A land space planning intelligent optimization method and system

Abstract

The application discloses a kind of land space planning intelligent optimization method and system, it is related to land space planning technical field, the method is realized by the construction of linkage pressure index ORI, realizes the multidimensional coupling calculation between red line boundary ecological elasticity and internal degradation potential, comprehensively judges the linkage pressure intensity that space ecological system bears, and constructs dynamic response strategy in combination with linkage interval threshold value.When the linkage pressure index ORI value is higher than the second linkage interval threshold F2, automatic trigger repair response;It is between linkage interval threshold value, and boundary development is inhibited;Below the first linkage interval threshold F1, maintain the original circle layer unmoved.The mechanism not only improves the response efficiency and accuracy of planning management decision, but also provides intelligent adjustment basis with operability for land space governance, ensures that the whole chain closed-loop ecological regulation system effectively operates from data acquisition, model calculation to strategy execution.

Description

Technical Field

[0001] This invention relates to the field of land and space planning technology, specifically to an intelligent optimization method and system for land and space planning. Background Technology

[0002] Territorial spatial planning, as a core component of the natural resource management system, is a comprehensive field that intersects and integrates spatial governance and ecological protection. This field encompasses important directions such as land use zoning, ecological protection red lines, and coordinated urban-rural development. With the development of digital technologies such as satellite remote sensing, Geographic Information Systems (GIS), and big data analytics, intelligent dynamic optimization of territorial space has gradually become a key research direction. Particularly in the specific application scenario of ecological protection red line management, how to achieve dynamic identification and scientific optimization of red line boundaries through intelligent sensing and data-driven approaches has become the focus of technological innovation.

[0003] Currently, in the existing practice of ecological red line management, the red line boundaries are usually delineated in a "single, rigid, and static" manner. This means that highly sensitive ecological areas are sealed off in a static map format, lacking a mechanism to perceive dynamic changes in ecological disturbances. This static boundary delineation method has the following significant drawbacks: it cannot respond in real time to ecosystem degradation changes, ignores the ecological volatility of the buffer zone near the red line boundary, and cannot provide data-driven dynamic grading basis for subsequent ecological restoration and development control. As a result, the problems of overly conservative or insufficient protection of the boundaries coexist, hindering the precise implementation of the principle of "ecological priority and moderate development."

[0004] The root of these problems lies in the fact that traditional methods of delineating red lines fail to incorporate an "ecological critical response model," neglecting the resilience and gradual degradation processes inherent in ecosystems. Once ecological disturbances occur in the boundary areas, such as increased heat island effects, vegetation fragmentation, or topographical disturbances, existing mechanisms are unable to identify and respond promptly, leading to a widening ecological gap between the areas inside and outside the red line. Furthermore, this imbalance can trigger a phenomenon known as "border protection with internal loss," where over-protection of the boundary leads to undetected internal degradation, while simultaneously hindering the initiation of appropriate ecological restoration projects outside the boundary. Ultimately, this affects the overall ecological stability and scientific management of the national land space system. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an intelligent optimization method and system for land spatial planning, which solves the problems mentioned in the background section.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a method for intelligent optimization of land spatial planning, comprising the following steps:

[0007] S1. By collecting remote sensing images of the ecological red line boundary in five phases, multi-temporal remote sensing image data extraction was performed to obtain differential data groups;

[0008] S2. Based on the differential data set, perform feature extraction to obtain differential feature vectors, and then perform preprocessing based on the differential feature vectors to obtain a standardized differential feature vector set;

[0009] S3. Based on the standardized differential feature vector group, calculate the output boundary resilience identification index (EBI), and conduct a preliminary comparative assessment based on the output results of the boundary resilience identification index (EBI), and trigger ecological degradation analysis based on the preliminary comparative assessment results.

[0010] S4. Based on the five periods of remote sensing image data, extract standardized degradation vector sets, and then calculate and output the degradation potential index (DRI) through the standardized degradation vector sets.

[0011] S5. Based on the boundary elasticity identification index (EBI) and degradation potential index (DRI), a comprehensive calculation is performed to output the linkage pressure index (ORI). The linkage interval threshold is set and the linkage pressure index (ORI) is compared and evaluated for a second time, and a response strategy is executed.

[0012] Preferably, S1 includes S11 and S12;

[0013] S11. Access the remote sensing satellite platform through the API application interface, collect five phases of remote sensing image data of the ecological red line boundary, the five phases of remote sensing image data are on an annual scale; and perform image preprocessing on the five phases of remote sensing image data to obtain a standardized remote sensing image set, the image preprocessing includes geometric correction and registration, atmospheric correction, radiometric correction and regional cropping.

[0014] The standardized remote sensing image set includes the first phase of standard remote sensing images, the second phase of standard remote sensing images, the third phase of standard remote sensing images, the fourth phase of standard remote sensing images, and the fifth phase of standard remote sensing images;

[0015] The geometric correction and registration are performed automatically on the five phases of remote sensing image data using the geospatial analysis platform GEE, eliminating image position errors caused by angle, orbit offset and terrain undulation in the five phases of remote sensing image data;

[0016] The atmospheric correction is assisted by using a medium resolution imaging spectrometer (MODIS) to remove atmospheric interference with the remote sensing signal.

[0017] The radiometric correction automatically converts the digital value DN into reflectance for physical applications using the SNAP remote sensing data processing software platform;

[0018] The region clipping was performed using the QGIS raster clipping tool to clip the area within ±500 of the ecological redline boundary in the five phases of remote sensing image data to obtain a buffer zone.

[0019] S12. Based on the standardized remote sensing image set obtained after image preprocessing, perform multi-temporal remote sensing image data extraction for each standardized remote sensing image set, and summarize to obtain the difference data group.

[0020] The differential data set includes the vegetation index NDVI from the most recent period t. t Soil Adjusted Vegetation Index (SAVI) in period t t Albedo, the surface albedo index for period t t LST (Land Surface Temperature Difference Index) in period t t and the topographic disturbance index DEM in period t t ;

[0021] The vegetation index NDVI in period t t By using multispectral satellite imagery, the near-infrared (NIR) and red (RED) bands of each remote sensing image in the standardized remote sensing image set are extracted. Then, the differences between the near-infrared (NIR) and red (RED) bands are analyzed using the raster clipping tool QGIS to obtain the vegetation index NDVI of the buffer zone in each standardized remote sensing image set.

[0022] The soil-adjusted vegetation index (SAVI) in period t. t Soil brightness correction factor L is extracted by inserting based on the difference between the near-infrared band (NIR) and the red band (RED). The soil brightness correction factor L is a dimensionless value of 0.5 and is used for areas with low vegetation cover or bare soil.

[0023] The surface albedo index for period t t The reflectance of blue, green, red, near-infrared (NIR), shortwave infrared 1 (SIR1), and shortwave infrared 2 (SIR2) bands is extracted from multispectral satellite imagery with a resolution grid of 10m. The albedo formula Landsat is then used to calculate and output the reflectance based on the blue, green, red, near-infrared (NIR), shortwave infrared 1 (SIR1), and shortwave infrared 2 (SIR2) bands.

[0024] The surface temperature difference index LST in period t t By using thermal infrared satellite imagery, the thermal infrared bands of each period of remote sensing imagery in a standardized remote sensing image set are extracted, and then the surface temperature (LST) is obtained by inversion using the built-in functions of the remote sensing image processing platform ENVI.

[0025] The topographic disturbance index DEM in period t t The data download platform is used to obtain the elevation data of each period of remote sensing images from the standardized remote sensing image set. The GIS analysis platform is then used to perform raster alignment to ensure that the two elevation data are aligned. The elevation data is the terrain disturbance index (DEM).

[0026] Preferably, S2 includes S21 and S22;

[0027] S21. Based on the acquired difference data set, feature extraction is performed to obtain the difference feature vector. The feature extraction is performed by extracting a single data point from each period in the difference data set and performing interpolation calculation.

[0028] The difference feature vectors include the vegetation index difference △NDVI(t, t-1) between the remote sensing images of period t and t-1, the soil-adjusted vegetation index difference △SAVI(t, t-1) between the remote sensing images of period t and t-1, the surface albedo index difference △Albedo(t, t-1) between the remote sensing images of period t and t-1, the surface temperature difference index difference △LST(t, t-1) between the remote sensing images of period t and t-1, and the topographic disturbance index difference △DEM(t, t-1) between the remote sensing images of period t and t-1.

[0029] S22. Preprocess the differential feature vectors by using z-score standardization to eliminate the influence of the dimensions of all parameters in the differential feature vectors and obtain a standardized differential feature vector set.

[0030] Preferably, S3 includes S31;

[0031] S31. Based on the standardized differential feature vector set, perform fusion calculation to output the boundary resilience identification index (EBI) and quantify the ecological sensitivity of each location at the red line boundary.

[0032] The boundary resilience identification index (EBI) is calculated and output using the following algorithm formula;

[0033] ;

[0034] In the formula, EBI(t, t-1) represents the boundary elasticity identification index of the remote sensing images in period t and period t-1, and ln represents the natural logarithm function.

[0035] Preferably, S3 further includes S32;

[0036] S32. Based on the boundary resilience identification index (EBI(t, t-1) output results of the remote sensing images of period t and period t-1, a preliminary comparative assessment is conducted, and boundary layers are delineated based on the preliminary comparative assessment results to determine the ecological sensitivity at various points along the red line boundary. The specific assessment content is as follows:

[0037] When the boundary elasticity identification index EBI(t, t-1) of the remote sensing images in period t and t-1 is greater than 0.8, it is classified as an EC1 boundary layer;

[0038] When the boundary elasticity identification index EBI(t, t-1) ∈ [0.4, 0.8] of the remote sensing images of period t and period t-1, it is divided into EC2 boundary layers;

[0039] When the boundary elasticity identification index EBI(t, t-1) of the remote sensing images in period t and period t-1 is less than 0.4, it is classified as an EC3 boundary layer.

[0040] Preferably, S4 includes S41;

[0041] S41. When the preliminary comparative assessment classifies the boundary layer as EC1, ecological degradation analysis is automatically triggered. The ecological degradation analysis is carried out by combining differential data groups with remote sensing image data from 5 periods. Ecological degradation data features are extracted to obtain an ecological degradation feature dataset. Then, the ecological degradation feature dataset is standardized using z-score to obtain a standardized degradation vector group.

[0042] The standardized degradation vector set includes the ecological change frequency Fv(T) in the T-th time window, the night light change index Hg(T) in the T-th time window, the thermal anomaly area Ha(T) in the T-th time window, the hydrological fluctuation raster value Wv(T) in the T-th time window, and the vegetation fragmentation Br(T) in the T-th time window.

[0043] Wherein, the T time window represents the changes within a certain time interval, taking 5 periods of remote sensing image data, that is, within 5 years;

[0044] The frequency of ecological changes Fv(T) within the T-th time window is obtained by extracting the vegetation index NDVI over five years. The user sets a degradation threshold. If the vegetation index NDVI is less than the degradation threshold, it indicates degradation; otherwise, it indicates no degradation. The number of times the vegetation index NDVI is less than the degradation threshold is counted.

[0045] The night light change index Hg(T) within the T-th time window is accessed through the API application interface of the remote sensing satellite platform to collect five phases of night light remote sensing images of the ecological red line boundary. After image preprocessing, the standard deviation of brightness of each pixel in each phase of the night light remote sensing image is calculated.

[0046] The thermal anomaly region Ha(T) within the T-th time window is obtained by extracting the surface temperature index LST over five years, with the user setting a thermal anomaly threshold, extracting the surface temperature index LST exceeding the thermal anomaly threshold, and calculating the average value.

[0047] The hydrological fluctuation raster value Wv(T) within the Tth time window is obtained by using the Normalized Water Index (NDWI) database to monitor water bodies in five periods of remote sensing image data, counting the number of pixels involved in water bodies, and then dividing by the ratio.

[0048] The vegetation fragmentation Br(T) within the T-th time window was analyzed using the landscape fragmentation analysis tool FRAGSTATS, based on the vegetation fragmentation Br in five periods of remote sensing image data.

[0049] Preferably, S4 further includes S42;

[0050] S42. Calculate based on standardized degradation vector set, output degradation potential index DRI, and quantify the degree of degradation of ecological red line boundary;

[0051] The Degradation Potential Index (DRI) is calculated and output using the following algorithm formula;

[0052] ;

[0053] In the formula, ln represents the natural logarithm function, and DRI(T) represents the degradation potential index DRI within the T-th time window.

[0054] Preferably, S5 includes S51;

[0055] S51. Based on the current Boundary Resilience Identification Index (EBI) and Degradation Potential Index (DRI), a comprehensive calculation is performed to output the Linkage Pressure Index (ORI), which comprehensively measures the spatial ecological linkage pressure borne by the redline boundary as a whole.

[0056] The linkage pressure index ORI is calculated and output using the following algorithm formula;

[0057] ;

[0058] In the formula, a1 and a2 represent the preset weight values ​​of the Boundary Resilience Identification Index (EBI) and the Degradation Potential Index (DRI), respectively. Their specific values ​​are set by the user, and a1 + a2 = 1.

[0059] Preferably, S5 further includes S52;

[0060] S52. Using the ORI of the linkage pressure index of multiple typical buffer zones, identify the 25th percentile and the 75th percentile, and set the linkage interval threshold. The linkage interval threshold includes the first linkage interval threshold F1 and the second linkage interval threshold F2.

[0061] The real-time acquired ORI (Organizational Pressure Index) is then compared and evaluated with the threshold of the linkage interval to determine the strength and internal degradation of the buffer zone. The specific evaluation content is as follows:

[0062] When the linkage pressure index ORI is greater than the second linkage interval threshold F2, it indicates that the buffer zone is critically unbalanced, and at this time, artificial ecological restoration projects are initiated.

[0063] When the first linkage interval threshold F1 ≤ linkage pressure index ORI ≤ the second linkage interval threshold F2, it indicates that there is pressure in the buffer zone. At this time, the boundary development of the current buffer zone is reduced by 50%.

[0064] When the linkage pressure index ORI is less than the threshold F1 of the first linkage interval, it indicates that the linkage of the buffer zone is stable, and the existing boundary layer is maintained.

[0065] A smart optimization system for land spatial planning includes a remote sensing difference extraction module, a feature extraction module, a boundary elasticity identification module, a degradation potential analysis module, and a linkage pressure analysis module;

[0066] The remote sensing difference extraction module extracts difference data by collecting remote sensing images of the ecological red line boundary in five phases and performing multi-temporal remote sensing image data extraction.

[0067] The feature extraction module extracts features based on the difference data set to obtain difference feature vectors, and then preprocesses the difference feature vectors to obtain a standardized difference feature vector set.

[0068] The boundary resilience identification module calculates and outputs the boundary resilience identification index (EBI) based on a standardized differential feature vector set, performs a preliminary comparative assessment based on the output of the boundary resilience identification index (EBI), and triggers ecological degradation analysis based on the preliminary comparative assessment results.

[0069] The degradation potential analysis module extracts a standardized degradation vector set based on five periods of remote sensing image data, and then calculates and outputs the degradation potential index (DRI) using the standardized degradation vector set.

[0070] The linkage pressure analysis module performs a comprehensive calculation based on the boundary elasticity identification index (EBI) and the degradation potential index (DRI) to output the linkage pressure index (ORI). It also sets a linkage interval threshold and performs a secondary comparison and evaluation with the linkage pressure index (ORI) before executing a response strategy.

[0071] This invention provides an intelligent optimization method and system for land spatial planning. It has the following beneficial effects:

[0072] (1) This method, by constructing the Boundary Resilience Identification Index (EBI), enables time-series remote sensing monitoring and index extraction of the ±500-meter buffer zone of the ecological red line, accurately quantifying the ecological disturbance and regulation capacity of the boundary area. This index system, combined with five types of ecologically sensitive factors in the differential data set, enables the tracking of ecological state changes in different years and has the ability to dynamically identify ecological critical zones. Compared with the traditional single rigid boundary delineation method, it can refine the division into three types of boundary response levels: rigid boundary EC1, flexible boundary EC2, and elastic boundary EC3, constructing an ecological management framework of "soft boundary and hard core," significantly improving the adaptability and scientific nature of the red line boundary.

[0073] 2) This method automatically triggers the ecological degradation analysis module based on the preliminary assessment results of the Boundary Resilience Index (EBI), constructing a Degradation Potential Index (DRI) and a standardized degradation vector set of five degradation factors to comprehensively reflect the internal degradation trend and external pressure disturbance intensity of the ecosystem. Through this mechanism, potential degraded patches within the red line can be dynamically identified and monitored, effectively filling the technical gap in existing planning systems that lack dynamic quantitative analysis of ecosystem stability and evolution trends, and significantly improving the management efficiency of "early identification and early intervention" in degraded areas.

[0074] (3) This method, through the construction of the linkage pressure index ORI, realizes the multi-dimensional coupling calculation between the ecological resilience of the red line boundary and the internal degradation potential, comprehensively judges the linkage pressure intensity borne by the spatial ecosystem, and constructs a dynamic response strategy by setting the first linkage interval threshold F1 and the second linkage interval threshold F2 in combination with the quantile value. When the linkage pressure index ORI value is higher than the second linkage interval threshold F2, a repair response is automatically triggered; when it is between the linkage interval thresholds, boundary development is suppressed; when it is lower than the first linkage interval threshold F1, the original sphere remains unchanged. This mechanism not only improves the response efficiency and accuracy of planning and management decisions, but also provides an operable intelligent adjustment basis for land space governance, ensuring the effective operation of the closed-loop ecological regulation system from data collection, model calculation to strategy execution. Attached Figure Description

[0075] Figure 1 This is a schematic diagram illustrating the steps of an intelligent optimization method for land spatial planning according to the present invention;

[0076] Figure 2 This is a schematic diagram of the process of an intelligent optimization system for land spatial planning according to the present invention;

[0077] Figure 3 Dynamic trend charts of the boundary resilience identification index (EBI), degradation potential index (DRI), and linkage stress index (ORI). Detailed Implementation

[0078] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0079] Example 1: Please refer to Figure 1 and Figure 3 This invention provides an intelligent optimization method for land spatial planning. To achieve the above objectives, this invention employs the following technical solution, comprising the following steps:

[0080] S1. By collecting remote sensing images of the ecological red line boundary in five phases, multi-temporal remote sensing image data extraction was performed to obtain differential data groups;

[0081] S2. Based on the differential data set, perform feature extraction to obtain differential feature vectors, and then perform preprocessing based on the differential feature vectors to obtain a standardized differential feature vector set;

[0082] S3. Based on the standardized differential feature vector group, calculate the output boundary resilience identification index (EBI), and conduct a preliminary comparative assessment based on the output results of the boundary resilience identification index (EBI), and trigger ecological degradation analysis based on the preliminary comparative assessment results.

[0083] S4. Based on the five periods of remote sensing image data, extract standardized degradation vector sets, and then calculate and output the degradation potential index (DRI) through the standardized degradation vector sets.

[0084] S5. Based on the boundary elasticity identification index (EBI) and degradation potential index (DRI), a comprehensive calculation is performed to output the linkage pressure index (ORI). The linkage interval threshold is set and the linkage pressure index (ORI) is compared and evaluated for a second time, and a response strategy is executed.

[0085] In this embodiment, the method transforms the ecological red line boundary from static delineation to dynamic and flexible regulation by introducing multi-temporal remote sensing image analysis, ecological disturbance identification index fusion modeling, and a zonal response strategy execution mechanism. During implementation, the method first acquires five periods of remote sensing images of the ±500-meter buffer zone of the ecological red line using remote sensing satellites, extracts and standardizes the differential data sets, and then constructs the Boundary Resilience Identification Index (EBI) to identify the boundary sensitivity and stability levels. Subsequently, triggered by the EBI, standardized degradation vector sets are further extracted to construct the Ecological Degradation Potential Index (DRI), which quantifies the ecological degradation trend within the red line. Finally, based on the joint construction of the EBI and DRI, the ORI (Organizational Pressure Index) is established to assess the intensity of system linkage within the boundary. A threshold response mechanism is then constructed based on typical quantile values ​​to automatically identify critical degradation zones and areas requiring restoration, forming a feedback-enabled, adjustable, and iterative spatial governance closed loop. This implementation method significantly enhances the ability to sensitively identify ecological red line boundaries, automatically diagnose internal degraded patches, and improve the adaptability and responsiveness of spatial governance strategies in territorial spatial planning. Compared to the traditional red line management mechanism that relies primarily on experience-based delineation and periodic revisions, this scheme achieves quantitative, intelligent, and dynamic control based on real remote sensing data. It effectively avoids problems such as delayed ecological early warnings and overly rigid or lenient red line delineation, enhancing the scientific and forward-looking nature of ecological protection. It has significant practical significance and promotional value for building a regional flexible governance system, improving ecological restoration efficiency, and preventing the spread of ecological degradation.

[0086] Example 2: Please refer to Figure 1 Specifically: S1 includes S11 and S12;

[0087] S11. Access the remote sensing satellite platform through the API application interface, collect five phases of remote sensing image data of the ecological red line boundary. The five phases of remote sensing image data are scaled on an annual basis, with one phase representing one year. Perform image preprocessing on the five phases of remote sensing image data to obtain a standardized remote sensing image set. Image preprocessing includes geometric correction and registration, atmospheric correction, radiometric correction, and regional cropping.

[0088] The standardized remote sensing image set includes the first, second, third, fourth and fifth phases of standardized remote sensing images.

[0089] Geometric correction and registration were performed by using the geospatial analysis platform GEE to automatically register the five phases of remote sensing image data, eliminating image position errors caused by angle, orbit offset and terrain undulation in the five phases of remote sensing image data, and ensuring that all images corresponded pixel to pixel in space.

[0090] Atmospheric correction is performed using the Moderate Resolution Imaging Spectroradiometer (MODIS) to remove atmospheric interference with remote sensing signals.

[0091] Radiometric correction automatically converts digital values ​​(DN) into reflectance for physical applications using the SNAP remote sensing data processing software platform.

[0092] Regional cropping was performed using the QGIS raster cropping tool to crop the ecological redline boundary ±500 area in the five phases of remote sensing image data to obtain a buffer zone.

[0093] S12. Based on the standardized remote sensing image set obtained after image preprocessing, perform multi-temporal remote sensing image data extraction for each standardized remote sensing image set, and summarize to obtain the difference data group.

[0094] The differential data group includes the vegetation index NDVI from the most recent period t. t Soil Adjusted Vegetation Index (SAVI) in period t t Albedo, the surface albedo index for period t t LST (Land Surface Temperature Difference Index) in period t t and the topographic disturbance index DEM in period t t ;

[0095] NDVI (Disease Index) at period t t By using multispectral satellite imagery, the near-infrared (NIR) and red (RED) bands of each remote sensing image in the standardized remote sensing image set are extracted. Then, the differences between the near-infrared (NIR) and red (RED) bands are analyzed using the raster cropping tool QGIS. The vegetation index NDVI of the buffer zone of each standardized remote sensing image set is obtained, reflecting vegetation vitality and coverage.

[0096] Soil Adjusted Vegetation Index (SAVI) in period t t Based on the difference between the near-infrared (NIR) band and the red (RED) band, a soil brightness correction factor L is inserted for extraction. The soil brightness correction factor L is a dimensionless value of 0.5, which is used in areas with low vegetation cover or bare soil to reduce the influence of soil background on NDVI.

[0097] Albedo, the surface albedo index in period t tThe reflectance of blue, green, red, near-infrared (NIR), shortwave infrared 1 (SIR1), and shortwave infrared 2 (SIR2) bands was extracted from multispectral satellite imagery with a resolution grid of 10m. The simplified albedo formula Landsat was then used to calculate and output the albedo based on the reflectance of the blue, green, red, near-infrared (NIR), shortwave infrared 1 (SIR1), and shortwave infrared 2 bands. The surface albedo index is the ability of the Earth's surface to reflect solar radiation back into the atmosphere. Changes in surface materials, such as changes in grassland or buildings, will significantly alter the albedo.

[0098] LST (Large Surface Temperature Difference Index) in Period t t By using thermal infrared satellite imagery, the thermal infrared bands of each period of remote sensing imagery in a standardized remote sensing image set are extracted, and then the surface temperature (LST) is obtained by inversion using the built-in functions of the remote sensing image processing platform ENVI.

[0099] Terrain Disturbance Index (DEM) in period t t The data download platform acquires the elevation data of each period of remote sensing images from the standardized remote sensing image set, and uses the GIS analysis platform to perform raster alignment to ensure that the two elevation data are aligned. The elevation data is the Topographic Disturbance Index (DEM), which is used to identify human-caused topographic disturbances such as excavation, filling, and landslides.

[0100] In this embodiment, the method establishes a complete set of standardized remote sensing data acquisition and differential index extraction processes, realizing multi-phase dynamic monitoring and ecological disturbance perception within a ±500-meter buffer zone of the ecological red line boundary. In stage S11, the method automatically accesses the remote sensing satellite platform via API, acquiring five phases of remote sensing images annually. Utilizing tools such as the geospatial analysis platform GEE, the medium-resolution imaging spectrometer MODIS, the remote sensing data processing software platform SNAP, and the raster clipping tool QGIS, multi-dimensional image preprocessing, including geometric correction, atmospheric correction, radiometric correction, and regional clipping, is performed, constructing a standardized remote sensing image set with uniform spatial accuracy and strong spectral comparability. In the subsequent step S12, based on the preprocessed images, five core ecological disturbance indicators from the differential data set are extracted. Combined with multi-source data such as red light, near-infrared, shortwave infrared, thermal infrared, and elevation data, the temporal evolution of vegetation vitality, soil background, surface energy reflection, heat flux, and topographic disturbance is comprehensively captured. This implementation effectively achieves refined, multi-dimensional, and comparable analysis of the ecological state of the red line boundary, significantly improving the scientific rigor and sensitivity of ecological monitoring. Its beneficial effects are reflected in three aspects: First, it provides a high-quality differential characteristic data foundation for subsequent model inputs such as the Boundary Resilience Identification Index (EBI) and the Degradation Potential Index (DRI), enhancing the accuracy and interpretability of model outputs; second, it opens up an automated pathway for the entire process of remote sensing data acquisition, preprocessing, and index extraction, reducing human intervention errors and improving operational efficiency and repeatability; third, it lays the foundation for dynamic monitoring capabilities based on ecological remote sensing, providing quantitative support for subsequent sphere adjustment, risk warning, and ecological governance, and promoting the scientific transformation of territorial spatial governance from experience-based to data-driven, and from static red lines to flexible boundaries.

[0101] Example 3: Please refer to Figure 1 Specifically: S2 includes S21 and S22;

[0102] S21. Based on the acquired differential data set, perform feature extraction to obtain differential feature vectors. Feature extraction is performed by extracting a single data point from each period in the differential data set and performing interpolation calculations.

[0103] The difference feature vectors include the difference in vegetation index between the remote sensing images of period t and period t-1, △NDVI(t, t-1), the difference in soil-adjusted vegetation index between the remote sensing images of period t and period t-1, △SAVI(t, t-1), the difference in surface albedo index between the remote sensing images of period t and period t-1, △Albedo(t, t-1), the difference in surface temperature difference index between the remote sensing images of period t and period t-1, △LST(t, t-1), and the difference in topographic disturbance index between the remote sensing images of period t and period t-1, △DEM(t, t-1).

[0104] S22. Preprocess the differential feature vectors by using z-score standardization to eliminate the influence of the dimensions of all parameters in the differential feature vectors and obtain a standardized differential feature vector set.

[0105] In this embodiment, the method constructs a feature vector extraction and standardization mechanism based on multi-temporal remote sensing difference data. In step S21, the method extracts difference feature vectors for each pair of adjacent remote sensing image data. These difference features fully reflect the ecological change trend and disturbance response of the target area over consecutive years, ensuring the dynamic capture and horizontal comparative analysis capability of ecological evolution. In step S22, to address the issues of inconsistent units and numerical scale differences among the indicators in the difference feature vectors, the method introduces z-score standardization to unify the dimensional characteristics of each parameter, forming a standardized difference feature vector set with strong comparability and normalization. The goal of this implementation is to provide an accurate, normalized, and temporally consistent input foundation for the subsequent modeling process of complex ecological indicators such as the Boundary Resilience Identification Index (EBI), thereby significantly improving the stability of model calculations and the interpretability of output results. Its beneficial effects are reflected in the following aspects: First, it effectively improves the quality of multi-source fusion of remote sensing ecological indicators, enabling remote sensing data from multiple periods to participate in ecological assessment and analysis at a unified scale; second, the construction of difference vectors strengthens the focus on "variability" itself, making the model more sensitive and timely to ecological disturbances; third, the standardization process eliminates redundant scale interference, providing more operable and scientifically based input conditions for subsequent cluster analysis, index construction, and ecological zoning, thus laying a solid foundation for building a dynamic, data-driven redline elastic boundary model.

[0106] Example 4: Please refer to Figure 1 and Figure 3 Specifically: S3 includes S31;

[0107] S31. Based on the standardized differential feature vector set, perform fusion calculation to output the boundary resilience identification index (EBI) and quantify the ecological sensitivity of each location at the red line boundary.

[0108] The Boundary Resilience Identification Index (EBI) is calculated and output using the following algorithm formula;

[0109] ;

[0110] In the formula, EBI(t, t-1) represents the boundary elasticity identification index of the remote sensing images in period t and period t-1, and ln represents the natural logarithm function;

[0111] The calculation logic of the formula: It represents the intensity and growth trend of ecological disturbance. The product represents the coordinated change among the three indicators. The power exponent of 1.1 introduces non-linear weighting and amplifies large-scale disturbances.

[0112] The topographic and soil response moderating factors represent the system's ability to regulate disturbances. A high △DEM indicates drastic topographic changes that are difficult to recover; a high △SAVI reflects large fluctuations in bare soil areas. The natural logarithm function is used to control divergence under extreme disturbances.

[0113] The physical meaning of the formula, the Boundary Resilience Identification Index (EBI), describes the ratio of the intensity of ecological disturbance experienced by an ecological boundary unit within a certain time series to its natural buffering capacity. It is a physically meaningful ecological stress response ratio that reflects: whether the boundary area has exceeded the self-regulation threshold of the ecosystem; whether it has the ability to recover naturally from degradation trends; and whether human intervention, such as restoration or buffer zone establishment, is needed to maintain the stability of the red line boundary.

[0114] S3 also includes S32;

[0115] S32. Based on the boundary resilience identification index (EBI(t, t-1) output results of the remote sensing images of period t and period t-1, a preliminary comparative assessment is conducted, and boundary layers are delineated based on the preliminary comparative assessment results to determine the ecological sensitivity at various points along the red line boundary. The specific assessment content is as follows:

[0116] When the boundary elasticity identification index EBI(t, t-1) of the remote sensing images of period t and period t-1 is greater than 0.8, it is classified as EC1 boundary layer, that is, rigid boundary, which indicates an extremely fragile ecological critical zone that should be completely closed off and prohibited from activity.

[0117] When the boundary elasticity identification index EBI(t, t-1) ∈ [0.4, 0.8] of the remote sensing images of period t and period t-1, it is divided into EC2 boundary layers, that is, flexible boundaries, which indicate that they can be moderately adjusted, such as ecological corridors and restoration projects;

[0118] When the boundary elasticity identification index EBI(t, t-1) of the remote sensing images in period t and period t-1 is less than 0.4, it is classified as an EC3 boundary layer, i.e. an elastic boundary, indicating that the state is stable and can be considered for eco-friendly utilization.

[0119] In this embodiment, the method integrates and calculates the Boundary Resilience Identification Index (EBI) based on the standardized differential feature vector set obtained in the previous stage. The intensity of ecological disturbance is reflected by the product of the vegetation index difference (ΔNDVI), the surface temperature difference index difference (ΔLST), and the surface albedo index difference (ΔAlbedo). Combining the topographic disturbance index difference (ΔDEM) and the soil-adjusted vegetation index difference (ΔSAVI), and modulating with the natural logarithm function ln, a nonlinear sensitization coefficient of 1.1 is introduced to quantify the stress response and self-regulation capacity of the ecosystem within the boundary area. This index, as a ratiomatic indicator of ecological response, effectively expresses the carrying capacity and recovery potential of the ecosystem at the spatial boundary level after being stressed. Subsequently, in S32, based on the output of the Boundary Resilience Identification Index (EBI), a three-layer concentric model of rigid boundary EC1, flexible boundary EC2, and elastic boundary EC3 is constructed, realizing the dynamic sensitivity classification of the ecological red line boundary. This classification not only has the characteristics of clear thresholds and controllable quantification, but also provides an intuitive basis for subsequent strategic responses. The goal of this step is to accurately identify ecologically critical and adjustable areas within the red line boundaries and dynamically classify different ecological strata. The main benefits include: first, significantly improving the resolution and response accuracy of red line boundary identification, enabling spatial planning to shift from static boundaries to dynamic, flexible boundaries; second, enhancing the adaptability of ecological management, ensuring that governance measures match boundary sensitivity, and avoiding one-size-fits-all control; and third, realizing a closed-loop response mechanism from data-driven to strategy-linked approaches through index modeling, thereby promoting the national land space ecological protection system towards zoned governance, quantitative early warning, and flexible regulation.

[0120] Example 5: Please refer to Figure 1 and Figure 3 Specifically: S4 includes S41;

[0121] S41. When the preliminary comparative assessment classifies the boundary layer as EC1, ecological degradation analysis is automatically triggered. Ecological degradation analysis is carried out by combining differential data groups with remote sensing image data from 5 periods. Ecological degradation data features are extracted to obtain an ecological degradation feature dataset. Then, the ecological degradation feature dataset is standardized using z-score to obtain a standardized degradation vector group.

[0122] The standardized degradation vector set includes the frequency of ecological change Fv(T) in the T-th time window, the night light change index Hg(T) in the T-th time window, the thermal anomaly area Ha(T) in the T-th time window, the hydrological fluctuation raster value Wv(T) in the T-th time window, and the vegetation fragmentation Br(T) in the T-th time window.

[0123] Wherein, the T time window represents the changes within a certain time interval, taking 5 periods of remote sensing image data, that is, within 5 years;

[0124] The frequency of ecological changes Fv(T) within the T-th time window is obtained by extracting the vegetation index NDVI over five years. The user sets a degradation threshold. If the vegetation index NDVI is less than the degradation threshold, it indicates degradation; otherwise, it indicates no degradation. The number of times the vegetation index NDVI is less than the degradation threshold is counted.

[0125] The nighttime light change index Hg(T) within the T-th time window is accessed through the API application interface of the remote sensing satellite platform. Five periods of nighttime light remote sensing images of the ecological red line boundary are collected, and the radiometric brightness of the nighttime light remote sensing images is normalized to eliminate cross-year sensor bias. The brightness change amplitude of each pixel between period t2 and period t1 is extracted, and the standard deviation method is used to measure the degree of brightness change. This is used to characterize the trend of human activity intensity in the region over time, reflecting the temporal change intensity of human activities such as urban expansion, settlement extension, and industrial intervention.

[0126] The thermal anomaly region Ha(T) within the T-th time window is obtained by extracting the surface temperature index LST over five years, with the user setting a thermal anomaly threshold, extracting the surface temperature index LST exceeding the thermal anomaly threshold, and calculating the average value.

[0127] The hydrological fluctuation raster value Wv(T) within the T-th time window is obtained by extracting the pixel water body W from each of the five remote sensing image data periods based on the Normalized Water Index (NDWI) database. Where W... i =1 indicates that the data in the i-th period of remote sensing imagery is a water body; otherwise, W i =0, after superimposing the remote sensing image data of 5 periods, count the number of times each pixel appears as a "water body", divide by the total number of periods (e.g. 5) to obtain the water body existence frequency at the pixel level, which is used to measure the frequency of water body changes and the instability of the hydrological system in a specific area. It is especially suitable for monitoring the dynamic water surface of ecologically sensitive areas such as wetlands, lakes, and floodplains.

[0128] The vegetation fragmentation Br(T) within the T-th time window was analyzed using the landscape fragmentation analysis tool FRAGSTATS, based on five periods of remote sensing image data.

[0129] S4 also includes S42;

[0130] S42. Calculate based on standardized degradation vector set, output degradation potential index DRI, and quantify the degree of degradation of ecological red line boundary;

[0131] The Degradation Potential Index (DRI) is calculated and output using the following algorithm formula;

[0132] ;

[0133] In the formula, ln represents the natural logarithm function, and DRI(T) represents the degradation potential index DRI within the T-th time window;

[0134] The calculation logic of the formula: This represents the endogenous degradation pressure of the ecosystem, and the vegetation fluctuation multiplied by the thermal anomaly disturbance. A joint analysis is performed to determine the frequency of ecosystem vitality imbalances. The denominator is... Used to analyze recovery potential, high fragmentation indicates poor connectivity and therefore poor recovery potential. Adding 1 is to avoid dividing by 0. This part of the result represents the strength of the ecosystem's own degradation trend.

[0135] The external pressure factor is represented by the nighttime light change index Hg(T) within the T-th time window, which is used to measure the development intensity. The hydrological fluctuation grid value Wv(T) within the T-th time window is used to measure the instability of the hydrological system, which leads to wetland degradation, groundwater anomalies, etc. The natural logarithm function ln is used to introduce nonlinear triggered weighting, showing that slight changes can be adjusted, while drastic changes will exponentially increase the disturbance.

[0136] The physical meaning of the formula: The Degradation Potential Index (DRI) is a product model of an ecosystem, which is "stability divided by degradation trend multiplied by external disturbance gain". When an ecosystem fluctuates frequently, experiences severe thermal disturbance, and has high vegetation fragmentation, its internal structure is already in an unbalanced state. If this is combined with increased nocturnal light and development intrusion, along with water body instability, it indicates that the ecosystem is likely to evolve towards full degradation. At this time, the DRI will increase significantly, providing a data-driven triggering mechanism for governance.

[0137] In this embodiment, after identifying the highly sensitive area of ​​the rigid boundary EC1, the method automatically triggers ecological degradation analysis, integrates five periods of remote sensing data, and extracts an ecological degradation feature dataset from multi-source time-series images. This dataset encompasses five core standardized degradation vector groups, and uses Z-score standardization to unify the dimensions, forming standardized degradation vector groups. Each parameter is a key response indicator to the dynamic imbalance within the ecosystem and external development disturbances, describing the dynamic stability and stress state of the ecosystem from different dimensions. In S42, the method uses the above-mentioned standardized degradation vector groups to calculate and output the degradation potential index DRI through a multi-factor nonlinear coupling model. This index constructs the endogenous degradation pressure and system recovery potential of the ecosystem as the basic evaluation structure, and further introduces the nighttime light change index Hg(T) and the hydrological fluctuation raster value Wv(T) as pressure disturbance terms. After modulation by the natural logarithm function ln, a nonlinear assessment model that can distinguish between mild disturbances and severe shocks is formed, effectively quantifying the potential risk of the ecosystem evolving from the initial degradation state to critical instability. The purpose of this step is to accurately identify, quantify, and predict potential degradation trends within the rigid boundaries of the ecological red line. Its beneficial effects are mainly reflected in three aspects: First, it establishes a degradation identification mechanism that highly integrates remote sensing change detection and ecological factor calculation, realizing the transformation analysis from "surface changes" to "structural degradation"; second, the DRI model can output regional-level ecological degradation risk values ​​in real time, empowering managers to dynamically formulate restoration, intervention, or early warning strategies; and third, it significantly improves the data triggering accuracy and spatial resolution efficiency of degradation responses, promoting ecological red line supervision from static assessment to multi-temporal dynamic early warning, and greatly enhancing the intelligence level and forward-looking intervention capabilities of spatial governance.

[0138] Example 6: Please refer to Figure 1 and Figure 3 Specifically: S5 includes S51;

[0139] S51. Based on the current Boundary Resilience Identification Index (EBI) and Degradation Potential Index (DRI), a comprehensive calculation is performed to output the Linkage Pressure Index (ORI), which comprehensively measures the spatial ecological linkage pressure borne by the redline boundary as a whole.

[0140] The ORI (Oriented Stress Index) is calculated and output using the following algorithm formula;

[0141] ;

[0142] In the formula, a1 and a2 represent the preset weight values ​​of the Boundary Resilience Identification Index (EBI) and the Degradation Potential Index (DRI), respectively. Their specific values ​​are set by the user, and a1 + a2 = 1.

[0143] Physical meaning of the formula: The boundary resilience identification index (EBI) and the degradation potential index (DRI) are summarized by a weighted formula, and users assign corresponding weights to them. This comprehensively measures the degradation and diffusion trend under the dual effects of boundary fragility and internal degradation.

[0144] S5 also includes S52;

[0145] S52. Using the ORI of multiple typical buffer zones, identify the 25th percentile and 75th percentile values, and set the linkage interval threshold. The linkage interval threshold includes the first linkage interval threshold F1 and the second linkage interval threshold F2.

[0146] Among them, the first linkage interval threshold F1 represents the 25th percentile value, corresponding to normal ecological degradation, and the second linkage interval threshold F2 represents the 75th percentile value, corresponding to abnormal ecological degradation.

[0147] The real-time acquired ORI (Organizational Pressure Index) is then compared and evaluated with the threshold of the linkage interval to determine the strength and internal degradation of the buffer zone, i.e., the boundary of the ecological red line. The specific evaluation content is as follows:

[0148] When the linkage pressure index ORI is greater than the second linkage interval threshold F2, it indicates that the buffer zone is critically unbalanced, and at this time, artificial ecological restoration projects are initiated.

[0149] When the first linkage interval threshold F1 ≤ linkage pressure index ORI ≤ the second linkage interval threshold F2, it indicates that there is pressure in the buffer zone. At this time, the boundary development of the current buffer zone is reduced by 50%.

[0150] When the linkage pressure index ORI is less than the threshold F1 of the first linkage interval, it indicates that the linkage of the buffer zone is stable and the ecosystem is in a recoverable state. At this time, the existing boundary layer should be maintained.

[0151] In this embodiment, the method first weights and fuses the Boundary Resilience Identification Index (EBI) and the Degradation Potential Index (DRI) to form the Linked Pressure Index (ORI), which comprehensively characterizes the combined strength of structural vulnerability and internal degradation trends experienced by red-line boundary units within the same time window. The calculation formula for this index introduces adjustable weight parameters a1 and a2, allowing users to flexibly configure the weight ratio of boundary sensitivity and degradation risk according to management needs. This makes ORI an integrated metric with strategic flexibility and indicator linkage, effectively improving the collaborative judgment capability in red-line supervision. Subsequently, in S52, by performing distribution analysis on the ORI of multiple typical buffer zones, the 25th and 75th percentile values ​​are extracted to scientifically construct the linkage interval threshold, establishing a dynamic, hierarchical, and quantitative ecological response threshold mechanism. This mechanism implements a three-tiered judgment logic for the current ecological state of the red line boundary: when the ORI (Oriented Pressure Index) is higher than the second linkage interval threshold F2, it is identified as a critical imbalance state, triggering a timely artificial restoration response; when the ORI is between the first linkage interval threshold F1 and the second linkage interval threshold F2, it is determined that a moderate pressure zone exists, automatically guiding the boundary development intensity to be halved; when the ORI is lower than the first linkage interval threshold F1, it is marked as a stable and recoverable state, requiring no intervention and maintaining the status quo. The implementation of this step achieves a closed-loop evaluation process from indicator integration to risk classification to strategy response, significantly improving the sensitivity and strategy execution capabilities of land spatial planning in addressing the multi-factor linkage pressure of the red line boundary. The beneficial effects are mainly reflected in the following aspects: First, by linking the ORI (Orientation of Increasing Pressure) index, a quantitative evaluation bridge is built across the boundary vulnerability and degradation trend; second, the quantile threshold division mechanism effectively breaks through the limitations of traditional experience and establishes a dynamic and adaptive ecological early warning system; third, the management of red line spatial boundaries has been upgraded and transformed from static control to intelligent zoning and dynamic intervention, enhancing the timeliness, accuracy and sustainability of ecological security prevention and control.

[0152] Example 7: Please refer to Figure 1 and Figure 2 A smart optimization system for land spatial planning includes a remote sensing difference extraction module, a feature extraction module, a boundary elasticity identification module, a degradation potential analysis module, and a linkage pressure analysis module.

[0153] The remote sensing difference extraction module extracts differential data by collecting five phases of remote sensing images of the ecological red line boundary, and extracts differential data groups from multi-temporal remote sensing image data.

[0154] The feature extraction module extracts features based on the differential data set to obtain differential feature vectors, and then preprocesses the differential feature vectors to obtain a standardized differential feature vector set.

[0155] The boundary resilience identification module calculates and outputs the boundary resilience identification index (EBI) based on a standardized set of differential feature vectors, performs a preliminary comparative assessment based on the output of the boundary resilience identification index (EBI), and triggers ecological degradation analysis based on the preliminary comparative assessment results.

[0156] The degradation potential analysis module extracts standardized degradation vector sets based on five periods of remote sensing image data, and then calculates and outputs the degradation potential index (DRI) using the standardized degradation vector sets.

[0157] The linkage pressure analysis module performs a comprehensive calculation based on the boundary elasticity identification index (EBI) and degradation potential index (DRI) to output the linkage pressure index (ORI). It also sets a linkage interval threshold and performs a secondary comparison and evaluation with the linkage pressure index (ORI) before executing a response strategy.

[0158] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

Claims

1. A method for intelligent optimization of territorial space planning, characterized by: The method comprises the following steps: S1, acquiring 5-phase remote sensing images of the ecological red line boundary, performing multi-temporal remote sensing image data extraction, and obtaining a difference data set; S2, performing feature extraction based on the difference data set to obtain a difference feature vector, and performing preprocessing based on the difference feature vector to obtain a standardized difference feature vector set; The difference feature vector comprises a vegetation index difference value △NDVI(t, t-1) of the t-phase and t-1-phase remote sensing images, a soil-adjusted vegetation index difference value △SAVI(t, t-1) of the t-phase and t-1-phase remote sensing images, a surface albedo index difference value △Albedo(t, t-1) of the t-phase and t-1-phase remote sensing images, a surface temperature difference index difference value △LST(t, t-1) of the t-phase and t-1-phase remote sensing images, and a terrain disturbance index difference value △DEM(t, t-1) of the t-phase and t-1-phase remote sensing images; S3, calculating an output boundary elasticity identification index EBI based on the standardized difference feature vector set, performing preliminary comparative evaluation based on the output result of the boundary elasticity identification index EBI, and triggering ecological degradation analysis based on the preliminary comparative evaluation result; The boundary elasticity identification index EBI is calculated and output by the following algorithm formula: ； In the formula, EBI(t, t-1) represents the boundary elasticity identification index of the t-phase and t-1-phase remote sensing images, and ln represents a natural logarithm function; S4, extracting a standardized degradation vector set based on 5-phase remote sensing image data, and calculating and outputting a degradation potential index DRI through the standardized degradation vector set; The standardized degradation vector set comprises an ecological change frequency Fv(T) within a T-time window, a night light change index Hg(T) within the T-time window, a thermal anomaly region Ha(T) within the T-time window, a hydrological fluctuation grid value Wv(T) within the T-time window, and a plant fragmentation degree Br(T) within the T-time window; The degradation potential index DRI is calculated and output by the following algorithm formula: ； In the formula, ln represents a natural logarithm function, and DRI(T) represents the degradation potential index DRI within the T-time window; S5, performing comprehensive calculation based on the boundary elasticity identification index EBI and the degradation potential index DRI to output a linkage pressure index ORI, performing secondary comparative evaluation of the linkage pressure index ORI by setting a linkage interval threshold value, and executing a response strategy; The linkage pressure index ORI is calculated and output by the following algorithm formula: ； In the formula, a1 and a2 respectively represent preset weight values of the boundary elasticity identification index EBI and the degradation potential index DRI, the specific values of which are set by a user, and a1+a2=1.

2. The method of claim 1, wherein: The S1 comprises S11 and S12; S11, accessing a remote sensing satellite platform through an API application program interface to collect 5-phase remote sensing image data of the ecological red line boundary, the 5-phase remote sensing image data being collected by taking years as a scale; and performing image preprocessing on the 5-phase remote sensing image data to obtain a standardized remote sensing image set, the image preprocessing comprising geometric correction and registration, atmospheric correction, radiation correction, and regional cropping; The standardized remote sensing image set includes a first period standard remote sensing image, a second period standard remote sensing image, a third period standard remote sensing image, a fourth period standard remote sensing image and a fifth period standard remote sensing image; The geometric correction and registration automatically registers the fifth period remote sensing image data by using a geographic spatial analysis platform GEE, and eliminates image position errors caused by angles, track offsets and terrain undulations in the fifth period remote sensing image data; The atmospheric correction removes the interference of the atmosphere on the remote sensing signal by using a moderate resolution imaging spectrometer MODIS for correction assistance; The radiation correction automatically converts digital values DN into physical application reflectivity by using a remote sensing data processing software platform SNAP; The regional clipping clips the ±500 region of the ecological red line boundary in the fifth period remote sensing image data by using a grid clipping tool QGIS, and obtains a buffer zone; S12, based on the standardized remote sensing image set obtained after image preprocessing, multi-temporal remote sensing image data is extracted from each period of the standardized remote sensing image set, and difference data groups are obtained by summarizing; said difference data set comprising a near t-th period vegetation index NDVI t , a t-th period soil-adjusted vegetation index SAVI t , a t-th period land surface albedo index Albedo t , a t-th period land surface temperature difference index LST t , and a t-th period terrain disturbance index DEM t ; The tth vegetation index NDVI t By using multispectral satellite images, the near-infrared band NIR and the red band RED of each remote sensing image in a standardized remote sensing image set are extracted, the difference between the near-infrared band NIR and the red band RED is analyzed by using a grid clipping tool QGIS, and the vegetation index NDVI of a buffer zone in each standardized remote sensing image set is obtained. The soil-adjusted vegetation index SAVI of the tth period t The soil brightness correction factor L is inserted for extraction based on the difference between the near-infrared band NIR and the red light band RED, and the soil brightness correction factor L has a value of 0.5 dimensionless value, which is used for low vegetation coverage or bare soil area. the tth surface albedo index Albedo t The reflectivity of blue light band, green light band, red light band, near-infrared band NIR, short-wave infrared 1 and short-wave infrared 2 is extracted through multi-spectral satellite image, and the resolution grid is 10m; and the reflectivity of blue light band, green light band, red light band, near-infrared band NIR, short-wave infrared 1 and short-wave infrared 2 is calculated and output based on the reflectivity of blue light band, green light band, red light band, near-infrared band NIR, short-wave infrared 1 and short-wave infrared 2 using the albedo formula Landsat. The tth land surface temperature difference index LST t By using thermal infrared band satellite image, extraction standardization remote sensing image set, each remote sensing image thermal infrared band, and then use the built-in function of remote sensing image processing platform ENVI, to obtain the land surface temperature LST. The tth terrain disturbance index DEM t Through the data download platform, the elevation data of each period of remote sensing image in the standardized remote sensing image set is obtained, and raster alignment is performed using a GIS analysis platform to ensure that the two elevation data are aligned, and the elevation data is the terrain disturbance index DEM.

3. The method of claim 2, wherein: The S2 includes S21 and S22; S21, based on the obtained difference data group, feature extraction is performed to obtain a difference feature vector, and the feature extraction is performed by extracting each single data in the difference data group and performing interpolation calculation and extraction; S22, the difference feature vector is preprocessed, and the preprocessing is performed by using z-score standardization to standardize the difference feature vector, eliminating the dimension influence of all parameters in the difference feature vector, and obtaining a standardized difference feature vector set.

4. The method of claim 3, wherein: The S3 includes S31; S31, based on the standardized difference feature vector set, fusion calculation is performed to output a boundary elasticity recognition index EBI, and the ecological sensitivity of each part of the red line boundary is quantified.

5. The method of claim 4, wherein: The S3 also includes S32; S32, based on the boundary elasticity recognition index EBI(t, t-1) output result of the t period and the t-1 period remote sensing image, preliminary comparative evaluation is performed, and based on the preliminary comparative evaluation result, boundary circle layer division is performed to judge the ecological sensitivity of each part of the red line boundary, and the specific evaluation content is as follows: When the boundary elasticity recognition index EBI(t, t-1) of the t period and the t-1 period remote sensing image is greater than 0.8, it is divided into EC1 boundary circle layer; When the boundary elasticity recognition index EBI(t, t-1) of the t period and the t-1 period remote sensing image is in the range of [0.4, 0.8], it is divided into EC2 boundary circle layer; When the boundary elasticity recognition index EBI(t, t-1) of the t period and the t-1 period remote sensing image is less than 0.4, it is divided into EC3 boundary circle layer.

6. The method of claim 5, wherein: The S4 includes S41; S41, when the preliminary comparative evaluation is divided into EC1 boundary circle layer, automatic triggering of ecological degradation analysis is performed, and the ecological degradation analysis is performed based on the difference data group combined with the fifth period remote sensing image data, ecological degradation data feature extraction, obtaining an ecological degradation feature data set, and then using z-score standardization on the ecological degradation feature data set, standardizing the ecological degradation feature data set to obtain a standardized degradation vector group; The ecological change frequency Fv(T) in the first T time window is obtained by extracting the vegetation index NDVI in five years, setting a degradation threshold by a user, and counting the number of times that the vegetation index NDVI is less than the degradation threshold. The noctilucent change index Hg(T) in the first T time window is obtained by accessing a remote sensing satellite platform through an API application interface, collecting five-phase noctilucent remote sensing images of the ecological red line boundary, and calculating the standard deviation of the brightness of each pixel point in each phase of the noctilucent remote sensing image after image preprocessing. The thermal anomaly area Ha(T) in the first T time window is obtained by extracting the land surface temperature index LST in five years, setting a thermal anomaly threshold by a user, extracting the land surface temperature index LST that exceeds the thermal anomaly threshold, and performing mean value calculation. The hydrological fluctuation grid value Wv(T) in the first T time window is obtained by using a normalized water index database NDWI to monitor water bodies in five-phase remote sensing image data, and then dividing the number of pixels related to the water bodies by a proportion. The plant fragmentation Br(T) in the first T time window is obtained by using a landscape fragmentation analysis tool FRAGSTATS to analyze the vegetation fragmentation Br in five-phase remote sensing image data.

7. The method of claim 6, wherein: The S4 further includes S42. S42, based on the standardized degradation vector group, calculates and outputs a degradation potential index DRI to quantify the degradation degree of the ecological red line boundary.

8. The method of claim 7, wherein: The S5 includes S51. S51, based on the current boundary elasticity recognition index EBI and the degradation potential index DRI, performs comprehensive calculation and outputs a linkage pressure index ORI to comprehensively measure the spatial ecological linkage pressure borne by the red line boundary as a whole.

9. The method of claim 8, wherein: The S5 further includes S52. S52, using the linkage pressure indexes ORI of a plurality of typical buffer zones, identifies the 25% quantile value and the 75% quantile value to set linkage interval thresholds, wherein the linkage interval thresholds include a first linkage interval threshold F1 and a second linkage interval threshold F2. The real-time obtained linkage pressure index ORI is compared with the linkage interval thresholds again to evaluate the strength of the buffer zone and the internal degradation, and the specific evaluation content is as follows: When the linkage pressure index ORI is greater than the second linkage interval threshold F2, it indicates that the buffer zone is critically unbalanced, and at this time, artificial ecological restoration engineering is started; When the first linkage interval threshold F1 is less than or equal to the linkage pressure index ORI and is less than or equal to the second linkage interval threshold F2, it indicates that the buffer zone is under pressure, and at this time, the boundary development of the current buffer zone is reduced by 50%; When the linkage pressure index ORI is less than the first linkage interval threshold F1, it indicates that the buffer zone is stable, and at this time, the existing boundary circle is maintained.

10. A land space planning intelligent optimization system, applied to the land space planning intelligent optimization method of any one of claims 1-9, characterized in that: The remote sensing difference extraction module, the feature extraction module, the boundary elasticity recognition module, the degradation potential analysis module, and the linkage pressure analysis module are included. The remote sensing difference extraction module acquires five-phase remote sensing images of the ecological red line boundary, performs multi-temporal remote sensing image data extraction, and obtains a difference data group. The feature extraction module extracts features based on the difference data set, obtains a difference feature vector, and pre-processes the difference feature vector based on the difference feature vector to obtain a standardized difference feature vector set; The boundary elasticity recognition module calculates an output boundary elasticity recognition index EBI based on the standardized difference feature vector set, performs preliminary comparative evaluation based on the output result of the boundary elasticity recognition index EBI, and triggers an ecological degradation analysis based on the preliminary comparative evaluation result; The degradation potential analysis module extracts a standardized degradation vector set based on the 5-phase remote sensing image data, and calculates an output degradation potential index DRI based on the standardized degradation vector set; The linkage pressure analysis module performs comprehensive calculation based on the boundary elasticity recognition index EBI and the degradation potential index DRI to output a linkage pressure index ORI, performs secondary comparative evaluation of the linkage pressure index ORI with a linkage interval threshold value, and executes a response strategy.

Images