherein, for the adjustment of the spectral range, , where is the initial spectral range, respectively representing the lower and upper limits of the initial spectral range, Represents the learning rate of the spectral range, which determines the step size each time the spectral range is adjusted. is the adjusted spectral range; for resolution adjustment, , where Indicates the adjusted resolution. Indicates the initial resolution, that is, the resolution set at the beginning of remote sensing monitoring. Represents the learning rate of resolution, which is used to control the step size of resolution adjustment; for acquisition time adjustment, , where represents the adjusted acquisition time, represents the initial acquisition time, is the learning rate of the acquisition time, which is used to control the step size of the acquisition time adjustment; Step 304: Outputting the adjusted remote sensing monitoring parameters .

[0020] The grassland monitoring network layout optimization module divides the grassland monitoring area into grid units and configures remote sensing detection units based on the output of the adaptive remote sensing parameter configuration module, and adjusts the layout of the remote sensing monitoring network by calculating and analyzing the ecological environment data quality indicators. The steps include: Step 401: Define the grassland monitoring area as , and discretized into grid cells, and the coordinates of each grid cell are ;in, Respectively represent the grid cell indexes of the grassland monitoring area in the north-south and east-west directions, and , They represent the number of grid cells divided into the grassland monitoring area in the north-south and east-west directions respectively; Step 402: For each grid unit, each grid unit is equipped with a remote sensing detection unit for obtaining ecological environment parameters; In step 402, the remote sensing detection unit includes a sensor component, a platform, a processor and an output device. The sensor component is the core component of the remote sensing detection unit, and includes different types of sensors for simultaneously capturing electromagnetic wave energy emitted or reflected by multiple objects such as vegetation and ground, and converting it into a signal that can be recorded or transmitted; the platform is the carrier of the remote sensing detection unit, such as a satellite, an airplane or a drone, etc., and is responsible for selecting the best observation position to ensure high-quality observation data; the processor performs necessary processing on the signal output by the sensor component, including signal amplification, filtering, transformation, etc., to extract useful information and convert it into a data format that can be used for further analysis or display; the output device is responsible for converting the signal processed by the processor into an observable or storable form, that is, ecological environment parameters (vegetation coverage, soil moisture and surface temperature); Step 403: Obtaining a remote sensing detection unit location set , where Represents the number of remote sensing detection units; The position of a remote sensing detection unit is expressed as , the location of each remote sensing detection unit is located within the constructed grid unit, that is, , Represents the index of the remote sensing detection unit; Step 404: Simultaneous remote sensing detection unit ,satisfy The constraint condition is, Indicates the actual size of the monitoring area of ​​the remote sensing detection unit. It represents the sensing radius of the remote sensing detection unit, that is, the farthest distance at which the remote sensing detection unit can obtain data; Step 405: Calculate the ecological environment data quality index of the remote sensing detection unit according to the ecological environment parameters obtained by the remote sensing detection unit. : , where are preset different proportional coefficients, which respectively represent the relative importance of vegetation coverage, soil moisture and surface temperature in calculating the ecological environment data quality index; among them, the ecological environment data quality index The vegetation coverage in the ecological environment parameters is used , soil moisture and surface temperature As a measure; Step 406: collect the ecological environment data quality indicators within the coverage area of ​​all remote sensing detection units, and compare them with the preset standard values ​​in sequence; if the ecological environment data quality indicator is lower than the preset standard value, mark the remote sensing detection unit for layout optimization; otherwise, do not mark it; Step 407: When optimizing the grassland monitoring network layout, by adjusting the coverage of the marked remote sensing detection units, including the location, quantity or sensing radius of the remote sensing detection units, the ecological environment data quality index of the entire grassland monitoring area is optimized, thereby outputting the optimized grassland remote sensing monitoring network layout; In step 407, the number of remote sensing detection units can be adjusted to increase the number of remote sensing detection units in the grassland monitoring area, thereby improving the data coverage and accuracy; the position of the remote sensing detection units can be adjusted according to the monitoring targets and geographical features to ensure that the entire grassland monitoring area can be effectively covered; the perception radius of the remote sensing detection units can be adjusted to further control the monitoring range and the precision of data collection, and increasing the perception radius can expand the monitoring area of ​​a single remote sensing detection unit, thereby improving the data coverage when the number of remote sensing detection units is limited.

[0021] The remote sensing observation and intelligent estimation module uses the semivariogram function to analyze grassland observation points and uses spatial interpolation to estimate the values ​​of future observation points. The steps include: Step 501: Determine the optimized grassland remote sensing monitoring network layout information. Grassland observation point ; Among them, the observation points represent the observation data at different locations within the grassland monitoring area; Step 502: Establishing a semivariogram , where Represents the grassland observation point index, which is used to traverse all grassland observation point pairs involved in the calculation, and ; represents the number of grassland observation points, that is, the number of grassland observation point pairs involved in calculating the semivariogram; Indicates The location of the grassland observation points; Indicates at location The observation value at a certain location, where the observation value can be any variable related to grassland monitoring, such as vegetation coverage, soil moisture or surface temperature data measured at a certain geographical coordinate; is the spatial lag distance, which represents the spatial interval between two observation locations. For example, if the grassland observation point The coordinates of , grassland observation point The coordinates of ,but , reflects the distance relationship between two observation points in space; Step 503: Get the parameters of the semivariogram by fitting the model to the semivariogram: , In the formula, Indicates the value of the membership fee. represents the partial sill value, represents the range; the purpose of fitting the model is to find the parameters of the semivariogram The value of makes the model curve close to the semivariogram value; In step 503, the semivariogram is an important tool in geostatistics, which describes the statistical law of the difference degree of a variable value (such as vegetation coverage, soil moisture, etc.) between two points in space. By solving the parameters of the semivariogram (such as the base value, nugget value, range, etc.), the characteristics of spatial variability in the grassland monitoring area can be deeply understood, including the intensity and range of spatial correlation; Step 504: After obtaining the parameters of the semivariogram model, use the spatial interpolation method to estimate the values ​​of future observation points: mark the coordinates of the future observation points to be estimated as , get known grassland observation points ; According to the semivariogram model, the interpolation equations are: , where represents the weight coefficient, represents the Lagrange multiplier; solve the interpolation equations to obtain the weight coefficients , where the weight coefficients and Lagrange multipliers can be used as unknowns to construct a linear equation system, and the weight coefficients can be extracted from the solved linear equation system; using the formula Estimating future observation points The value of In step 504, spatial interpolation can be performed using the solved semivariogram parameters. Spatial interpolation is a method of inferring unknown point data based on known point data, and it plays an important role in spatial analysis. Based on the characteristics of the semivariogram, the data of future observation points can be inferred. These interpolation results provide more comprehensive data distribution information in the grassland monitoring area. Step 505: storing the estimated values ​​of future observation points in the background data management center as a basis for adjusting the grassland remote sensing monitoring frequency; In step 505, the adjustment of the grassland remote sensing monitoring frequency is specifically as follows: formulating a monitoring frequency adjustment logic based on remote sensing data. For example, when remote sensing monitoring finds that the grassland ecological status may change significantly (such as a significant decrease in vegetation coverage, a sharp change in soil moisture, and other obvious changes in remote sensing image features), the remote sensing monitoring frequency should be increased to obtain more detailed and accurate grassland ecological data; at the same time, the adjustment logic should also comprehensively consider other factors, such as differences in vegetation growth cycles caused by seasonal changes, the impact of weather conditions on the quality of remote sensing data, and the potential impact of human activities (such as grazing, farming, etc.) on the grassland ecosystem, so as to ensure the rationality and pertinence of the remote sensing monitoring frequency and provide a scientific basis for the management and protection of grassland resources.

[0022] In the embodiment of the present invention, data dimension reduction is achieved through covariance matrix analysis and eigenvector projection, key information is retained, and data processing complexity is reduced. By calculating the average change rate and standard deviation of key features, real-time monitoring of grassland ecological environment changes is achieved. By adjusting the remote sensing parameter configuration, it is ensured that the monitoring parameters match the actual environment and the accuracy of the monitoring results is improved. By planning the layout of the grassland monitoring network, the coverage and accuracy of the data can be improved. By using the semivariogram function and spatial interpolation method, spatial analysis of the grassland ecological environment is achieved, so as to make more accurate grassland management decisions. In summary, the example of the present invention solves the problem of poor grassland monitoring and management results. In actual situations, more data and context information may be needed to make specific decisions and optimization plans.

[0023] Reference Figure 2 , an adjustable grassland remote sensing monitoring method, comprising: Step 1: Acquire historical remote sensing data and ground measured data, preprocess the acquired historical remote sensing data and ground measured data to obtain data preprocessing results; Step 2: Combine the data preprocessing results to extract key feature vectors, use the formula to calculate the average change rate and standard deviation for each key feature, and analyze the change trend of each key feature; Step 3: According to the trend analysis results of key features, adjust the remote sensing monitoring parameters by establishing the objective function; Step 4: Based on the adjusted remote sensing monitoring parameters, the grassland monitoring area is divided into grid units and equipped with remote sensing detection units, and the layout of the remote sensing monitoring network is adjusted by calculating and analyzing the ecological environment data quality indicators; Step 5: Use the semivariogram function to analyze the grassland observation points and use spatial interpolation to estimate the values ​​of future observation points.

[0024] In addition, the formulas involved in the above are all calculated by removing dimensions and taking their numerical values. They are a formula that is closest to the actual situation obtained by collecting a large amount of data and performing software simulation. The proportional coefficient in the formula and the various preset thresholds in the analysis process are set by technical personnel in this field according to actual conditions or obtained by simulating a large amount of data; the size of the proportional coefficient is to quantify each parameter to obtain a specific value for subsequent comparison. The size of the proportional coefficient depends on the amount of sample data and the preliminary setting of the corresponding processing coefficient for each group of sample data by technical personnel in this field; as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0025] Each embodiment in this specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device embodiment, since it is basically based on the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0026] For the convenience of description, the above device is described in various units according to their functions. Of course, when implementing the present application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0027] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0028] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0029] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0030] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0031] Secondly: In the drawings of the embodiments disclosed in the present invention, only the structures related to the embodiments disclosed in the present invention are involved, and other structures can refer to the general design. In the absence of conflict, the same embodiment and different embodiments of the present invention can be combined with each other; Finally: The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An adjustable grassland remote sensing monitoring system, characterized in that: It includes data preprocessing module, change trend analysis module, adaptive remote sensing parameter configuration module, grassland monitoring network layout optimization module and remote sensing observation and intelligent estimation module; Data preprocessing module: obtain historical remote sensing data and ground measured data, preprocess the acquired historical remote sensing data and ground measured data, and obtain data preprocessing results; Change trend analysis module: Combine the data preprocessing results to extract key feature vectors, use the formula to calculate the average change rate and standard deviation for each key feature, and analyze the change trend of each key feature; Adaptive remote sensing parameter configuration module: According to the trend analysis results of key features, the remote sensing monitoring parameters are adjusted by establishing the objective function; Grassland monitoring network layout optimization module: Based on the output of the adaptive remote sensing parameter configuration module, the grassland monitoring area is divided into grid units and equipped with remote sensing detection units, and the layout of the remote sensing monitoring network is adjusted by calculating and analyzing the ecological environment data quality indicators; Remote sensing observation and intelligent estimation module: The semivariogram function is used to analyze grassland observation points, and the spatial interpolation method is used to estimate the values ​​of future observation points.

2. The adjustable grassland remote sensing monitoring system according to claim 1, characterized in that: The data preprocessing module obtains historical remote sensing data and ground measured data, and preprocesses the obtained historical remote sensing data and ground measured data. The process of obtaining the data preprocessing results includes: Get a collection of historical remote sensing data and mark it as ;in, Indicates Historical remote sensing data samples, represents the number of historical remote sensing data samples, represents the number of features of each historical remote sensing data sample; Get the ground truth data set ;in, Indicates Ground truth data samples, is the number of ground truth data samples, is the number of features for each ground truth data sample; The original data matrix is ​​composed of historical remote sensing data and ground-based measured data ,Right now ;in, Represents the total number of data samples, which is the sum of the number of historical remote sensing data samples and the number of ground measured data samples; It represents the total number of features of the data sample, which is the sum of the number of features of the historical remote sensing data and the number of features of the ground measured data; represents the real number threshold; By calculating the covariance matrix Measuring the correlation between different features of the data: , where Represents the transpose of a matrix; Calculate the covariance matrix The characteristic equation of , and get the eigenvalue and the corresponding eigenvector ,in, represents the eigenvalue; represents the identity matrix; Calculate the cumulative contribution rate using the formula , when the cumulative contribution rate reaches the preset threshold, the previous The eigenvectors form the projection matrix ;in, represents the feature vector index, represents the number of eigenvectors determined according to the cumulative contribution rate; Project the original matrix into a lower dimensional space , and obtain a set of features in the low-dimensional space; among them, represents the data matrix after dimensionality reduction, which is the projection result of historical remote sensing data and ground measured data in the direction of the selected eigenvector; Key features are selected from this set of features and used as the result of data preprocessing.

3. The adjustable grassland remote sensing monitoring system according to claim 2, characterized in that: The change trend analysis module extracts key feature vectors based on the data preprocessing results, and uses the formula to calculate the average change rate and standard deviation for each key feature. The process of analyzing the change trend of each key feature includes: Extract key feature vectors and mark them as ; For each key feature , whose time series is ;in, represents the index of the key feature, Indicates time, and , Indicates the total number of time points; calculate key features using the formula In different time windows The average rate of change within: , where Respectively represent the start and end time points of the time window; Calculate key characteristics using formula The standard deviation of , where Representing key features In the time window The average value within , standard deviation To reflect key features In the time window The degree of fluctuation within According to the average rate of change and standard deviation , identify key features In the time window The changing trends within.

4. The adjustable grassland remote sensing monitoring system according to claim 3, characterized in that: The adaptive remote sensing parameter configuration module adjusts the remote sensing monitoring parameters by establishing an objective function based on the trend analysis results of key features. The process includes: Define the remote sensing monitoring parameter vector as ;in, represent the spectral range, resolution, and acquisition time, respectively; Establish the objective function: , where Indicates The weight of the key features; Indicates The expected rate of change of each key characteristic; A weight indicating the degree of volatility; Indicates The expected standard deviation of the key features; According to the objective function, the remote sensing monitoring parameters are adjusted, including the spectral range, resolution and acquisition time; among them, for the adjustment of the spectral range, , where is the initial spectral range, respectively representing the lower and upper limits of the initial spectral range, Represents the learning rate of the spectral range, which determines the step size each time the spectral range is adjusted. is the adjusted spectral range; for resolution adjustment, , where Indicates the adjusted resolution. Indicates the initial resolution, that is, the resolution set at the beginning of remote sensing monitoring. Represents the learning rate of resolution, which is used to control the step size of resolution adjustment; for acquisition time adjustment, , where represents the adjusted acquisition time, represents the initial acquisition time, is the learning rate of the acquisition time, which is used to control the step size of the acquisition time adjustment; Output adjusted remote sensing monitoring parameters .

5. The adjustable grassland remote sensing monitoring system according to claim 1, characterized in that: The grassland monitoring network layout optimization module divides the grassland monitoring area into grid units and configures remote sensing detection units based on the output of the adaptive remote sensing parameter configuration module, and adjusts the layout of the remote sensing monitoring network by calculating and analyzing the ecological environment data quality indicators. The process includes: The grassland monitoring area is defined as , and discretized into grid cells, and the coordinates of each grid cell are ;in, Respectively represent the grid cell indexes of the grassland monitoring area in the north-south and east-west directions, and , They represent the number of grid cells divided into the grassland monitoring area in the north-south and east-west directions respectively; For each grid unit, each grid unit is equipped with a remote sensing detection unit to obtain ecological environment parameters; Get the remote sensing detection unit location collection , where Represents the number of remote sensing detection units; The position of a remote sensing detection unit is expressed as , the location of each remote sensing detection unit is located within the constructed grid unit, that is, , Represents the index of the remote sensing detection unit; Simultaneous remote sensing unit ,satisfy The constraint condition is, Indicates the actual size of the monitoring area of ​​the remote sensing detection unit. Indicates the sensing radius of the remote sensing detection unit; According to the ecological environment parameters obtained by the remote sensing detection unit, calculate the ecological environment data quality index of the remote sensing detection unit : , where are preset different proportional coefficients, which respectively represent the relative importance of vegetation coverage, soil moisture and surface temperature in calculating the ecological environment data quality index; among them, the ecological environment data quality index The vegetation coverage in the ecological environment parameters is used , soil moisture and surface temperature As a measure; Collect the ecological environment data quality indicators within the coverage area of ​​all remote sensing detection units and compare them with the preset standard values ​​in turn; if the ecological environment data quality indicators are lower than the preset standard values, the remote sensing detection unit will be marked for layout optimization; otherwise, it will not be marked; When optimizing the layout of the grassland monitoring network, the coverage of the marked remote sensing detection units is adjusted, including the location, number or perception radius of the remote sensing detection units, in order to optimize the ecological environmental data quality indicators of the entire grassland monitoring area, thereby outputting the optimized grassland remote sensing monitoring network layout.

6. The adjustable grassland remote sensing monitoring system according to claim 1, characterized in that: The remote sensing observation and intelligent estimation module uses the semivariogram function to analyze grassland observation points and uses the spatial interpolation method to estimate the values ​​of future observation points. The process includes: Combined with the optimized grassland remote sensing monitoring network layout information, determine Grassland observation point ; Among them, the observation points represent the observation data at different locations within the grassland monitoring area; Building a semivariogram , where Represents the grassland observation point index, which is used to traverse all grassland observation point pairs involved in the calculation, and ; represents the number of grassland observation points, that is, the number of grassland observation point pairs involved in calculating the semivariogram; Indicates The location of the grassland observation points; Indicates at location The observed value at is the spatial lag distance; By fitting the model to the semivariogram, the parameters of the semivariogram are obtained: ， In the formula, Indicates the value of the membership fee. represents the partial sill value, represents the range; the purpose of fitting the model is to find the parameters of the semivariogram The value of makes the model curve close to the semivariogram value; After obtaining the parameters of the semivariogram model, the spatial interpolation method is used to estimate the values ​​of future observation points: the coordinates of the future observation points to be estimated are marked as , get known grassland observation points ; According to the semivariogram model, the interpolation equations are: , where represents the weight coefficient, represents the Lagrange multiplier; solve the interpolation equations to obtain the weight coefficients ; Using the formula Estimating future observation points The value of The estimated values ​​of future observation points are stored in the background data management center as the basis for adjusting the frequency of grassland remote sensing monitoring.

7. An adjustable grassland remote sensing monitoring method, characterized in that: An adjustable grassland remote sensing monitoring system as claimed in any one of claims 1 to 6, the method comprising: Step 1: Acquire historical remote sensing data and ground measured data, preprocess the acquired historical remote sensing data and ground measured data to obtain data preprocessing results; Step 2: Combine the data preprocessing results to extract key feature vectors, use the formula to calculate the average change rate and standard deviation for each key feature, and analyze the change trend of each key feature; Step 3: According to the trend analysis results of key features, adjust the remote sensing monitoring parameters by establishing the objective function; Step 4: Based on the adjusted remote sensing monitoring parameters, the grassland monitoring area is divided into grid units and equipped with remote sensing detection units, and the layout of the remote sensing monitoring network is adjusted by calculating and analyzing the ecological environment data quality indicators; Step 5: Use the semivariogram function to analyze the grassland observation points and use spatial interpolation to estimate the values ​​of future observation points.

Images