Intelligent forest grass remote sensing monitoring method based on GIS and big data

Abstract

The invention relates to the technical field of remote sensing monitoring, in particular to an intelligent forest and grass remote sensing monitoring method based on a GIS and big data, which comprises the following steps: acquiring coordinates of a forest coverage area based on remote sensing data, performing overlay analysis on the data in combination with the GIS, and identifying boundary and internal structure features of the forest area. And calculating the number of connected components and holes, and generating a primary topological feature. According to the method, the coordinates of the forest coverage area are accurately captured by combining the GIS and remote sensing data, and the data are subjected to overlay analysis, so that the recognition of boundary and internal structure features is finer, the space-time analysis capability of the data is expanded by utilizing the dynamically adjusted clustering granularity, the flexibility of data processing is improved, and the accuracy of data processing is improved. The introduction of the recurrent neural network optimizes the accuracy of data prediction, enables the periodic change and seasonal adjustment to be accurately responded, monitors the key ecological change through a multi-scale time window, and improves the monitoring precision of the ecological parameter change.

Description

Technical Field

[0001] The present invention relates to the field of remote sensing monitoring technology, and in particular to a smart forest and grassland remote sensing monitoring method based on GIS and big data. Background Art

[0002] Remote sensing monitoring technology is a method of using sensors on satellites or aircraft to collect data from a distance to monitor and analyze the earth's surface and its environmental conditions. This technology can observe and analyze the target without directly contacting it by capturing radiation of different wavelengths. It has a wide range of applications, including environmental monitoring, agricultural assessment, forestry management, urban planning, etc. Remote sensing technology can provide continuous time series data, which is of great significance for understanding environmental changes, resource distribution and the impact of natural disasters.

[0003] Among them, the smart forest and grassland remote sensing monitoring method refers to a technical means of real-time monitoring and management of forest and grassland resources using advanced remote sensing technology combined with geographic information system (GIS) and big data analysis. This method can effectively monitor changes in forest and grassland coverage, the occurrence of pests and diseases, and the health of vegetation, thereby guiding the formulation of forestry management and protection measures. The main purpose of smart forest and grassland remote sensing monitoring is to improve the efficiency and accuracy of forest resource management, protect the ecological environment, and promote sustainable development.

[0004] Although traditional remote sensing monitoring technology can monitor environmental conditions, it is insufficient in real-time and multi-scale temporal analysis, especially in capturing ecological changes in the short term. This limitation may lead to untimely assessments of forest health, which in turn affects the effective response to forest degradation or pests and diseases. In addition, traditional methods are inefficient in integrating different data sources and processing large-scale data sets, which limits their application potential in the modern big data environment, hinders accurate resource management and decision-making, and reduces the sustainability of ecosystem management. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and propose a smart forest and grassland remote sensing monitoring method based on GIS and big data.

[0006] In order to achieve the above purpose, the present invention adopts the following technical solution: a smart forest and grass remote sensing monitoring method based on GIS and big data, comprising the following steps:

[0007] S1: Based on remote sensing data, the coordinates of forest-covered areas are obtained, and the data are overlaid and analyzed with GIS to identify the boundaries and internal structural features of the forest area, calculate the number of connected components and holes, and generate primary topological features;

[0008] S2: taking the primary topological features as input, setting the scale parameters of the analysis, dynamically adjusting the clustering granularity, covering the analysis with multiple time scales including days, months and years, performing structural clustering, and generating a clustering topology map;

[0009] S3: According to the cluster topology, set the parameters of the recursive neural network model, perform deep learning on the characteristics of each cluster, adjust the recursive neural network model to match the periodic changes and seasonal adjustments, and generate time series analysis results;

[0010] S4: Using the time series analysis results, set multi-scale time windows, define vegetation coverage as an ecological indicator, observe key ecological changes, subdivide the changes in ecological parameters within each time window, identify key ecological change time nodes, including vegetation growth peaks and degradation rates, and generate ecological change indicators;

[0011] S5: Combined with the above ecological change indicators, spatial analysis is performed using GIS data to assess forest health, adjust and optimize model parameters to match differentiated environmental conditions, and generate forest health assessment results.

[0012] The primary topological features specifically include connected components and the number of holes; the clustering topology diagram specifically includes structured clustering results and time scale clustering; the time series analysis results include periodic changes and seasonal adjustments; the ecological change indicators specifically include vegetation growth peak and degradation rate; the forest health assessment results include vegetation coverage, changes in ecological parameters, and environmental adaptability.

[0013] As a further solution of the present invention, the step of acquiring the primary topological features is specifically as follows:

[0014] S111: Obtain the coordinates of forest-covered areas from remote sensing data, spatially overlay the coordinate data, analyze the data through a geographic information system, generate a geographic information model of the forest area, and obtain preliminary features of the forest boundary and internal structure;

[0015] S112: using the preliminary features of the forest boundary and internal structure, performing spectral analysis on the forest boundary, using edge detection technology to determine the location of the forest boundary, and applying texture analysis to the internal structure to identify differentiated tree species and dense areas, and calculating the number of connected components and holes;

[0016] S113: Using the number of connected components and holes, through structural analysis and correlation evaluation of the number of nodes, a weighted formula is adopted:

[0017]

[0018] Calculate and generate primary topological features;

[0019] Among them, w i Represents the relative critical weight of each connected component, which is used to weight the contribution of each node, x i represents the number of nodes within a component, reflecting the size of the component, and n is the total number of connected components included in the analysis, which is used for normalization.

[0020] As a further solution of the present invention, the step of obtaining the cluster topology map is specifically as follows:

[0021] S211: Based on the primary topological features, analysis scale parameters are set, daily, monthly and annual data sets are integrated, and time data aggregation is performed to obtain multi-scale time aggregation data;

[0022] S212: dynamically adjusting the granularity of the clustering algorithm by using the multi-scale time aggregation data, determining parameter optimization and clustering centers of the clustering algorithm, and obtaining an adjusted clustering parameter set;

[0023] S213: Perform structural clustering according to the adjusted clustering parameter set, using the formula:

[0024]

[0025] Calculate and obtain the topological structure score of each cluster and generate a cluster topology graph;

[0026] Where C(i) represents the topological structure score of the i-th cluster, d ij is the distance between the i-th cluster and the j-th data point, α j is the adjustment coefficient, reflecting the criticality of distance in each cluster, and m is the total number of data points in the cluster.

[0027] As a further solution of the present invention, the step of obtaining the time series analysis result is specifically:

[0028] S311: receiving the cluster topology map, adjusting the parameters of the recursive neural network model, initializing the network weights and biases according to the multiple clustering features, configuring the network to match the characteristics of the input data, and generating a preliminary network configuration;

[0029] S312: Based on the preliminary network configuration, feature learning is performed for multiple clusters, neural network parameters including weights and biases are adjusted according to periodic changes to match seasonal changes, and the adjusted network parameters are output;

[0030] S313: Utilize the adjusted network parameters to perform sequence analysis using the formula:

[0031]

[0032] Generate time series analysis results;

[0033] Among them, S(t) is the output sequence at time t, which represents the time-dependent result of the model output, β k represents the period adjustment coefficient, which is used to adjust the model's response sensitivity to changes in the differentiated time period, γ k Represents the growth rate, which is used to adjust the impact of time growth on output, σ k is the normalization factor used for data standardization, and t represents the time variable.

[0034] As a further solution of the present invention, the steps for obtaining the ecological change index are specifically as follows:

[0035] S411: Using the time series analysis results, setting a multi-scale time window, including daily, monthly, and annual data views, indexing changes in ecological parameters at multiple time levels, and generating a time window configuration;

[0036] S412: Based on the time window configuration, define vegetation coverage as an ecological indicator, record the growth and decline trends of vegetation by comparing data in multiple time windows, and generate ecological change observation data;

[0037] S413: Based on the ecological change observation data, identify key ecological change time nodes, including vegetation growth peak and degradation rate, using the formula:

[0038]

[0039] Generate indicators of ecological change;

[0040] Among them, ΔV(t) represents the change in vegetation coverage at time t, which is used to measure ecological changes in the target time period. Δt represents the time interval for the change, which is the time basis for calculating the rate of change. λ is the baseline adjustment coefficient, which is used to adjust the basic value of the model's sensitivity to differentiated ecological data.

[0041] As a further solution of the present invention, the steps for obtaining the forest health assessment results are specifically as follows:

[0042] S511: Use GIS data to conduct spatial analysis, integrate the ecological change indicators into the geographic information system, set a multi-scale analysis benchmark for forest ecological changes, and generate a spatial analysis configuration;

[0043] S512: Based on the spatial analysis configuration, analyze the change of forest vegetation coverage, evaluate vegetation growth and decline through GIS, and determine key ecological change indicators to generate preliminary forest health assessment data;

[0044] S513: Apply the preliminary forest health assessment data to optimize model parameters to match changes in environmental conditions, using the formula:

[0045]

[0046] Produce forest health assessment results;

[0047] Among them, F(x) represents the forest health assessment result, which means the overall health status of the forest obtained by calculation, and c i is the parameter adjustment coefficient, which is used to adjust the influence of multiple ecological change indicators in the overall assessment. i Represents multiple ecological indicators extracted from GIS data, including vegetation coverage, soil and water conservation capacity, a i is the variable weight, which adjusts the model's sensitivity to changes in indicators, and b i is the variable adjustment power, which is used to enhance or weaken the impact of the target indicator and is adjusted according to the contribution of the indicator to forest health. i is the offset factor used to balance the baseline effect.

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

[0049] In the present invention, by combining GIS and remote sensing data, the coordinates of forest-covered areas are accurately captured, and these data are overlaid and analyzed, so that the identification of boundaries and internal structural features is more refined. The dynamically adjusted clustering granularity is used to expand the spatiotemporal resolution of data and improve the flexibility of data processing. The introduction of recursive neural networks optimizes the accuracy of data prediction, allowing accurate response to periodic changes and seasonal adjustments. By monitoring key ecological changes through multi-scale time windows, the monitoring accuracy of ecological parameter changes is improved, effectively guiding the management and protection of forest resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic diagram of the main steps of the present invention;

[0051] Figure 2 A flow chart of the steps for obtaining the primary topological features of the present invention;

[0052] Figure 3 A flow chart of the steps for obtaining a cluster topology map of the present invention;

[0053] Figure 4 A flow chart of the steps for obtaining the time series analysis results of the present invention;

[0054] Figure 5 A flow chart of the steps for obtaining ecological change indicators of the present invention;

[0055] Figure 6This is a flow chart of the steps for obtaining the forest health assessment results of the present invention. DETAILED DESCRIPTION

[0056] 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.

[0057] 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.

[0058] Embodiment 1

[0059] See also Figure 1 The present invention provides a technical solution: a smart forest and grass remote sensing monitoring method based on GIS and big data, comprising the following steps:

[0060] S1: Based on remote sensing data, the coordinates of forest-covered areas are obtained, and the data are overlaid and analyzed with GIS to identify the boundaries and internal structural features of the forest area, calculate the number of connected components and holes, and generate primary topological features;

[0061] S2: Taking the primary topological features as input, setting the scale parameters of the analysis, dynamically adjusting the clustering granularity, covering the analysis on multiple time scales including days, months and years, performing structural clustering, and generating a clustering topology map;

[0062] S3: According to the cluster topology, set the parameters of the recursive neural network model, perform deep learning on the characteristics of each cluster, adjust the recursive neural network model to match the periodic changes and seasonal adjustments, and generate time series analysis results;

[0063] S4: Use the results of time series analysis to set multi-scale time windows, define vegetation coverage as an ecological indicator, observe key ecological changes, subdivide the changes in ecological parameters within each time window, identify key ecological change time nodes, including vegetation growth peaks and degradation rates, and generate ecological change indicators;

[0064] S5: Combine ecological change indicators and use GIS data for spatial analysis to assess forest health, adjust and optimize model parameters to match differentiated environmental conditions, and generate forest health assessment results.

[0065] The primary topological features include connected components and the number of holes. The clustering topology diagram includes structured clustering results and time scale clustering. The time series analysis results include periodic changes and seasonal adjustments. The ecological change indicators include vegetation growth peak and degradation rate. The forest health assessment results include vegetation coverage, changes in ecological parameters, and environmental adaptability.

[0066] See also Figure 2 , the steps for obtaining primary topological features are as follows:

[0067] S111: Obtain the coordinates of forest-covered areas from remote sensing data, spatially overlay the coordinate data, analyze the data through a geographic information system, generate a geographic information model of the forest area, and obtain preliminary features of the forest boundary and internal structure;

[0068] Spatial overlay analysis of remote sensing data, processing data in the GIS system to obtain preliminary characteristics of forest boundaries and internal structures. The analysis involves basic data overlay, detecting and determining the location and shape of boundaries through spatial analysis, and identifying the boundary between vegetation and non-vegetation areas based on the differentiated wavelength information reflected by multispectral images. This analysis also includes vegetation density detection, which is carried out by comparing the differences in spectral reflectance. The reflectance of densely vegetated areas is higher than that of open areas. Using these data, the GIS system can output a highly detailed forest structure map, which lays the foundation for the next step of analysis and obtains preliminary characteristics of forest boundaries and internal structures.

[0069] S112: Using preliminary features of forest boundaries and internal structures, spectral analysis of forest boundaries is performed, edge detection techniques are used to determine the location of forest boundaries, and texture analysis is applied to internal structures to identify differentiated tree species and dense areas, and the number of connected components and holes is calculated;

[0070] Based on the preliminary feature data of the forest boundary and internal structure, the refined boundary identification process involves using edge detection technology to determine the location. Edge detection determines the boundary by analyzing the changing intensity of pixel color in the image. This method is particularly effective in determining the edges of trees. In addition, texture analysis technology is applied to the identification of the internal structure of the forest. Texture analysis relies on the repeatability of similar patterns in the image. Different tree species and forests under various growth conditions show differentiated texture characteristics. These technologies combined not only enhance the accuracy of image recognition, but also provide a measurement of the number of connected components and holes. Calculating these data provides key quantitative information for forest management and obtains the number of connected components and holes.

[0071] S113: Using the number of connected components and holes, a weighted formula is used through structural analysis and correlation evaluation of the number of nodes:

[0072]

[0073] Calculate and generate primary topological features;

[0074] Among them, w i Represents the relative critical weight of each connected component, which is used to weight the contribution of each node, x i represents the number of nodes within a component, reflecting the size of the component, and n is the total number of connected components included in the analysis, which is used for normalization.

[0075] formula:

[0076]

[0077] The benefit of the formula is that by introducing a weighting mechanism and absolute value, the impact of outliers can be reduced when dealing with noisy data, while normalization ensures that connected components of different scales can be considered fairly.

[0078] Detailed explanation of the formula and the process of formula calculation and derivation:

[0079] Set the number of connected components n to 100 and assume a scenario where the critical weights of all connected components are w i are all set to 1, and each connected component includes the number of nodes x i Assuming it is 50, the calculation process is as follows:

[0080]

[0081] The results show that the average contribution of each connected component to the overall structure is quantified as 7.07, and the numerical value can help analyze the uniformity and connectivity of forest structure, thereby optimizing forest management and protection strategies.

[0082] See also Figure 3 , the specific steps for obtaining the cluster topology map are:

[0083] S211: Based on the primary topological features, set the analysis scale parameters, integrate the daily, monthly and annual data sets, and perform time data aggregation to obtain multi-scale time aggregation data;

[0084] Based on the primary topological features, the analysis scale parameters are set, the daily, monthly, and annual data sets are integrated, and time data aggregation is performed. The data are grouped by the target time window, and the data in each window are merged by average calculation to ensure data consistency and feasibility of analysis. Multi-scale time aggregation data is obtained. This data set becomes the basis for iterative clustering analysis and will provide the necessary input data for the next step of dynamic adjustment.

[0085] S212: using multi-scale time aggregation data, dynamically adjusting the granularity of the clustering algorithm, determining parameter optimization and clustering centers of the clustering algorithm, and obtaining an adjusted clustering parameter set;

[0086] The granularity of the clustering algorithm is dynamically adjusted by using multi-scale time aggregation data, including parameter optimization of the clustering algorithm and determination of the clustering center. By analyzing the changing trend of the aggregation data in real time, the sensitivity and threshold parameters in the clustering algorithm are dynamically adjusted to ensure the accuracy of the clustering results. The adjusted clustering parameter set is obtained, which will directly affect the quality of clustering and the output results.

[0087] S213: Perform structural clustering according to the adjusted clustering parameter set, using the formula:

[0088]

[0089] Calculate and obtain the topological structure score of each cluster and generate a cluster topology graph;

[0090] Where C(i) represents the topological structure score of the i-th cluster, d ij is the distance between the i-th cluster and the j-th data point, α j is the adjustment coefficient, reflecting the criticality of distance in each cluster, and m is the total number of data points in the cluster.

[0091] formula:

[0092]

[0093] The benefit of the formula is that by introducing the relative critical weight of each connected component and the number of nodes included in the multi-components, the influence of differentiated data points in the clustering process can be flexibly adjusted, thereby more accurately reflecting the true topological relationship of the structural clustering.

[0094] Detailed explanation of the formula and the process of formula calculation and derivation:

[0095] Set the number of connected components to m = 5, and the distance from the multi-connected component to the cluster center to d ij =

[0096] [2, 5, 1, 3, 4], with a relative critical weight of α j=[0.5, 0.2, 1.0, 0.3, 0.4], calculate the adjustment value of the multi-component weight and the distance weighted value, and substitute them into the formula one by one:

[0097]

[0098] The results show that the average score of the weight-adjusted structural clustering is 2.87, which reflects the average topological influence of multiple components in the clustering structure. The clustering topology diagram describes the relationships between differentiated components and the contribution of each component to the overall structure.

[0099] See also Figure 4 , the specific steps for obtaining the time series analysis results are:

[0100] S311: receiving a cluster topology map, adjusting parameters of a recursive neural network model, initializing network weights and biases according to multiple clustering features, configuring the network to match characteristics of the input data, and generating a preliminary network configuration;

[0101] Based on the cluster topology diagram, parameter adjustment begins with defining the initial weight values ​​of each cluster node. These values ​​are adjusted according to the connectivity of the nodes and the cluster density. Through the forward propagation and back propagation process of the network, the weights and biases are gradually optimized to ensure that the model can capture the key patterns in the topological structure. This process is iterated until the network parameters are stable and a preliminary network configuration plan is generated.

[0102] S312: Based on the preliminary network configuration, feature learning is performed for multiple clusters, and the neural network parameters, including weights and biases, are adjusted according to periodic changes to match seasonal changes, and the adjusted network parameters are output;

[0103] In this step, feature learning depends on the quality of clustering results. First, the consistency of data within each cluster is evaluated, and then the weight is adjusted according to the variability of the data. The weight adjustment refers to the periodicity and trend of the reference data. If a cluster shows obvious activity in winter, the response weight to this period is increased. The weight adjustment is refined through the gradient descent method. The output parameters reflect the behavior pattern of each cluster in differentiated seasons, laying a solid foundation for subsequent time series analysis.

[0104] S313: Perform sequence analysis using the adjusted network parameters using the formula:

[0105]

[0106] Generate time series analysis results;

[0107] Among them, S(t) is the output sequence at time t, which represents the time-dependent result of the model output, β krepresents the period adjustment coefficient, which is used to adjust the model's response sensitivity to changes in the differentiated time period, γ k Represents the growth rate, which is used to adjust the impact of time growth on output, σ k is the normalization factor used for data standardization, and t represents the time variable.

[0108] formula:

[0109]

[0110] The benefit of the formula is that by adjusting the coefficient β k , growth rate γ k , and the normalization factor σ k With the participation of , the formula can effectively match the periodic changes of time series data, improve the accuracy of prediction and the response of the model to time sensitivity, which is crucial for matching in a dynamic data environment.

[0111] Detailed explanation of the formula and the process of formula calculation and derivation:

[0112] The following parameter values ​​are set, β k =0.3,γ k =0.05,σ k =2, and t=10. Then the calculation process of the formula is:

[0113]

[0114] The results show that at time t=10, the model prediction output value is 0.1428, which means that the model has a good match and response to the behavior of the time point, which is obtained by adjusting the target growth rate and adjustment coefficient, reflecting the effectiveness of the model in processing actual time series data.

[0115] See also Figure 5 , the specific steps for obtaining ecological change indicators are:

[0116] S411: Using the time series analysis results, set up multi-scale time windows, including daily, monthly, and annual data views, index the changes in ecological parameters at multiple time levels, and generate time window configurations;

[0117] Setting up multi-scale time windows involves strategies for comprehensive analysis and adjustment of time series data. According to differentiated ecological change needs, multiple time levels of days, months, and years are defined to capture the micro and macro changes in ecological dynamics. Through time segmentation, seasonal growth patterns and sudden ecological events can be effectively distinguished. This method provides a framework for synchronizing and comparing ecological data at differentiated time scales, providing support for long-term ecological research. At the same time, the configuration of time windows should refer to the availability of data and the periodicity of ecological phenomena to ensure that the set time windows can effectively cover key ecological changes.

[0118] S412: Based on the time window configuration, define the vegetation coverage rate as an ecological indicator, record the growth and decline trends of vegetation by comparing the data in multiple time windows, and generate ecological change observation data;

[0119] In the process of defining vegetation coverage as an ecological indicator, it is first necessary to measure the vegetation coverage through existing remote sensing data or data collected on-site. By analyzing the data, the growth or decline trend of vegetation in differentiated time windows can be determined, including comparing vegetation data at consecutive time points and evaluating the increase or decrease in vegetation coverage. Such analysis helps to identify the health status of the ecosystem and changes in biodiversity, which is crucial for the formulation of management and protection strategies. Changes in vegetation coverage can indicate the effectiveness of soil and water conservation, the impact of climate change, and interference from human activities.

[0120] S413: Based on the ecological change observation data, identify the key ecological change time nodes, including vegetation growth peak and degradation rate, using the formula:

[0121]

[0122] Generate indicators of ecological change;

[0123] Among them, ΔV(t) represents the change in vegetation coverage at time t, which is used to measure ecological changes in the target time period. Δt represents the time interval for the change, which is the time basis for calculating the rate of change. λ is the baseline adjustment coefficient, which is used to adjust the basic value of the model's sensitivity to differentiated ecological data.

[0124] formula:

[0125]

[0126] The benefit of the formula is that by combining the rate of change of vegetation coverage and the time adjustment factor, the trend of vegetation change can be predicted more accurately, and the influence of time on the rate of change can be referred to, so that the model can match differentiated ecological monitoring cycles.

[0127] Detailed explanation of the formula and the process of formula calculation and derivation:

[0128] Set the target time window, the vegetation coverage rate increases from 50% to 55%, the time span is 5 years, then ΔV(t) = 55-50 = 5, Δt = 5. Set λ = 0.1 and δ = 0.03, and calculate:

[0129]

[0130] The results showed that the rate of vegetation cover degradation increased by 10.4% per year over a given period of time. This information is crucial for environmental scientists and policymakers to evaluate the effectiveness of ecological protection measures or adjust related policies.

[0131] See also Figure 6 , the specific steps for obtaining forest health assessment results are:

[0132] S511: Use GIS data for spatial analysis, integrate ecological change indicators into the geographic information system, set multi-scale analysis benchmarks for forest ecological change, and generate spatial analysis configurations;

[0133] Using GIS data for spatial analysis, integrating ecological change indicators into the geographic information system, and setting a multi-scale analysis benchmark for forest ecological change, this process first involves acquiring a variety of GIS data, including satellite imaging, topographic maps, and vegetation distribution maps. The data is collected through remote sensing technology and integrated and analyzed through GIS. During the integration process, the data is corrected through the spatial data management system to ensure data accuracy and validity, provide basic data for assessing forest health, and generate spatial analysis configurations. The key to this process is to match multiple data in time and space, and use matching algorithms to analyze and extract ecological change indicators.

[0134] S512: Based on spatial analysis configuration, analyze changes in forest vegetation cover, assess vegetation growth and decline through GIS, and identify key ecological change indicators to generate preliminary forest health assessment data;

[0135] Based on the configured spatial analysis foundation, the changes in forest vegetation coverage are analyzed, and GIS is used to evaluate vegetation growth and decline. First, it is necessary to conduct a time series analysis of the vegetation coverage in the GIS data to identify growth or decline trends. This involves comparing vegetation coverage data at consecutive time points through time series analysis to determine key ecological change indicators and generate preliminary forest health assessment data. This step relies critically on data input and trend analysis algorithms to ensure the accuracy and practicality of the assessment results.

[0136] S513: Apply preliminary forest health assessment data to optimize model parameters to match changing environmental conditions using the formula:

[0137]

[0138] Produce forest health assessment results;

[0139] Among them, F(x) represents the forest health assessment result, which means the overall health status of the forest obtained by calculation, and c i is the parameter adjustment coefficient, which is used to adjust the influence of multiple ecological change indicators in the overall assessment. i Represents multiple ecological indicators extracted from GIS data, including vegetation coverage, soil and water conservation capacity, a i is the variable weight, which adjusts the model's sensitivity to changes in indicators, and b i is the variable adjustment power, which is used to enhance or weaken the impact of the target indicator and is adjusted according to the contribution of the indicator to forest health. i is the offset factor used to balance the baseline effect.

[0140] formula:

[0141]

[0142] The benefit of the formula is that by dynamically adjusting the parameter c i 、a i 、b i ,d i To match the changing environmental conditions and improve the model's matching ability and prediction accuracy for environmental changes.

[0143] Detailed explanation of the formula and the process of formula calculation and derivation:

[0144] The forest health assessment involves four key ecological variables, and the measured value of each variable is x i Obtained from field sampling data, for example, x1 = 0.5, x2 = 0.3, x3 = 0.7, x4 = 0.4, parameter c i 、a i 、b i ,d i are set to values ​​based on past research experience and field data analysis,

[0145] c=[1.2, 0.8, 1.5, 1.1], a=[1, 1, 2, 2], b=[2, 2, 1, 1], d=

[0146] [1, 1, 1, 1]:

[0147]

[0148] F(x)≈0.48+0.22+0.43+0.126=1.256

[0149] The results show that the calculated forest health index is 1.256. The higher the value, the better the health of the forest. The results can be used directly to guide the formulation of forest management and protection policies.

[0150] 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 smart forest and grassland remote sensing monitoring method based on GIS and big data, characterized in that: The following steps are involved: Based on remote sensing data, the coordinates of forest-covered areas are obtained, and the data are overlaid and analyzed with GIS. The boundaries and internal structural features of the forest area are identified, the number of connected components and holes is calculated, and primary topological features are generated. Taking the primary topological features as input, setting the scale parameters of the analysis, dynamically adjusting the clustering granularity, covering the analysis with multiple time scales including days, months and years, performing structural clustering, and generating a clustering topology map; According to the cluster topology, the parameters of the recursive neural network model are set, the characteristics of each cluster are deeply learned, the recursive neural network model is adjusted to match the periodic changes and seasonal adjustments, and the time series analysis results are generated; Using the time series analysis results, multi-scale time windows are set, vegetation coverage is defined as an ecological indicator, key ecological changes are observed, changes in ecological parameters within each time window are subdivided, key ecological change time nodes are identified, including vegetation growth peaks and degradation rates, and ecological change indicators are generated; Combined with the ecological change indicators, GIS data was used for spatial analysis to assess forest health, adjust and optimize model parameters to match differentiated environmental conditions, and generate forest health assessment results.

2. The smart forest and grassland remote sensing monitoring method based on GIS and big data according to claim 1 is characterized in that: The primary topological features specifically include connected components and the number of holes; the clustering topology diagram specifically includes structured clustering results and time scale clustering; the time series analysis results include periodic changes and seasonal adjustments; the ecological change indicators specifically include vegetation growth peak and degradation rate; the forest health assessment results include vegetation coverage, changes in ecological parameters, and environmental adaptability.

3. The smart forest and grassland remote sensing monitoring method based on GIS and big data according to claim 2 is characterized in that: The steps of obtaining the primary topological features are specifically as follows: Obtain the coordinates of forest-covered areas from remote sensing data, spatially overlay the coordinate data, analyze the data through a geographic information system, generate a geographic information model of the forest area, and obtain preliminary features of the forest boundary and internal structure; Using the preliminary features of the forest boundary and internal structure, a spectral analysis is performed on the forest boundary, an edge detection technique is used to determine the location of the forest boundary, and a texture analysis is applied to the internal structure to identify differentiated tree species and dense areas, and the number of connected components and holes is calculated; Using the number of connected components and holes, a weighted formula is used through structural analysis and the correlation evaluation of the number of nodes: Calculate and generate primary topological features; Among them, w i Represents the relative critical weight of each connected component, which is used to weight the contribution of each node, x i represents the number of nodes within a component, reflecting the size of the component, and n is the total number of connected components included in the analysis, which is used for normalization.

4. The smart forest and grassland remote sensing monitoring method based on GIS and big data according to claim 3 is characterized in that: The steps of obtaining the cluster topology map are specifically as follows: Based on the primary topological features, analysis scale parameters are set, daily, monthly and annual data sets are integrated, and time data aggregation is performed to obtain multi-scale time aggregation data; Using the multi-scale time aggregation data, dynamically adjusting the granularity of the clustering algorithm, determining parameter optimization and clustering centers of the clustering algorithm, and obtaining an adjusted clustering parameter set; According to the adjusted clustering parameter set, structural clustering is performed using the formula: Calculate and obtain the topological structure score of each cluster and generate a cluster topology graph; Where C(i) represents the topological structure score of the i-th cluster, d ij is the distance between the i-th cluster and the j-th data point, α j is the adjustment coefficient, reflecting the criticality of distance in each cluster, and m is the total number of data points in the cluster.

5. The method for intelligent forest and grassland remote sensing monitoring based on GIS and big data according to claim 4 is characterized in that: The steps for obtaining the time series analysis results are specifically as follows: Receiving the cluster topology map, adjusting the parameters of the recursive neural network model, initializing the network weights and biases according to the plurality of clustering features, configuring the network to match the characteristics of the input data, and generating a preliminary network configuration; Based on the preliminary network configuration, feature learning is performed for multiple clusters, neural network parameters, including weights and biases, are adjusted according to periodic changes to match seasonal changes, and the adjusted network parameters are output; Using the adjusted network parameters, sequence analysis was performed using the formula: Generate time series analysis results; Among them, S(t) is the output sequence at time t, which represents the time-dependent result of the model output, β k represents the period adjustment coefficient, which is used to adjust the model's response sensitivity to changes in the differentiated time period, γ k Represents the growth rate, which is used to adjust the impact of time growth on output, σ k is the normalization factor used for data standardization, and t represents the time variable.

6. The method for intelligent forest and grassland remote sensing monitoring based on GIS and big data according to claim 5 is characterized in that: The steps for obtaining the ecological change indicators are specifically as follows: Using the time series analysis results, set up multi-scale time windows, including daily, monthly, and annual data views, index the changes in ecological parameters at multiple time levels, and generate time window configurations; Based on the time window configuration, the vegetation coverage rate is defined as an ecological indicator, and the growth and decline trends of vegetation are recorded by comparing the data in multiple time windows to generate ecological change observation data; Based on the ecological change observation data, key ecological change time nodes are identified, including vegetation growth peak and degradation rate, using the formula: Generate indicators of ecological change; Among them, ΔV(t) represents the change in vegetation coverage at time t, which is used to measure ecological changes in the target time period. Δt represents the time interval for the change, which is the time basis for calculating the rate of change. λ is the baseline adjustment coefficient, which is used to adjust the basic value of the model's sensitivity to differentiated ecological data. δ is the time adjustment coefficient, which adjusts the model's response speed to time changes. e is the base of the natural logarithm.

7. The method for intelligent forest and grassland remote sensing monitoring based on GIS and big data according to claim 6 is characterized in that: The specific steps for obtaining the forest health assessment results are as follows: Use GIS data for spatial analysis, integrate the ecological change indicators into the geographic information system, set a multi-scale analysis benchmark for forest ecological change, and generate a spatial analysis configuration; Based on the spatial analysis configuration, analyze the changes in forest vegetation cover, evaluate vegetation growth and decline through GIS, and identify key ecological change indicators to generate preliminary forest health assessment data; Using the preliminary forest health assessment data, the model parameters were optimized to match the changing environmental conditions using the formula: Produce forest health assessment results; Among them, F(x) represents the forest health assessment result, which means the overall health status of the forest obtained by calculation, and c i is the parameter adjustment coefficient, which is used to adjust the influence of multiple ecological change indicators in the overall assessment. i Represents multiple ecological indicators extracted from GIS data, including vegetation coverage, soil and water conservation capacity, a i is the variable weight, which adjusts the model's sensitivity to changes in indicators, and b i is the variable adjustment power, which is used to enhance or weaken the impact of the target indicator and is adjusted according to the contribution of the indicator to forest health. i is the offset factor used to balance the baseline effect.

Images