A real-time monitoring system for ecological restoration process

Abstract

The present invention relates to the field of environmental monitoring technology, specifically a real-time monitoring system for ecological restoration process, the system comprising: based on remote sensing technology and ground monitoring, collecting normalized vegetation index, soil chemical composition, water quality index data, calibrating remote sensing data and ground data, adjusting sampling frequency and data resolution, and generating basic environmental data sets. In the present invention, the integration of normalized vegetation index, soil chemical composition and water quality index data improves comprehensiveness and comparability, makes data analysis more accurate, adjusts sampling frequency and data resolution to meet the needs of large-scale or complex environmental monitoring projects, and provides a high degree of flexibility. The application of time window analysis and local time series strengthens the ability to identify environmental change trends and provides a basis for decision-making. By integrating multiple model outputs, parameter settings are optimized, and the consistency and application value of predictions are improved, thereby enhancing the efficiency of project management and the practicality of scientific research data.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring, and in particular to a real-time monitoring system for ecological restoration process. Background Art

[0002] The field of environmental monitoring technology focuses on the use of various sensors, monitoring tools and data analysis technologies to monitor and assess environmental conditions in real time. This includes continuous monitoring of air quality, water quality, soil conditions, biodiversity and other ecological indicators. Environmental monitoring systems can provide key data to help researchers, governments and businesses understand current environmental conditions, predict future changes, and develop effective environmental policies and management measures. This technology uses high-tech equipment such as satellites, drones, ground stations and networked sensor arrays, combined with advanced data processing and model prediction technologies to ensure accurate data collection and processing, thereby supporting the goals of sustainable development and environmental protection.

[0003] Among them, the real-time monitoring system for ecological restoration is a technical system specially designed to monitor and evaluate the progress of ecological restoration projects. The system uses environmental monitoring technology to track ecological parameters of the restoration area in real time, such as vegetation cover, species recovery, soil quality and water status. Through real-time data collection and analysis, this system can help project managers understand the ecological restoration effects in a timely manner and adjust management measures to ensure the success of the restoration project. In addition, the system can also provide data support for scientific research, deepen the understanding of the ecosystem restoration process, and promote the research and development of ecology and environmental science.

[0004] Although existing technologies have made some progress in the field of environmental monitoring, they are still insufficient in data integration and real-time performance. Especially in large-scale and complex environmental projects, existing systems often have data discontinuity and discontinuity problems, which limit the application scope and effectiveness of monitoring data. In addition, existing systems lack responsiveness in terms of flexible adjustment of data sampling frequency and resolution, which leads to an unstable basis for decision-making under changing environmental conditions. Existing technologies have not yet fully realized their potential in dynamically capturing environmental changes and real-time data analysis, which limits timely response and accurate prediction of environmental change trends, resulting in poor implementation of restoration projects and environmental policies. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a real-time monitoring system for the ecological restoration process.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme: A real-time monitoring system for ecological restoration process comprises:

[0007] The ecological monitoring module is based on remote sensing technology and ground monitoring, collects normalized vegetation index, soil chemical composition, water quality index data, calibrates remote sensing data and ground data, adjusts sampling frequency and data resolution, and generates basic environmental data sets;

[0008] The trend analysis module analyzes the change trends of vegetation spectral index and soil and water quality data through the basic environmental data set, uses the time window to analyze the local time series, identifies the time series correlation of key variables, extracts the time and space change characteristics from them, and obtains the dynamic correlation analysis results;

[0009] Based on the dynamic correlation analysis results, the model tuning module adjusts the input parameters of the simulation model, compares the deviation between the model output and the actual monitoring data, optimizes the parameter settings to reduce the error demand, merges the outputs of multiple simulation models, and establishes a comprehensive model optimization result.

[0010] As a further solution of the present invention, the basic environmental data set includes remote sensing spectral data, soil parameter data, and water quality parameter data. The dynamic correlation analysis results are specifically correlation intensity index, time lag characteristics, and spatial distribution maps. The comprehensive model optimization results include model parameter adjustment, prediction accuracy improvement, and strategy optimization solutions.

[0011] As a further solution of the present invention, the method for collecting the normalized vegetation index, soil chemical composition, and water quality index data is specifically as follows:

[0012] Using remote sensing technology and ground monitoring, we capture surface spectral data and ground sample data, process remote sensing images separately to obtain a preliminary normalized vegetation index NDVI, and collect soil and water samples for chemical analysis. The formula for calculating the normalized vegetation index is:

[0013]

[0014] Among them, NIR represents the near-infrared band reflectance obtained from remote sensing images, and Red represents the red light band reflectance;

[0015] Calibrate the preliminary data obtained by remote sensing and the ground measured data, adjust the sampling frequency and data resolution, and simultaneously optimize the data accuracy of the normalized vegetation index, soil chemical composition, and water quality indicators through the calibration method. Apply the calibration formula:

[0016]

[0017] in, represents any preliminary determination value, represents the corresponding ground measured value, is the calibration factor, adjust Value to minimize and The difference between is the data weight parameter, is the adjustment coefficient of the error sum of squares, represents the sum of squared errors, is the adjustment factor for the mean absolute deviation, represents the mean absolute deviation, adjusted Increase the impact of the average difference in data on the calibration process, Indicates calibrated data.

[0018] As a further solution of the present invention, the steps of acquiring the basic environment data set are specifically as follows:

[0019] The calibrated normalized vegetation index, soil chemical composition and water quality index data are integrated, and the remote sensing data and ground monitoring data are integrated through data fusion technology. The data fusion formula is applied:

[0020]

[0021] in, Indicates the basic environment index represents the calibrated normalized difference vegetation index, Calibrated soil chemical composition data, Represents calibrated water quality index data, is the weight coefficient used to adjust NDVI, soil and water quality, , is the weight coefficient, which is used to adjust the impact of time difference on the environmental index. is an exponential decay function, represents the time difference between remote sensing data and ground monitoring data, is the time decay factor, used to adjust the effect of time difference;

[0022] The fused data is verified again to correct data deviations and outliers, optimize data structure, field configuration and data type, generate basic environment data sets, and the standardized results are calculated by the formula:

[0023]

[0024] in, represents the original environment dataset, Indicated in The maximum value of the data points in the dataset, The function is from Capture the maximum value in the data set, Indicated in The minimum value of a data point in the dataset, The function is from Capture the minimum value in the data set, It is the result after standardization.

[0025] As a further solution of the present invention, the step of obtaining the dynamic association analysis result is specifically:

[0026] The vegetation spectral index, soil and water quality data are selected as input variables, and the data in the basic environmental data set are used to calculate the average value and change trend of each variable in the target time window. The calculation formula is:

[0027]

[0028] in, Indicates at a point in time data points, is the number of data points in the time window, It's in time In the window The average value of the data;

[0029] Apply local time series analysis to calculate time series correlation step by step through a sliding window, using the formula:

[0030]

[0031] in, and is time series data of differentiated data types, and is the average value of the corresponding time window, For in time of and relevance;

[0032] Combine the time series analysis results of multiple variables to identify the dynamic associations between key variables, using the association measurement formula:

[0033]

[0034] in, is the normalization factor, is the attenuation parameter, is the current time point, To quantify the dynamic correlation between variables, dynamic correlation analysis results are generated.

[0035] As a further solution of the present invention, the method for obtaining the comprehensive model optimization result is specifically:

[0036] Based on the dynamic correlation analysis results, the input parameters of the simulation model were adjusted, and the parameters including vegetation growth rate, water quality concentration threshold and soil nutrient level were re-evaluated and set, using the parameter adjustment formula:

[0037]

[0038] in is the original parameter, is the target parameter adjusted based on actual data. is the adjustment factor, is the new parameter value;

[0039] Compare the deviation between the model output and the actual monitoring data, and calculate the error of each model. The formula is:

[0040]

[0041] in, is the model output value, is the actual monitoring value, is the model error;

[0042] To reduce the error requirement, optimize the parameter settings, and merge the outputs of multiple simulation models, the weighted average method is used to optimize and fuse the results of the differentiated models, and the comprehensive model optimization formula is:

[0043]

[0044] in, It is The output of the model It is The weight of the model is the result correction factor, For the The corrected constant of the model, is the optimized model output;

[0045] The weight calculation formula uses:

[0046]

[0047] in, It is The error of the model It is The weight of the model, is the regulating factor, is the error adjustment index.

[0048] Compared with the prior art, the advantages and positive effects of the present invention are:

[0049] In the present invention, the accuracy and continuity of environmental monitoring data are improved by integrating remote sensing technology with ground monitoring data. The integration of normalized vegetation index, soil chemical composition and water quality index data improves comprehensiveness and comparability, making data analysis more accurate. Adjusting the sampling frequency and data resolution meets the needs of large-scale or complex environmental monitoring projects and provides a high degree of flexibility. The application of time window analysis and local time series enhances the ability to identify environmental change trends and provides a basis for decision-making. By integrating multiple model outputs, parameter settings are optimized, the consistency of predictions and application value are improved, thereby enhancing the efficiency of project management and the practicality of scientific research data. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a system flow chart of the present invention;

[0051] Figure 2 It is a flow chart of the method for collecting data of normalized vegetation index, soil chemical composition and water quality index of the present invention;

[0052] Figure 3 A flowchart of the steps for obtaining the basic environment data set of the present invention;

[0053] Figure 4 A flowchart of the steps for obtaining the dynamic correlation analysis results of the present invention;

[0054] Figure 5 The present invention is a flowchart of a method for obtaining comprehensive model optimization results. DETAILED DESCRIPTION

[0055] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0056] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0057] Embodiment 1

[0058] See also Figure 1, a real-time monitoring system for ecological restoration process includes:

[0059] The ecological monitoring module is based on remote sensing technology and ground monitoring, collects normalized vegetation index, soil chemical composition, water quality index data, calibrates remote sensing data and ground data, adjusts sampling frequency and data resolution, and generates basic environmental data sets;

[0060] The trend analysis module uses the basic environmental data set to analyze the changing trends of vegetation spectral index and soil and water quality data, uses the time window to analyze the local time series, identifies the time series correlation of key variables, extracts the temporal and spatial change characteristics, and obtains the dynamic correlation analysis results;

[0061] The model tuning module adjusts the input parameters of the simulation model based on the results of dynamic correlation analysis, compares the deviation between the model output and the actual monitoring data, optimizes the parameter settings to reduce the error demand, merges the outputs of multiple simulation models, and establishes a comprehensive model optimization result.

[0062] The basic environmental data sets include remote sensing spectral data, soil parameter data, and water quality parameter data. The dynamic correlation analysis results are specifically the correlation intensity index, time lag characteristics, and spatial distribution maps. The comprehensive model optimization results include model parameter adjustment, prediction accuracy improvement, and strategy optimization solutions.

[0063] See also Figure 2 , the specific method for collecting data on normalized vegetation index, soil chemical composition, and water quality index is as follows:

[0064] Using remote sensing technology and ground monitoring, we capture surface spectral data and ground sample data, process remote sensing images separately to obtain a preliminary normalized vegetation index NDVI, and collect soil and water samples for chemical analysis. The formula for calculating the normalized vegetation index is:

[0065]

[0066] Among them, NIR represents the near-infrared band reflectance obtained from remote sensing images, and Red represents the red light band reflectance;

[0067] Calibrate the preliminary data obtained by remote sensing and the ground measured data, adjust the sampling frequency and data resolution, and simultaneously optimize the data accuracy of the normalized vegetation index, soil chemical composition, and water quality indicators through the calibration method. Apply the calibration formula:

[0068]

[0069] in, represents any preliminary determination value, represents the corresponding ground measured value, is the calibration factor, adjust Value to minimize and The difference between is the data weight parameter, is the adjustment coefficient of the error sum of squares, represents the sum of squared errors, is the adjustment factor for the mean absolute deviation, represents the mean absolute deviation, adjusted Increase the impact of the average difference in data on the calibration process, Indicates calibrated data.

[0070] Calculation of Normalized Difference Vegetation Index (NDVI)

[0071] Use remote sensing technology to obtain spectral data of the earth's surface, with a particular focus on the reflectance of the red and near-infrared bands. represents the reflectivity of red light band, and NIR represents the reflectivity of near infrared band.

[0072] The normalized difference vegetation index is calculated using the following formula:

[0073]

[0074] in, and It is a value directly obtained from remote sensing images.

[0075] Analysis of soil chemical composition

[0076] Soil samples were collected from the ground and the main chemical components in the samples, such as pH value and organic matter content, were determined by chemical analysis methods. Representative results of chemical composition of soil samples.

[0077] The chemical analysis results Record it for subsequent data calibration.

[0078] Determination of water quality indicators

[0079] Collect water samples and conduct water quality tests, including determination of dissolved oxygen, ammonia nitrogen, Value, etc. Represents the measurement results of water quality indicators.

[0080] The water quality test results Record it and use it for subsequent calibration of the data.

[0081] Data calibration process

[0082] Collecting remote sensing data , collect ground truth data .

[0083] Preliminary difference calculation:

[0084] Calculate the difference between remote sensing data and ground data:

[0085] Error calculation:

[0086] Calculate SSE and MAD:

[0087] SSE:

[0088] MAD:

[0089] Calibration parameters determination:

[0090] Determine the weight coefficient and the weight vector These are derived using optimization methods (such as regression analysis, machine learning, etc.) on a training dataset.

[0091] Substitute the above parameters and calculation results into the calibration formula to obtain the calibrated data .

[0092] Assume the following data and parameters:

[0093] (Remote sensing normalized difference vegetation index)

[0094] =0.6 (ground normalized difference vegetation index)

[0095]

[0096] (Here it is assumed that only a single indicator is calibrated)

[0097]

[0098] Substitute the calibration formula into the numerical calculation:

[0099]

[0100] See also Figure 3 ,The specific steps for obtaining the basic environment dataset are:

[0101] The calibrated normalized vegetation index, soil chemical composition and water quality index data are integrated, and the remote sensing data and ground monitoring data are integrated through data fusion technology. The data fusion formula is applied:

[0102]

[0103] in, represents the basic environment index, represents the calibrated normalized difference vegetation index, Calibrated soil chemical composition data, Represents calibrated water quality index data, is the weight coefficient used to adjust NDVI, soil and water quality, is the weight coefficient, which is used to adjust the impact of time difference on the environmental index. is an exponential decay function, represents the time difference between remote sensing data and ground monitoring data, is the time decay factor, used to adjust the effect of time difference;

[0104] The fused data is verified again to correct data deviations and outliers, optimize data structure, field configuration and data type, generate basic environment data sets, and the standardized results are calculated by the formula:

[0105]

[0106] in, represents the original environment dataset, Indicated in The maximum value of the data points in the dataset, The function is from Capture the maximum value in the data set, Indicated in The minimum value of the data points in the dataset, The function is from Capture the minimum value in the data set, It is the result after standardization.

[0107]

[0108] Calibration data: Calibrate the original data using a standardized method or empirical calibration formula , soil and water quality data.

[0109] Data weighting: Calculate the weighted sum.

[0110] Apply Time Decay Function: Calculates the exponentially decaying effect of time differences.

[0111] Integrate the data: Combine the weighted data with the time decay function to get .

[0112]

[0113] Find the extreme value: From Calculate the maximum and minimum values ​​in the data set.

[0114] Calculate the denominator: maximum value minus minimum value.

[0115] Standardization: Each The value is divided by the range to obtain a number between 0 and 1.

[0116] Algorithm calculation and process

[0117] The above calculations will be demonstrated using a hypothetical data set:

[0118] Assumptions:

[0119]

[0120]

[0121]

[0122]

[0123]

[0124]

[0125]

[0126] The calculation is as follows:

[0127] Compute the time decay term:

[0128]

[0129] Compute the sum in the denominator:

[0130]

[0131] Calculate the weighted average:

[0132]

[0133] Apply the renormalization formula, assuming that the calculated The value is 50 and the range is 100:

[0134]

[0135] See also Figure 4 , the specific steps for obtaining the dynamic association analysis results are:

[0136] The vegetation spectral index, soil and water quality data are selected as input variables, and the data in the basic environmental data set are used to calculate the average value and change trend of each variable in the target time window. The calculation formula is:

[0137]

[0138] in, Indicates at a point in time data points, is the number of data points in the time window, It's in time In the window The average value of the data;

[0139] Apply local time series analysis to calculate time series correlation step by step through a sliding window, using the formula:

[0140]

[0141] in, and is time series data of differentiated data types, and is the average value of the corresponding time window, For in time of and relevance;

[0142] Combine the time series analysis results of multiple variables to identify the dynamic associations between key variables, using the association measurement formula:

[0143]

[0144] in, is the normalization factor, is the attenuation parameter, is the current time point, To quantify the dynamic correlation between variables, dynamic correlation analysis results are generated.

[0145] Calculates the mean of a variable within a time window

[0146] First, it is necessary to calculate the average values ​​of various environmental variables (such as vegetation spectral index, soil and water quality data) within the time window. This calculation process involves the following steps:

[0147] Assumptions:

[0148] Indicates at time Vegetation spectral index data points.

[0149] is the number of data points in the time window.

[0150] Calculation process:

[0151] Determine the time window (e.g. from arrive ).

[0152] For each time point Data points within , perform cumulative summation:

[0153]

[0154] Calculate the average:

[0155]

[0156] Calculating time series correlation

[0157] Next, we need to calculate the time series correlation between different data types within a specific time window. This step involves the following calculations:

[0158] Assumptions:

[0159] and are two different data types (such as vegetation and soil) in time The sequence data.

[0160] and They are and The average value over a specific time window.

[0161] Calculation process:

[0162] Determine the time window and calculate and The average value of .

[0163] Calculate the difference between the data points at each time point:

[0164]

[0165] Multiply and sum the differences:

[0166]

[0167] Cumulative sum of the squared differences:

[0168]

[0169] Calculate the correlation coefficient:

[0170]

[0171] Comprehensive correlation to obtain dynamic correlation analysis results

[0172] Finally, the obtained time series correlations are combined to obtain the dynamic correlation analysis results:

[0173] Correlation coefficient for all time windows Perform cumulative summation.

[0174] Apply the simplified formula to calculate the dynamic correlation analysis results:

[0175]

[0176] See also Figure 5 , the method for obtaining the comprehensive model optimization results is as follows:

[0177] Based on the results of the dynamic correlation analysis, the input parameters of the simulation model were adjusted, and the parameters including vegetation growth rate, water quality concentration threshold and soil nutrient level were re-evaluated and set, using the parameter adjustment formula:

[0178]

[0179] in is the original parameter, is the target parameter adjusted based on actual data. is the adjustment factor, is the new parameter value;

[0180] Compare the deviation between the model output and the actual monitoring data, and calculate the error of each model. The formula is:

[0181]

[0182] in, is the model output value, is the actual monitoring value, is the model error;

[0183] To reduce the error requirement, optimize the parameter settings, and merge the outputs of multiple simulation models, the weighted average method is used to optimize and fuse the results of the differentiated models, and the comprehensive model optimization formula is:

[0184]

[0185] in, It is The output of the model It is The weight of the model is the result correction factor, For the The corrected constant of the model, is the optimized model output;

[0186] The weight calculation formula uses:

[0187]

[0188] in, It is The error of the model It is The weight of the model, is the regulating factor, is the error adjustment index.

[0189] Get and :

[0190] Obtained from existing models.

[0191] Optimize target determination based on new research data or models.

[0192] Calculate new parameters :

[0193]

[0194] is a scaling factor that is preset or adjusted based on experience.

[0195] collect and :

[0196] Obtained from model run results.

[0197] Obtained from field monitoring or experimental data.

[0198] Calculation error :

[0199]

[0200] The sum is traversed over all data points, the sum of the squares of the differences between the model output and the actual data is calculated, and finally the square root is taken.

[0201] Calculating weights for a single model :

[0202]

[0203] in It is The error of the model, Adjusts the sensitivity of the weights to error.

[0204] Merge multiple model outputs:

[0205]

[0206] It is The output of a model.

[0207] is a model correction term, such as that set according to specific requirements.

[0208] is the weight coefficient of the correction term.

[0209] Suppose there are two models and , we need to calculate the merged output:

[0210]

[0211]

[0212]

[0213]

[0214] First calculate the weights:

[0215]

[0216] Then merge the model outputs:

[0217]

[0218] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A real-time monitoring system for ecological restoration process, characterized in that: The system comprises: The ecological monitoring module is based on remote sensing technology and ground monitoring, collects normalized vegetation index, soil chemical composition, water quality index data, calibrates remote sensing data and ground data, adjusts sampling frequency and data resolution, and generates basic environmental data sets; The trend analysis module analyzes the change trends of vegetation spectral index and soil and water quality data through the basic environmental data set, uses the time window to analyze the local time series, identifies the time series correlation of key variables, extracts the time and space change characteristics from them, and obtains the dynamic correlation analysis results; The steps for obtaining the dynamic association analysis results are specifically as follows: The vegetation spectral index, soil and water quality data are selected as input variables, and the data in the basic environmental data set are used to calculate the average value and change trend of each variable in the target time window. The calculation formula is: in, Indicates at a point in time data points, is the number of data points in the time window, It's in time In the window The average value of the data; Apply local time series analysis to calculate time series correlation step by step through a sliding window, using the formula: in, and is time series data of differentiated data types, and is the average value of the corresponding time window, For in time of and relevance; Combine the time series analysis results of multiple variables to identify the dynamic associations between key variables, using the association measurement formula: in, is the normalization factor, is the attenuation parameter, is the current time point, To quantify the dynamic correlation between variables, dynamic correlation analysis results are generated; The model tuning module adjusts the input parameters of the simulation model based on the dynamic correlation analysis results, compares the deviation between the model output and the actual monitoring data, optimizes the parameter settings to reduce the error requirements, merges the outputs of multiple simulation models, and establishes a comprehensive model optimization result; The method for obtaining the comprehensive model optimization result is specifically as follows: Based on the dynamic correlation analysis results, the input parameters of the simulation model were adjusted, and the parameters including vegetation growth rate, water quality concentration threshold and soil nutrient level were re-evaluated and set, using the parameter adjustment formula: in is the original parameter, is the target parameter adjusted based on actual data. is the adjustment factor, is the new parameter value; Compare the deviation between the model output and the actual monitoring data, and calculate the error of each model. The formula is: in, is the model output value, is the actual monitoring value, is the model error; To reduce the error requirement, optimize the parameter settings, and merge the outputs of multiple simulation models, the weighted average method is used to optimize and fuse the results of the differentiated models, and the comprehensive model optimization formula is: in, It is The output of the model It is The weight of the model is the result correction factor, For the The corrected constant of the model, is the optimized model output; The weight calculation formula uses: in, It is The error of the model It is The weight of the model, is the regulating factor, is the error adjustment index.

2. The real-time monitoring system for ecological restoration process according to claim 1 is characterized in that: The method for collecting the normalized vegetation index, soil chemical composition, and water quality index data is specifically as follows: Using remote sensing technology and ground monitoring, we captured surface spectral data and ground sample data, processed remote sensing images to obtain a preliminary Normalized Difference Vegetation Index (NDVI), and collected soil and water samples for chemical analysis; Calibrate the preliminary data obtained by remote sensing and the ground measured data, adjust the sampling frequency and data resolution, and optimize the data accuracy of the normalized vegetation index, soil chemical composition, and water quality indicators through the calibration method. Apply the calibration formula: in, represents any preliminary determination value, represents the corresponding ground measured value, is the calibration factor, adjust Value to minimize and The difference between is the data weight parameter, is the adjustment coefficient of the error sum of squares, represents the sum of squared errors, is the adjustment factor for the mean absolute deviation, represents the mean absolute deviation, adjusted Increase the impact of the average difference in data on the calibration process, Indicates calibrated data.

3. The real-time monitoring system for ecological restoration process according to claim 1 is characterized in that: The steps for obtaining the basic environment data set are specifically as follows: The calibrated normalized vegetation index, soil chemical composition and water quality index data are integrated, and the remote sensing data and ground monitoring data are integrated through data fusion technology. The data fusion formula is applied: in, represents the basic environment index, represents the calibrated normalized difference vegetation index, Calibrated soil chemical composition data, Represents calibrated water quality index data, is the weight coefficient used to adjust NDVI, soil and water quality, is the weight coefficient, which is used to adjust the impact of time difference on the environmental index. is an exponential decay function, represents the time difference between remote sensing data and ground monitoring data, is the time decay factor, used to adjust the effect of time difference; The fused data is verified again to correct data deviations and outliers, optimize data structure, field configuration and data type, generate basic environment data sets, and the standardized results are calculated by the formula: in, represents the original environment dataset, Indicated in The maximum value of the data points in the dataset, The function is from Capture the maximum value in the data set, Indicated in The minimum value of a data point in the dataset, The function is from Capture the minimum value in the data set, It is the result after standardization.

Images