An ecological remote sensing data processing system

Abstract

The application relates to the technical field of geographic information data, in particular to an ecological remote sensing data processing system, which comprises a remote sensing layer collection module, a boundary position checking module, a regional code integration module, a spatial change splicing module and an ecological trend drawing module.In the application, the coverage range and positioning accuracy of the ground feature map patch are significantly improved by accurately matching the geographic recognition information and the spatial boundary value, time and space information are effectively integrated, the continuity and accuracy of the data are enhanced, the error of the geographic information is effectively reduced by the boundary resetting technology, the time-space continuity of the data is improved, the processing efficiency is greatly improved by constructing a continuous layer structure and optimizing a data retrieval process, the fine boundary splicing technology also allows ecological changes to be tracked in detail, more detailed data support is provided for ecological monitoring, the ecological trend analysis capability is enhanced, and thus the overall effect and application value of the ecological monitoring are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of geographic information data, and particularly relates to an ecological remote sensing data processing system. BACKGROUND

[0002] The technical field of geographic information data contains the collection, storage, processing and analysis technology of spatial position information and related attributes. Based on the geographic coordinate system, the technical field integrates remote sensing image environmental monitoring data and ecological elements to build a multi-dimensional spatio-temporal database. The core is to support natural resource assessment and ecological change tracking through coordinate registration data fusion and spatial modeling. It involves satellite data interpretation, ground observation data correction, and multi-scale ecological index calculation. It serves applications such as land planning, disaster warning, and biodiversity protection. Current technology is limited by low integration efficiency of heterogeneous data, insufficient spatio-temporal continuity, and differences in data standardization.

[0003] Among them, the ecological remote sensing data processing system refers to the technical system for collaborative processing of remote sensing data and environmental parameters in ecological monitoring scenarios. It forms a unified spatio-temporal reference by establishing an automatic matching mechanism for multi-source remote sensing data and ground observation data, constructs a dynamic monitoring framework for ecological elements, and realizes fast indexing of massive data using geographic grid-based data partition management technology. Specifically, it eliminates spatial deviations caused by sensor differences through coordinate conversion rules, separates basic geographic information and ecological thematic attributes using data hierarchical storage structure, extracts spatial distribution characteristics of ecological indicators such as vegetation cover and soil erosion using spatial overlay analysis method, realizes time series comparison of multi-period data based on standardized data format, supplements ecological parameters in missing areas relying on spatial interpolation technology, and outputs ecological assessment results of different scales through thematic mapping templates.

[0004] The existing technology has low efficiency in processing large-scale heterogeneous data. The lack of efficient automatic matching mechanism and unified spatio-temporal reference leads to inaccurate spatial correspondence and inconsistent levels during data integration. Spatial deviations between remote sensing data and ground observation data can cause errors in vegetation cover and soil erosion assessment, affecting the accuracy of decision-making. Differences in data standardization and spatio-temporal discontinuity can also cause breakpoints in ecological change trend monitoring, making long-term ecological change analysis incomplete and limiting effective assessment of ecological system health, thereby affecting the formulation of ecological protection and resource management strategies. SUMMARY

[0005] The purpose of the present application is to solve the shortcomings in the prior art and to provide an ecological remote sensing data processing system.

[0006] In order to achieve the above purpose, the present application adopts the following technical scheme: an ecological remote sensing data processing system comprises:

[0007] The remote sensing layer acquisition module acquires a scanning image of a remote sensing platform, identifies covered ground objects in combination with image geographic information and boundary values, matches known marks, constructs a plot range, labels upper and lower and left and right boundary points, and generates a boundary box information set;

[0008] The boundary position checking module extracts a projection plane position based on the boundary box information set, calls a map reference boundary mark line, performs boundary resetting according to the offset direction and distance of the corner points of the plot and the reference boundary, repositions the corner positions of the plot, and generates a repositioned plot coding layer;

[0009] The regional coding integration module calls the repositioned plot coding layer, constructs a layer in time sequence, retrieves coding continuous and adjacent plots, packs as a spatial segment, and generates a combined boundary structure set;

[0010] The spatial change splicing module calls the combined boundary structure set, selects a vegetation and wetland block, compares the contour coincidence degree of adjacent blocks, judges boundary drift, records the contour number, and generates a spliced block sequence;

[0011] The ecological trend drawing module calls the spliced block sequence, retrieves a repeated extension direction and an expansion form, matches a degradation zone and a variation point, combines converging blocks into a trend segment, and obtains an ecological remote sensing data set.

[0012] As a further scheme of the present application, the boundary box information set includes upper left boundary coordinates, lower right boundary coordinates, lower left boundary coordinates, and upper right boundary coordinates of a ground object plot, the repositioned plot coding layer includes a projection boundary plane coordinate set, a reference boundary mark line space parameter, an offset direction identifier, and a distance threshold parameter, the combined boundary structure set includes a geographic coding unique identifier, a time stamp index, an adjacent plot space topological relationship, and a segment combined closed boundary line, the spliced block sequence includes a drift judgment result table, a contour offset matrix, and a junction area splicing path vector set, and the ecological remote sensing data set includes a degradation continuous zone space trajectory coordinate, a variation concentrated area polygon vertex set, a trend segment vector superposition unit, and an extension direction repetition degree identifier.

[0013] As a further scheme of the present application, the remote sensing layer acquisition module includes:

[0014] The image geographic positioning submodule acquires ground scanning image data carried by a remote sensing platform, extracts geographic identification information carried by each image, compares the corresponding positions of coordinate points, image time markers, and space boundary values in the image, calculates the spatial mapping point distribution of the image data, acquires the positioning of the image in the geographic space, and generates an image space positioning coefficient value;

[0015] The figure spot boundary recognition submodule calls the area range in the image located within the boundary value based on the image space positioning coefficient value, calculates the similarity coefficient between the pixel feature value, the space distribution value in the area and the known ground object space identifier, and judges and screens the figure spot in the area according to the similarity coefficient and the set space matching threshold value, obtains the space partition value of the boundary range coverage area, and generates the space figure spot matching degree value;

[0016] The boundary box generation submodule identifies the contour edge of the figure spot covered by the space figure spot matching degree value, judges the up, down, left and right boundary point positions of the edge line segment in the image pixel grid, calculates the coordinate value of the boundary point set, and adopts the formula:

[0017]

[0018] The operation obtains the boundary envelope trend intensity value, constructs the boundary point set according to the boundary envelope trend intensity value, obtains the closed graphic area of the figure spot boundary, and generates the boundary box information set;

[0019] Wherein, B ij represents the boundary envelope trend intensity value, C ik represents the boundary coordinate difference value of the kth point in the ith area in the figure spot contour edge, D jk represents the center expansion distance of the kth point in the jth area in the figure spot contour edge, M ijk represents the weighted matching factor between the boundary point and the matching degree value, P i represents the horizontal coordinate difference value of the upper boundary point to the center, Q j represents the vertical coordinate difference value of the lower boundary point to the center, and n represents the number of boundary points participating in the calculation in each area.

[0020] As a further scheme of the application, the boundary position checking module comprises:

[0021] The boundary point extraction submodule extracts the projection plane boundary position used by the figure spot based on the boundary points labeled in the boundary box information set, identifies the coordinate data and the space position index corresponding to the boundary points, performs boundary point integrity detection and figure spot boundary topology relationship judgment, and obtains the boundary coordinate index value;

[0022] The diagonal offset calculation submodule calls the boundary coordinate index value, selects the figure spot diagonal points corresponding to the reference boundary according to the figure spot boundary point set corresponding to the boundary coordinate index value, respectively collects the horizontal and vertical coordinate offset values, analyzes the space position change trend between the figure spot diagonal points and the reference mark line, and obtains the figure spot offset angle value set;

[0023] The boundary position resetting submodule adjusts the four position coordinates of the boundary of the image patch according to the offset direction and offset distance of the image patch in the set of image patch offset angle values, performs coordinate difference value correction and spatial re-projection operations, and generates a repositioned image patch encoding layer.

[0024] As a further scheme of the present application, the region encoding integration module comprises:

[0025] The image patch encoding acquisition submodule calls the repositioned image patch encoding layer, acquires the geographic encoding and acquisition time marker of each image patch, extracts the latitude and longitude coordinate values and time stamp values of the image patch according to its unique identification value, and structures and organizes the acquired parameters according to the image patch ID as an index field, constructs an image patch parameter set table, and generates an image patch encoding information set;

[0026] The encoding sequence construction submodule calls the image patch encoding information set, sorts all image patches according to the acquisition time marker, reconstructs the image patch data sequence according to the shooting time sequence, identifies the encoding difference value and spatial adjacency relationship between adjacent image patches, groups the image patches with continuous encoding and spatial adjacency, and generates an adjacent image patch sequence set;

[0027] The spatial segment combination submodule extracts the boundary coordinate points of each group of image patches based on the adjacent image patch sequence set, calculates the boundary coincidence degree and cross expansion value of each combined image patch, and judges the integrity boundary value of the segment combination, using the formula:

[0028]

[0029] The operation obtains the combined boundary coincidence degree value, filters the corresponding image patch combination according to the result that the boundary coincidence degree value is greater than the combined boundary threshold value, establishes the closed boundary path, and generates a combined boundary structure set;

[0030] wherein, B s represents the combined boundary coincidence degree value, C z represents the number of boundary points of the zth image patch, L z represents the boundary line segment length value of the zth image patch, A z represents the area value of the zth image patch, M z represents the boundary coincidence distance value of the zth image patch and the adjacent image patch, represents the average spatial adjacency distance value of the whole image patch group, D u represents the outer package boundary expansion length of the uth group of image patches, T u represents the number of image patches contained in the uth group of image patches, V u represents the internal boundary jump degree of the uth group of image patches, w represents the total number of image patches contained in the currently processed image patch combination, and m represents the number of outer package boundary expansion groups involved in the current image patch combination.

[0031] As a further scheme of the present application, the spatial variation splicing module comprises:

[0032] The boundary line extraction submodule calls the combined boundary structure set, acquires the original layer content of the vegetation coverage region block and the wetland distribution block, detects the continuous time layer frame of the block, calls the adjacent time frame block boundary line data, compares the boundary trend difference of the adjacent time frame block, numbers and identifies the contour of each frame, and generates a contour boundary number sequence;

[0033] The contour coincidence discrimination submodule compares the geometric coincidence rate of the adjacent block boundary lines in the continuous layer according to the contour boundary number sequence, calculates the overlap rate value and the spatial offset distance value of the contour boundary of each pair of adjacent blocks, judges whether the set contour offset threshold and the spatial position offset threshold are exceeded, and uses the formula:

[0034]

[0035] The operation acquires the block spatial offset degree value, acquires each block with boundary drift behavior according to the case of exceeding the offset threshold, records the starting contour number and the ending contour number, and obtains a boundary drift contour interval list;

[0036] Wherein, R disp represents the block spatial offset degree value, L e represents the boundary line length of the e-th block, ΔC e represents the contour center displacement amount between the e-th block and the next frame, O e represents the boundary overlap rate value of the e-th block, S e represents the boundary line average offset amount of the e-th block, and g represents the total number of block contour numbers;

[0037] The splicing block generation submodule calls the boundary drift contour interval list, filters the contour combination with a spatial overlap greater than the block splicing reference threshold according to the boundary line continuity between the contours and the block junction coordinates, sorts the contour numbers to form the block connection sequence, acquires the block boundary corresponding relationship, and generates a splicing block sequence.

[0038] As a further scheme of the present application, the ecological trend drawing module comprises:

[0039] The boundary splicing submodule calls the boundary information of each block in the splicing block sequence, extracts the spatial contour edge line segment, compares the splicing relationship of the adjacent block edge line segment, constructs a splicing path, collects the extension trend of the path in the two-dimensional space, and generates a block splicing extension line;

[0040] The direction recognition sub-module calls the tile splicing extension line, extracts an extension path set of multiple time periods, calculates a principal axis vector of the path on a spatial coordinate axis, compares the principal axis vector directions of different time periods in the same block, filters a direction distribution line that appears repeatedly for a set number of times, and obtains a repeated extension direction value;

[0041] The trend combination sub-module selects a line segment with consistent direction in the extension line according to the repeated extension direction value, extracts a spatial block identifier corresponding to the extension direction, judges spatial overlap degrees between the spatial block identifier and positioning points of a degradation continuous belt distribution line and a variation concentration area in the ecological scene, filters a line segment combination with an overlap degree exceeding a set threshold, splices to form a continuous spatial segment, and generates an ecological remote sensing data set.

[0042] Compared with the prior art, the advantages and positive effects of the present application are that:

[0043] In the present application, by accurately matching the geographical recognition information and the spatial boundary value, the coverage range and positioning accuracy of the feature map patch are significantly improved, the time and space information are effectively integrated, the data continuity and accuracy are enhanced, the boundary resetting technology effectively reduces the error of geographical information, the spatiotemporal continuity of data is improved, the continuous layer structure is constructed, and the data retrieval process is optimized, which greatly improves the processing efficiency, the fine boundary splicing technology also allows detailed tracking of ecological changes, provides more detailed data support for ecological monitoring, enhances the ability of ecological trend analysis, and thus improves the overall effect and application value of ecological monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 The system flowchart of the present application;

[0045] Figure 2 The remote sensing layer acquisition module flowchart of the present application;

[0046] Figure 3 The boundary position checking module flowchart of the present application;

[0047] Figure 4 The regional code integration module flowchart of the present application;

[0048] Figure 5 The spatial change splicing module flowchart of the present application;

[0049] Figure 6 The ecological trend drawing module flowchart of the present application. DETAILED DESCRIPTION

[0050] In order to make the objects, technical solutions and advantages of the present application clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and not to limit the present application.

[0051] In the description of the present application, it should be understood that the terms "length", "width", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like indicate the orientation or positional relationship shown in the drawings, which are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as limiting the present application. In addition, in the description of the present application, the meaning of "a plurality of" is two or more, unless otherwise explicitly and specifically limited.

[0052] Please refer to Figure 1 An ecological remote sensing data processing system comprises:

[0053] The remote sensing layer acquisition module acquires ground scanning image data carried on a remote sensing platform, identifies covered ground objects within the belonging area in combination with geographical identification information and spatial boundary values carried by each image, constructs corresponding image patch coverage ranges by position matching with known ground object spatial identifiers, and labels upper and lower boundary points of the image patches to acquire a boundary box information set;

[0054] The boundary position checking module extracts the projection plane boundary positions used based on the boundary points labeled in the boundary box information set, calls reference boundary markings in the target map, performs boundary resetting operations according to the diagonal point offset direction and distance size between the image patches and the reference boundaries, locates new four-corner positions of each image patch, and acquires a repositioned image patch coding layer;

[0055] The area coding integration module calls geographical coding and acquisition time markers of each image patch in the repositioned image patch coding layer, constructs a continuous layer structure according to the shooting time sequence, retrieves image patches that are continuously coded and spatially adjacent, and packs them into spatial fragment combinations, establishes spatial boundary lines for each combination, acquires a combination boundary structure set, and acquires a combination boundary structure set;

[0056] The spatial change splicing module calls the continuous time content of the fragments in the combination boundary structure set and the corresponding boundary lines, selects vegetation coverage area blocks and wetland distribution area blocks in the ecological remote sensing scene as combination objects, compares the coincidence degree of the contour positions of adjacent blocks in the continuous layer, judges whether there is a boundary drift behavior, records the start and end contour numbers of the blocks that have drifted, and generates a spliced block sequence according to the block boundary area;

[0057] The ecological trend drawing module calls the splicing path and spatial extension line of the splicing blocks in the splicing block sequence, retrieves the repeated extension direction and area expansion form appearing in multiple time periods, matches the degradation continuous strip distribution line and variation concentration area positioning point in the ecological scene, combines the spatial segments with repeated extension and convergent boundaries in the blocks into trend segments, and obtains an ecological remote sensing data set.

[0058] The bounding box information set includes the upper left boundary coordinates, the lower right boundary coordinates, the lower left boundary coordinates, and the upper right boundary coordinates of the feature map block, the repositioned block code layer includes the projected boundary plane coordinate set, the reference boundary mark space parameter, the offset direction identifier, and the distance threshold parameter, the combined boundary structure set includes the geographic code unique identifier, the time stamp index, the adjacent block spatial topology relationship, and the segment combination closed boundary line, the splicing block sequence includes the drift judgment result table, the contour offset matrix, and the intersection area splicing path vector set, and the ecological remote sensing data set includes the degradation continuous strip spatial trajectory coordinates, the variation concentration area polygon vertex set, the trend segment vector superposition unit, and the extension direction repetition identifier.

[0059] Please refer to Figure 2 , the remote sensing layer collection module includes:

[0060] The image geographic positioning submodule collects the ground scanning image data carried by the remote sensing platform, extracts the geographic recognition information carried by each image, compares the corresponding positions of the coordinate points, image time markers, and spatial boundary values in the image, calculates the spatial mapping point distribution of the image data, obtains the relative positioning result of the image in the geographic space, and generates an image space positioning coefficient value;

[0061] When collecting ground scanning image data carried by a remote sensing platform, the scanning path range and corresponding scanning resolution parameters need to be specified, for example, the platform is set to scan along the north-south direction with a spacing of 10 meters and a spatial resolution of 0.5 meters. The acquisition time of each image is recorded according to the image acquisition timestamp, and the original GPS coordinate information recorded by the image sensor is bound, such as the longitude and latitude coordinates carried by the image collected at a certain time (117.1524°E, 39.1236°N). Through coordinate correction, it is projected into the preset UTM projection system, and the mapping coordinates become (457200.3, 4336200.1). A mapping matrix is established by corresponding the mapping coordinates with the image pixel row and column numbers. Then, based on the matrix, the image boundary values are analyzed, for example, the image width and height are 4000x3000 pixels, the projection area covers 1000x750 meters, and the mapping distance per unit pixel is 0.25 meters / pixel. The actual position of any pixel point in the geographical space is derived, such as the actual coordinates corresponding to the 1500th row and 2000th column pixel are (457200.3+500, 4336200.1+375), i.e. (457700.3, 4336575.1). The above steps are batch operated when processing image group data to traverse the geographical recognition information of each pixel point in each image, and finally the image space positioning coefficient value is generated.

[0062] The image patch boundary recognition submodule calls the area range in the image located within the boundary value based on the image space positioning coefficient value, calculates the similarity coefficient between the pixel feature value, spatial distribution value in the area and the known ground object space identifier, and determines and filters the image patch in the area according to the similarity coefficient and the set spatial matching threshold value, obtains the spatial partition value of the boundary range covered area, and generates the spatial image patch matching degree value.

[0063] Based on the image space positioning coefficient value, the specific projection position of each remote sensing image in the geographical space can be determined, the image area in the range is called, the pixel data in the target area is extracted, a 3x3 sliding window is set to traverse the image gray value, the median, variance and range in each window are obtained as the pixel feature index, for example, the region center pixel value is 135, the maximum and minimum difference value of the neighborhood is 20, and the range is 20. The value is compared with the standard characteristic value of the known spatial identification of the ground object. Assuming that the standard characteristic value of the known ground object type A is [130, 150], the current pixel meets the preliminary matching condition, and the spatial distribution value is calculated based on the average gray mean value of the image region block and the spatial azimuth angle, for example, the direction of a certain plot is north, the average gray value is 140, and the spatial weight value is set to 0.9. The similarity coefficient between the normalized spatial matching function and type A is calculated, and if the similarity is greater than the set spatial matching threshold 0.85, it is considered as a successful match. The matching threshold is set to refer to the stable value interval of the spatial characteristics in the historical region, and the similarity mean value and standard deviation of 100 sample regions are calculated to obtain the matching threshold value, that is, threshold = 0.91-0.06 = 0.85. If the matching is successful, the plot region boundary is drawn, the spatial partition value corresponding to the region is obtained, and the spatial plot matching degree value is finally generated.

[0064] The boundary box generation submodule identifies the contour edge of the plot covered by the spatial plot matching degree value, judges the up-down-left-right boundary point position of the edge line segment in the image pixel grid, calculates the coordinate value of the boundary point set, and uses the formula:

[0065]

[0066] The operation obtains the boundary envelope trend intensity value, constructs the boundary point set according to the boundary envelope trend intensity value, obtains the closed graph area of the plot boundary, and generates the boundary box information set;

[0067] Wherein, B ij represents the boundary envelope trend intensity value, C ik represents the boundary coordinate difference value of the kth point in the ith region of the plot contour edge, D jk represents the center extension distance of the kth point in the jth region of the plot contour edge, M ijk represents the weighted matching factor between the boundary point and the matching degree value, P i represents the horizontal coordinate difference value of the upper boundary point to the center, Q j represents the vertical coordinate difference value of the lower boundary point to the center, and n represents the number of boundary points participating in the calculation in each region.

[0068] According to the spatial graph spot matching degree value, the boundary points of the spot contour are extracted, and the boundary envelope trend intensity value is calculated. After the boundary points are extracted, the horizontal and vertical coordinate difference values of the upper, lower, left and right boundary points are recorded. Assuming that in region 1, the upper boundary point is (500, 650), the lower boundary point is (500, 700), the left boundary point is (470, 675), and the right boundary point is (530, 675), then the horizontal difference value P i = 530-470 = 60, the vertical difference value Q j

[0069] = 700-650 = 50, the distance of each boundary point to the center point (500, 675) is measured in turn, the coordinate difference C i1 of point 1 is 30, the expansion distance D j1 is 25, the matching factor M ij1 is 0.88, the difference value C i2

[0070] of point 2 is 20, the expansion distance D j2 is 22, the matching factor M ij2 is 0.86, and the formula is:

[0071]

[0072] Molecular part calculation:

[0073] (30+25)·0.88 = 55·0.88 = 48.4;

[0074] (20+22)·0.86 = 42·0.86 = 36.12;

[0075] The sum of the numerator and denominator is 48.4+36.12 = 84.52;

[0076] Denominator part calculation:

[0077]

[0078] Final calculation:

[0079]

[0080] The result shows that the current spot region boundary intensity value is 1.082, indicating that the spot contour convergence is strong, and a closed boundary box can be generated by corresponding upper, lower, left and right boundary points. The value is within the set boundary envelope intensity reference interval [0.85, 1.15], and is determined to be valid. Finally, the boundary box information set is generated. By integrating the boundary difference value, the center expansion and the matching factor, and adding the normalization scale denominator, the boundary intensity index has better geometric scale adaptability and feature fusion, thereby improving the accuracy of closed boundary construction.

[0081] Please refer toFigure 3 The boundary position checking module comprises:

[0082] The boundary point extraction submodule extracts the boundary position of the projection plane used by the polygon based on the boundary points labeled in the boundary box information set, identifies the coordinate data and spatial position index corresponding to the boundary points, performs boundary point integrity detection and polygon boundary topology relationship judgment, and obtains the boundary coordinate index value.

[0083] First, the boundary point coordinate data corresponding to each polygon is extracted. This process is usually based on the polygon vector file derived from the geographic information platform. Each polygon has boundary points, which are marked with the southwest, southeast, northeast, and northwest corner points. The polygon boundary is stored as a closed polygon. In the implementation, polygon T001 is taken as an example. The four corner points are (352000.25, 4070000.75), (352000.25, 4070002.65), (352002.15, 4070002.65), and (352002.15, 4070000.75), respectively. The boundary data is complete and meets the closure condition. When performing coordinate conversion operation on the polygon data, the projection coordinate system needs to be unified. In this implementation, the WGS84UTMZone50N coordinate system is used. After conversion, the last two digits after the decimal point are retained to ensure coordinate accuracy. The coordinate accuracy range is set to ±0.05 meters. Polygons exceeding this error range need to recalculate the original point position. Subsequently, boundary point integrity detection is performed. The four corner points are compared one by one to determine whether there are missing or duplicate records. In polygon T002, if there are two points with identical coordinates or one corner point is missing, it is judged as an incomplete polygon. Such polygons are temporarily excluded from subsequent processing after being marked as abnormal polygons. In addition to the number of boundary points, it is also necessary to determine whether the connection between the points is intersected or not closed. The distance continuity judgment method is used. If the distance between two adjacent points is greater than 1.5 meters and not closed, it is recorded as a structurally abnormal polygon. When judging the topology relationship, the polygon and its adjacent polygons need to be matched and detected. If the coincidence degree is less than 90%, it is judged that the boundary is not closed and continuous. The threshold value of the coincidence degree is set based on the actual distribution characteristics of the regional polygons. Based on the polygon density characteristics of urban construction areas, 90% is determined as the critical value of structural stability. This value is obtained by sampling and comparing a large number of boundary overlapping polygons. For example, 30 groups of polygons are taken as samples. The average boundary coincidence rate is 92.8%, and the standard deviation is 1.3%. The lower value of 90% is selected as the judgment basis. Finally, the boundary coordinate index set that has completed projection conversion and structure checking is output, and the boundary coordinate index value is obtained.

[0084] The diagonal offset calculation submodule calls the boundary coordinate index value, selects the polygon diagonal points corresponding to the reference boundary based on the polygon boundary point set corresponding to the boundary coordinate index value, collects the horizontal and vertical coordinate offset values, analyzes the spatial position change trend between the polygon diagonal points and the reference mark line, and obtains the polygon offset angle value set.

[0085] The boundary coordinate index value is called, and the spatial relationship between each polygon and the reference boundary is analyzed according to the corner point coordinates of the polygon positioned in the boundary coordinate index value. In actual operation, the corresponding area range of the polygon and the reference boundary needs to be locked first to ensure the effectiveness of the comparison between the same block polygons. The southwest corner point (352000.25, 4070000.75) of the polygon T001 is extracted, and the corresponding reference boundary point is (352002.15, 4070002.65). The horizontal distance is 1.90 meters and the vertical distance is 1.90 meters through coordinate difference calculation. The coordinate difference value is combined with the polygon direction angle. The polygon direction angle θ is set to 45.0 degrees, and the reference boundary angle β is 45.2 degrees. The angle difference between the two is 0.2 degrees. According to the engineering surveying and mapping accuracy standard, the angle difference threshold is set to 0.5 degrees. That is, if the direction difference exceeds this range, it is marked as a boundary direction abnormal polygon. The setting basis is the urban planning polygon edge error control standard. The average value of the angle difference in the regional sample is 0.36 degrees, so 0.5 degrees is selected as the maximum allowed deviation. The direction angle difference between the polygon T002 and the reference boundary is 0.5 degrees, which is critical. The difference value of T003 is 0.2 degrees, which is within the acceptable range. The corresponding offset values are 2.4 meters and 1.3 meters respectively. In the offset analysis, if the polygon offset exceeds 2.5 meters, it is marked as a serious offset polygon. This offset distance threshold comes from the analysis of the historical data of the existing polygon offset. The average value of the polygon offset in the urban area is 1.8 meters, and the safety threshold is set to 2.5 meters as the acceptable upper limit. T001, T002 and T003 are 1.9, 2.4 and 1.3 respectively, which do not exceed the set threshold. The polygon number, offset distance and direction angle difference value of each polygon are recorded as a structure vector for subsequent position adjustment call to obtain the polygon offset angle value set.

[0086] The boundary position resetting submodule adjusts the four position coordinates of the polygon boundary according to the offset direction and offset distance of the polygon in the polygon offset angle value set, performs coordinate difference correction and spatial re-projection operation, and generates a repositioned polygon encoding layer.

[0087] According to the offset direction and offset distance of each patch in the set of patch offset angle values, the four-corner position is repositioned by using the coordinate adjustment mode. Before the patch is repositioned, the superposition relationship between the four-corner points and the offset vector is analyzed. For example, the offset distance of the opposite corner of the T001 patch is 1.90 meters, and the direction angle is 45 degrees. According to the vector decomposition, the offset vector is decomposed into X and Y directions, and the decomposed values are 1.34 meters and 1.34 meters respectively. The X and Y coordinates of the four corner points of T001 are added respectively to obtain the new boundary point coordinates, which are (352001.59, 4070002.09), (352001.59, 4070000.89), (352003.49, 4070002.09), and (352003.49, 4070000.89) respectively. In this process, it is necessary to ensure that the four edges remain closed, that is, the boundary is repositioned while the topological structure of the original patch is retained. At the same time, it is checked whether the joint between adjacent patches is broken due to repositioning. If the distance between adjacent patches is less than 0.3 meters, the system automatically performs edge mending repair operation. The value is taken from the boundary precision error tolerance standard. The repositioned patch is again calculated by the coordinate index calculation method to generate a new boundary coordinate set, and is encoded according to the patch number. At the same time, the layer attribute field is updated, including patch number, four-corner coordinates, offset value, etc. Finally, all the patches are output as a new layer file to obtain the repositioned patch encoding layer.

[0088] Please refer to Figure 4 The regional encoding integration module includes:

[0089] The patch encoding acquisition submodule calls the repositioned patch encoding layer, acquires the geographic encoding and acquisition time mark of each patch, extracts the latitude and longitude coordinate values and timestamp values according to the unique identification value of the patch, and structures the collected parameters according to the patch ID as the index field to build a patch parameter set table and generate a patch encoding information set.

[0090] First, according to the plot coding layer, the unique identification value of the plot is extracted one by one, which can be a number such as GID001, GID002, etc. The corresponding geographic code and collection time mark are read as index fields. The geographic code is composed of longitude and latitude, for example, the center point coordinates of plot GID001 are (112.3512, 24.5678), and the collection time mark is "2023-09-1014:35:22". The longitude and latitude values are directly read from the geometry attribute in the layer data structure, and the collection time is retrieved from the attribute field. Then the above plot ID, longitude and latitude, and time stamp are combined into a structured record in the form of (GID001, 112.3512, 24.5678, 20230910143522). This process iterates through all plots in the layer to extract them. When the layer contains 500 plots, parameter grabbing needs to be executed 500 times in a loop, and is filled into the plot parameter set array in turn;

[0091] If the parameters of plot GID002 are (112.3531, 24.5682, 20230910143711);

[0092] GID003 is (112.3540, 24.5690, 20230910143900);

[0093] The generated parameter set is:

[0094] {(GID001, 112.3512, 24.5678, 20230910143522),

[0095] (GID002, 112.3531, 24.5682, 20230910143711),

[0096] (GID003, 112.3540, 24.5690, 20230910143900),…};

[0097] This set is used as the plot coding information set for subsequent sorting and grouping links.

[0098] The coding sequence construction submodule calls the plot coding information set, sorts all plots according to the collection time mark of the plot, reconstructs the plot data sequence according to the shooting time sequence, and identifies the coding difference and spatial adjacency relationship between adjacent plots. The plots with continuous coding and spatial adjacency are grouped to generate an adjacent plot sequence set.

[0099] Firstly, the graph patch encoding information set is called, all the graph patches are arranged in ascending order according to the timestamp field, and the execution process is standardized by processing the string format time mark '20230910143522', and then sorted one by one after conversion to a comparable format, such as the time of graph patch GID001, GID002, GID003 is 14:35:22, 14:37:11, 14:39:00, and the sorting result is GID001→GID002→

[0100] GID003, after sorting, the adjacent graph patch pairs are traversed in turn, and it is judged whether the difference between the graph patch encoding values is a unit increment, for example, if the GID suffix number is a continuous number, the difference is 1, the difference value is calculated by taking the difference between the two graph patch ID suffix numbers, such as GID002-GID001=2-1=1, then the encoding continuous condition is met, and the spatial distance between the center coordinates of the two graph patches is calculated, and the spherical distance simplified calculation formula is adopted:

[0101]

[0102] Where Δx is the difference in longitude, and Δy is the difference in latitude, in degrees, for example, GID001(112.3512, 24.5678), GID002(112.3531, 24.5682), then Δx=0.0019, Δy=0.0004,

[0103]

[0104] If the threshold is set to 0.5, it is recognized as spatial adjacency, and when the condition is met, the graph patch pair is classified into the same group, and the above judgment is continued for all sorted graph patches to obtain multiple combinations of groups that meet the conditions, such as combination group 1 {GID001, GID002, GID003}, combination group 2 {GID005, GID006}, etc., each group meets the encoding continuous and spatial adjacency conditions, and is merged to form an adjacent graph patch sequence set.

[0105] The spatial fragment combination submodule extracts the boundary coordinate points of each group of graph patches based on the adjacent graph patch sequence set, calculates the boundary coincidence degree and intersection expansion value of each combined graph patch, and judges the integrity boundary value of the fragment combination, using the formula:

[0106]

[0107] The combined boundary consistency value is obtained by operation, and the corresponding graph patch combination is screened according to the result that the boundary consistency value is greater than the combination boundary threshold value, a closed boundary path is established, and a combination boundary structure set is generated;

[0108] Where, B s represents the combined boundary consistency value, Cz L represents the number of boundary points of the z-th patch. z A represents the length of the boundary line segment of the z-th patch. z M represents the area value of the z-th patch. z This represents the overlap distance between the z-th polygon and its adjacent polygons. D represents the average spatial adjacency distance of the entire image patch group. u T represents the outer boundary extension length of the u-th patch group. u V represents the number of polygons contained in the u-th polygon group. u represents the internal boundary jump degree of the u-th group of patches, w represents the total number of patches contained in the current patch combination being processed, and m represents the number of outer boundary extension groups involved in the current patch combination;

[0109] After obtaining the set of adjacent polygon sequences, the consistency value of the combined boundaries needs to be calculated. The process first extracts parameters such as the number of boundary points, boundary segment length, area, and the length of the boundary overlapping with adjacent polygons for each group of polygons. Then, the average adjacency distance, boundary extension length, number of polygons, and boundary jump degree between the combined groups are further calculated. Taking the combined group {GID010, GID011, GID012} as an example, the parameters are as follows: GID010 has 35 boundary points, a boundary length of 145m, an area of ​​1620㎡, and a boundary overlap of 40m with GID011; GID011 has 33 points, a boundary length of 140m, an area of ​​1500㎡, and an overlap of 35m; GID012 has 36 points, a boundary length of 150m, an area of ​​1580㎡, and an overlap of 30m. The average adjacency distance of the combined groups is 13.2m, the boundary extension lengths are 12m and 10m respectively, the number of polygons is 3, and the boundary jump degrees are 2 and 1 respectively.

[0110] Substitute the above parameters into the consistency formula:

[0111]

[0112] The calculation steps are as follows:

[0113] Given C1 = 35, L1 = 145, and M1 = 40, we get: 35·(145 + 40.25)

[0114] =35·185.25=6483.75, the difference is |6483.75-40|=6443.75;

[0115] Given C² = 33, L² = 140, and M² = 35, we get: 33·(140 + 38.73)

[0116] = 33 · 178.73 = 5898.09, difference = |5898.09 - 35| = 5863.09;

[0117] C3 = 36, L3 = 150, M3 = 30, we get: 36 · (150 + 39.75)

[0118] = 36 · 189.75 = 6831, difference = |6831 - 30| = 6801;

[0119] The total sum of the molecules is: 6443.75 + 5863.09 + 6801 = 19107.84;

[0120] The denominator part: D1 = 12, T1 = 3, V1 = 2; V2 = 1;

[0121] The sum is: 13.2 + (6.93 - 2) + (5.77 - 1) = 13.2 + 4.93 + 4.77 = 22.9;

[0122] The denominator is: 3 · 22.9 = 68.7;

[0123] Final calculation result:

[0124]

[0125] The result indicates that the boundary consistency degree of the patch combination is 278.21, and if the boundary consistency degree threshold is set to 250, the determination condition is met, and the result indicates that a closed boundary path can be established and the combination boundary structure set of the combination is output. The benefit of the formula is that the spatial structure integrity is quantified explicitly through multi-parameter joint calculation (such as the deduction term of the boundary point number, the square root of the area, the boundary jump degree, etc.), which is helpful for screening patch combinations with stable boundary structures in a large data scenario.

[0126] Please refer to Figure 5 , the spatial change splicing module comprises:

[0127] The boundary line extraction submodule calls the combination boundary structure set, obtains the original layer content of the vegetation coverage area block and the wetland distribution area block, detects the continuous time layer frame of the block, calls the adjacent time frame block boundary line data, compares the boundary direction difference of the adjacent time frame blocks, numbers and identifies the contour of each frame of block, and generates a contour boundary number sequence;

[0128] Firstly, the continuous remote sensing image layers from May 2024 to August 2024 need to be read, and the corresponding vegetation tiles and wetland tiles in each month are extracted from the original layers through coordinate labels. For example, the area with a coordinate range of (100, 200)-(500, 600) is marked as the same area index in each frame of image to ensure time sequence consistency. After extraction, the boundary of each frame tile is expressed in the form of polygon contour, and after vectorization processing of the contour boundary, equidistant boundary nodes on the contour are selected, such as selecting a node every 5 meters to ensure consistent node number. The nodes are represented by two-dimensional coordinates. Then, for each contour node sequence, the angle change between nodes is calculated and the boundary direction is marked;

[0129] The boundary node of the first tile in the May 2024 frame is [(120.3, 210.4), (125.1, 213.7), (130.0, 215.6), (135.5, 214.2)];

[0130] And the corresponding tile nodes in June 2024 are [(120.9, 211.1), (125.8, 214.4),

[0131] (130.5, 216.0), (135.9, 214.8)], the contour change between adjacent time frames is analyzed by node-by-node difference analysis, and the boundary change point is marked at the node where the direction angle change exceeds 15 degrees. For the contour number, the layer time frame index and position index of the tile where the contour is located are used for double numbering, such as "202405_03" representing the third tile in May 2024. By processing all time frame tile contours as described above, a complete contour boundary numbering sequence is formed, such as [202405_01, 202405_02, …, 202408_01, 202408_02].

[0132] The contour coincidence discrimination submodule compares the geometric coincidence rate of the boundary lines of adjacent tiles in consecutive layers according to the contour boundary numbering sequence, calculates the overlap rate value and spatial offset distance value of each pair of adjacent tile contours, judges whether it exceeds the set contour offset threshold and spatial position offset threshold, and uses the formula:

[0133]

[0134] The operation obtains the spatial offset degree value of the tile, and according to the case of exceeding the offset threshold, it obtains each tile with boundary drift behavior and records the starting contour number and ending contour number to obtain the boundary drift contour interval list;

[0135] Where, R disp represents the spatial offset degree value of the tile, L e represents the length of the boundary line of the e-th tile, and ΔCe representing the contour center displacement between the two frames before and after the e-th tile, O e representing the boundary overlap rate value of the e-th tile, S e representing the average boundary offset of the e-th tile, g represents the total number of tile contour numbers;

[0136] According to the contour boundary number sequence, the tile space offset degree value R between the continuous layers disp Each item can be calculated in the following manner. Assuming that there are three tiles, e1, e2, and e3, and their corresponding boundary line lengths, contour center displacement amounts, boundary overlap rates, and average boundary offsets are as follows:

[0137] Tile e1: L1 = 110 m, ΔC1 = 4.2 m, O1 = 0.68, S1 = 2.0 m;

[0138] Tile e2: L2 = 125 m, ΔC2 = 3.6 m, O2 = 0.61, S2 = 1.9 m;

[0139] Tile e3: L3 = 98 m, ΔC3 = 5.1 m, O3 = 0.52, S3 = 2.4 m;

[0140] Substitute into the original formula:

[0141]

[0142] The items are expanded as follows:

[0143]

[0144] Finally, sum the three items:

[0145] R disp = |218.7 + 225.6 + 203.5| = |647.8| = 647.8;

[0146] Set the offset threshold to 400. Comparison shows that 647.8 > 400, so it is considered that there is a spatial offset phenomenon in the region, and the marked contour interval list is: [(202405_01)-(202406_01)], [(202406_02)-(202407_02)], and [(202407_03)-(202408_03)].

[0147] By introducing the product of the contour length and the center displacement combined with the boundary overlap rate and the average offset to construct the denominator, the strength of the influence of the tile size on the offset is quantitatively processed; the denominator part provides double regulation of the overlap accuracy and the average offset, so that the model has high sensitivity to local changes and overall trends.

[0148] The spliced tile generation submodule calls the boundary drift contour interval list, filters the contour combinations with spatial overlap greater than the tile splicing reference threshold according to the contour boundary line continuity and tile junction coordinates, sorts the contour numbers to form the tile connection sequence, and generates the spliced tile sequence after obtaining the tile boundary corresponding relationship;

[0149] After receiving the boundary drift contour interval list, the spliced tile generation submodule first analyzes the node continuity of the spatial boundary of each contour combination. For example, contour 202405_01 ends at point (300.0, 450.0), and adjacent contour 202406_01 starts at (300.7, 450.5), so the distance between the two points is less than the set continuity judgment threshold of 1.2 meters, which is judged to be continuous. Then the spatial overlap area between the two contours is calculated. If the overlap area of contour 202405_01 and 202406_01 is 43.6 square meters, and the combined area of the tiles is 125 square meters, then the overlap rate is 0.35, it can be judged that the current overlap rate is slightly low and does not meet the standard. If the other two contours 202406_02 and 202407_02 overlap an area of 58.5 square meters, and the combined area is 130 square meters, then the overlap rate is 0.45>0.35, which meets the splicing requirements. The contour numbers of all contours that meet the splicing conditions are sorted in ascending order, such as 202406_02<202407_02<202408_02, and the tile connection sequence is determined as [202406_02, 202407_02, 202408_02]. The unified splicing boundary is generated by weighted averaging the node corresponding relationship on the contour boundary, for example, the splicing point coordinates are:

[0150] The final spliced tile sequence is 202406_02~202408_02, and the number is "20240608_A".

[0151] Please refer to Figure 6 , the ecological trend drawing module includes:

[0152] The boundary splicing submodule calls the boundary information of each tile in the spliced tile sequence, extracts the spatial contour edge line segment, compares the splicing relationship of the edge line segments of adjacent tiles, constructs the splicing path, and collects the extension trend of the path in the two-dimensional space to generate the tile splicing extension line.

[0153] Based on the boundary information of each tile in the tile sequence, the boundary coordinate set of a single tile needs to be extracted first, which is usually derived from the vectorization output in the pre-processing stage of remote sensing data. For example, the boundary node sequence of tile A01 is [(105.2, 28.4), (106.0, 28.4), (106.0, 29.0), (105.2, 29.0)], after forming the boundary contour polyline, it can be further divided into an array of directional line segments, and the start and end coordinates of each line segment are used as the basic data for judging the splicing relationship. For the edge line segments of multiple tiles, the Euclidean distance between the end point and the start point of other tiles needs to be extracted, and it is judged whether there is a proximity condition. At the same time, the direction vector of the two line segments is extracted, and the angle value is calculated. On this basis, the splicing condition threshold is set, that is, the splicing distance threshold is set to 15 meters, and the splicing angle difference threshold is set to 20 degrees. The threshold is set based on the statistical analysis of regional feature scale characteristics, which is 0.25 times the average minimum boundary size of the regional land, and is determined by expert field survey. For example, if the boundary end point of tile A01 is (106.0, 28.4) and the boundary start point of tile A02 is (106.0, 28.55), the distance between the two points is 11.1 meters, and the angle is 13 degrees, which meets the splicing standard, so the two tiles are considered as boundary continuous. In this way, the boundary information between tiles is processed pair by pair, and finally a set of spatial continuous splicing paths is formed. The spatial trend of the splicing path is calculated by the coordinate increment of each line segment, for example, the line segment direction from A01 to A02 is north, the extension distance is 0.15 kilometers, and the path trend vector is (0, +0.15). All the splicing path trend information is summarized and recorded, and finally the tile splicing extension line is obtained.

[0154] The direction recognition submodule calls the tile splicing extension line, extracts the extension path set of multiple time periods, calculates the principal axis vector of the path on the spatial coordinate axis, compares the principal axis vector direction of different time periods in the same block, and selects the direction distribution line whose direction repeatedly appears for a set number of times to obtain the repeated extension direction value.

[0155] After calling the tile splicing extension line, the path trend at different time stages is uniformly analyzed and processed, and the years 2020, 2021, 2023, etc. are selected as the analysis time period, the extension line segment is uniformly mapped in the 1km×1km spatial grid, and the direction vector of each line segment in the grid is recorded, which is measured in the form of degree, such as the line segment extending from (105.2, 28.4) to (105.2, 29.4), the direction is 0 degrees. The path direction of the same block at different time periods is compared, and the included angle difference calculation method is used. If the direction in 2020 is 5 degrees, the direction in 2021 is 7 degrees, and the direction in 2023 is 3 degrees, the difference between the two is 2 degrees, 4 degrees, and 4 degrees, respectively, which are all within the set direction consistency threshold of 10 degrees. The threshold setting basis is that the maximum main direction fluctuation observed in local vegetation change does not exceed 10 degrees, and it is verified by regional wind direction data and stable direction of mountain trend. The number of times that each spatial grid meets the direction consistency determination condition is counted. If a direction appears more than twice in three periods, it is identified as a repeated direction. In the above example, the trend of about 5 degrees appears three times in a row, which meets the condition. All grid blocks that meet the requirements are counted and the direction angle is marked, and finally the repeated extension direction value is obtained.

[0156] The trend combination sub-module selects the line segments with consistent directions in the extension line according to the repeated extension direction value, extracts the spatial block identifier corresponding to the extension direction, judges the spatial overlap degree between the line segments and the positioning points in the degenerated continuous area of the variation center, selects the line segment combination with an overlap degree exceeding the set threshold, splices to form a continuous spatial segment, and generates an ecological remote sensing data set;

[0157] According to the repeated extension direction value, the line segments with similar directions in the tile splicing extension line are selected, for example, if the repeated direction is 5 degrees, the tile line segment combination with a direction angle between 2-8 degrees is selected, the corresponding spatial block numbers such as A02, B03, B04 are recorded, and the boundary vector range of these blocks is further extracted. The intersection operation is performed between the spatial superposition and the degenerated continuous area vector range to obtain the spatial overlap area data, and then the ratio operation is performed with the total area of the spatial unit where the tile is located. For example, the intersection area of B03 region and degenerated band is 0.48km 2 , the total area is 1.25km 2, the overlap ratio is calculated as 38.4%, which is higher than the set threshold of 30% overlap, and thus is retained. The overlap threshold is set based on the statistical analysis of typical degraded areas, combined with remote sensing interpretation data for more than 5 years to evaluate the minimum effective coverage proportion of spatial distribution concentration area. Further, the spatial range of the qualified tiles is compared with the positioning point data of the variation concentration area, and buffer analysis is used to generate a buffer zone with a radius of 100 meters centered on each positioning point. If the southern edge of tile B03 intersects with a buffer zone, it is considered a hit point. If the number of hits reaches 2, it meets the variation concentration determination standard, and thus B03 is confirmed to have multiple overlap characteristics. Finally, all tiles that meet the direction consistency, degradation zone overlap ratio, and variation positioning point hit are integrated to form a spatial continuous paragraph, and an ecological remote sensing data set is generated.

[0158] The above is only a preferred embodiment of the present application, and does not limit the present application in other forms. Any skilled person in the art can modify or change the above disclosed technical content to equivalent embodiments applied to other fields, but any simple modification, equivalent change and modification made to the above embodiments without departing from the technical solution content of the present application, according to the technical essence of the present application, still belongs to the protection scope of the technical solution of the present application.

Claims

1. An ecological remote sensing data processing system, characterized by: The system comprises: The remote sensing layer acquisition module acquires remote sensing platform scanning images, identifies covered ground objects in combination with image geographic information and boundary values, matches known identifiers, constructs plot ranges, labels upper, lower, left and right boundary points, and generates a boundary box information set; The boundary position checking module extracts a projection plane position based on the boundary box information set, calls a map reference boundary marking line, performs boundary resetting according to the offset direction and distance of the corner points of the plot and the reference boundary, repositions the four corner positions of the plot, and generates a repositioned plot encoding layer; The regional coding integration module calls the repositioned plot encoding layer, constructs a layer in chronological order, retrieves plots that are continuously coded and adjacent, packages them into spatial segments, and generates a combined boundary structure set; The spatial change splicing module calls the combined boundary structure set, selects vegetation and wetland blocks, compares the contour coincidence degree of adjacent blocks, judges boundary drift, records the contour number, and generates a spliced block sequence; The ecological trend drawing module calls the spliced block sequence, retrieves repeated extension directions and expansion morphologies, matches degradation zones and variation points, combines converging blocks into trend segments, and obtains an ecological remote sensing data set; The boundary position checking module comprises: The boundary point extraction submodule extracts the projection plane boundary position used by the plot based on the boundary points labeled in the boundary box information set, identifies the coordinate data and spatial position index corresponding to the boundary points, performs boundary point integrity detection and plot boundary topology relationship judgment, and obtains boundary coordinate index values; The diagonal offset calculation submodule calls the boundary coordinate index values, selects the plot diagonal points corresponding to the reference boundary according to the plot boundary point set corresponding to the boundary coordinate index values, collects horizontal and vertical coordinate offset values, analyzes the spatial position change trend between the plot diagonal points and the reference marking line, and obtains a plot offset angle value set; The boundary position resetting submodule adjusts the four position coordinates of the plot boundary according to the offset direction and offset distance of the plot in the plot offset angle value set, performs coordinate difference correction and spatial re-projection operations, and generates a repositioned plot encoding layer.

2. The ecological remote sensing data processing system of claim 1, wherein: The boundary box information set includes the upper left boundary coordinates, the lower right boundary coordinates, the lower left boundary coordinates, and the upper right boundary coordinates of the ground object plot, the repositioned plot encoding layer includes a projection boundary plane coordinate set, a reference boundary marking line space parameter, an offset direction identifier, and a distance threshold parameter, the combined boundary structure set includes a geographic coding unique identifier, a time stamp index, an adjacent plot spatial topology relationship, and a segment combination closed boundary line, the spliced block sequence includes a drift judgment result table, a contour offset matrix, and a junction area splicing path vector set, and the ecological remote sensing data set includes a degradation continuous zone space trajectory coordinate, a variation concentrated area polygon vertex set, a trend segment vector superposition unit, and a repeated direction extension degree identifier.

3. The ecological remote sensing data processing system of claim 1, wherein: The remote sensing layer acquisition module comprises: The image geolocation sub-module collects ground scanning image data carried by a remote sensing platform, extracts geographical identification information carried by each image, compares the corresponding positions of coordinate points in the image, image time markers and space boundary values, calculates the spatial mapping point distribution of the image data, obtains the positioning of the image in the geographical space, and generates image space positioning coefficient values; The image patch boundary recognition sub-module calls the area range in the image located within the boundary value based on the image space positioning coefficient values, calculates the similarity coefficient between the pixel feature value, the spatial distribution value in the area and the known spatial identifier of the ground object, and determines and screens the image patch in the area according to the similarity coefficient and the set spatial matching threshold value, obtains the spatial partition value of the boundary range coverage area, and generates a spatial image patch matching degree value; The boundary box generation sub-module identifies the contour edge of the covered image patch according to the spatial image patch matching degree value, judges the up, down, left and right boundary point positions of the edge line segment in the image pixel grid, calculates the coordinate values of the boundary point set, and obtains the boundary envelope trend intensity value by using the formula: ； The boundary box information set is generated by constructing the boundary point set according to the boundary envelope trend intensity value, obtaining the closed graphic area of the image patch boundary, and constructing the boundary box information set; wherein, representing a boundary envelope trend intensity value, representing a boundary coordinate difference value of the kth point in the ith region in the profile edge of the map plot, representing a center expansion distance of the kth point in the jth region in the profile edge of the map plot, representing a weighted matching factor between the boundary point and the matching degree value, representing a horizontal coordinate difference value of the upper boundary point to the center, representing a vertical coordinate difference value of the lower boundary point to the center, representing the number of boundary points participating in the calculation in each region.

4. The ecological remote sensing data processing system of claim 1, wherein: The region coding integration module comprises: The image patch coding acquisition sub-module calls the repositioned image patch coding layer, obtains the geographical coding and acquisition time marker of each image patch, respectively extracts the latitude and longitude coordinate values and time stamp values according to the unique identifier value of the image patch, and structurally organizes the collected parameters with the image patch as the index field, constructs an image patch parameter set table, and generates an image patch coding information set; The coding sequence construction sub-module calls the image patch coding information set, sorts all image patches according to the acquisition time marker of the image patch, reconstructs the image patch data sequence according to the shooting time sequence, identifies the coding difference value and spatial adjacency relationship between adjacent image patches, groups the image patches with continuous coding and spatial adjacency, and generates an adjacent image patch sequence set; The spatial segment combination sub-module extracts the boundary coordinate points of each group of image patches based on the adjacent image patch sequence set, calculates the boundary coincidence degree and cross expansion value of the image patch, and judges the integrity boundary value of the segment combination, and obtains the combination boundary consistency value by using the formula: ； The combination boundary structure set is generated by establishing a closed boundary path according to the result of screening the corresponding image patch combination according to the boundary consistency value greater than the combination boundary threshold value; wherein, representing the combination boundary consistency value, representing the number of boundary points of the representing the boundary line segment length value of the representing the area value of the representing the boundary coincidence distance value of the representing the average spatial adjacency distance value of the full group of map patches, representing the outer bounding boundary extension length of the representing the number of groups of map patches, representing the internal boundary jump degree of the representing the total number of map patches contained in the current processed map patch combination, representing the number of outer bounding boundary extension groups involved in the current map patch combination.​​​​​​​ 5. The ecological remote sensing data processing system of claim 1, wherein: The spatial change splicing module comprises: The boundary line extraction sub-module calls the combination boundary structure set, obtains the original layer content of the vegetation coverage region block and the wetland distribution area block, detects the continuous time layer frame of the block, calls the adjacent time frame block boundary line data, compares the boundary trend difference of the adjacent time frame blocks, numbers and identifies the contour of each frame block, and generates a contour boundary number sequence; The contour coincidence discrimination sub-module compares the geometric coincidence rate of the adjacent block boundary lines in the continuous layer according to the contour boundary number sequence, calculates the overlap rate value and spatial offset distance value of each pair of adjacent block contour boundaries, judges whether the set contour offset threshold value and the spatial position offset threshold value are exceeded, and obtains the combination boundary consistency value by using the formula: ； The operation obtains a tile spatial offset degree value, obtains each tile with a boundary shift behavior according to a case of exceeding an offset threshold, records a starting contour number and a terminal contour number, and obtains a boundary shift contour interval list; wherein, represents a tile space offset degree value, represents a boundary line length of the e-th tile, represents a contour center displacement amount between the two frames before and after the e-th tile, represents a boundary overlap rate value of the e-th tile, represents a boundary line average offset amount of the e-th tile, represents a total number of tile contour numbers; The spliced tile generation submodule calls the boundary shift contour interval list, filters contour combinations with spatial overlap greater than a tile splicing reference threshold according to boundary line continuity between contours and tile junction coordinates, sorts contour numbers to form a tile connection sequence, obtains a tile boundary corresponding relationship, and generates a spliced tile sequence.

6. The ecological remote sensing data processing system of claim 1, wherein: The ecological trend drawing module includes: The boundary splicing submodule calls boundary information of each tile in the spliced tile sequence, extracts spatial contour edge line segments, compares splicing relationships of adjacent tile edge line segments, constructs a splicing path, collects an extension trend of the path in a two-dimensional space, and generates a tile splicing extension line; The direction recognition submodule calls the tile splicing extension line, extracts a plurality of time period extension path sets, calculates a main axis vector of the path on a spatial coordinate axis, compares directions of the main axis vector of different time periods in the same block, filters a direction distribution line with a repeated direction reaching a set number of times, and obtains a repeated extension direction value; The trend combination submodule selects line segments with consistent directions in the extension line according to the repeated extension direction value, extracts spatial block identifiers corresponding to the extension direction, judges spatial overlap degrees between the spatial block identifiers and positioning points of a degradation continuous belt distribution line and a mutation concentrated area in an ecological scene, filters line segment combinations with an overlap degree exceeding a set threshold, splices to form a continuous spatial segment, and generates an ecological remote sensing data set.

Images