A dynamic change detection result positioning display method for illegal encroachment behavior

Abstract

The present application relates to the technical field of land analysis, and particularly relates to a dynamic change detection result positioning display method for illegal occupation behavior, comprising the following steps: based on continuous time series remote sensing images of a target area, extracting geometric deformation trajectories and expansion directions of a change area, and generating an occupation behavior mode vector; performing space-time matching on the occupation behavior mode vector and a preset ownership database, and outputting a responsibility subject binding list; superimposing the occupation behavior mode vector and the responsibility subject binding list on a digital elevation model, and rendering an interactive occupation behavior positioning simulation sand table. Through a surface curvature compensation algorithm for fine adjustment of the vertical position of a cube, the present application ensures that in complex terrain areas such as slope land and mine pits, the spatial positioning accuracy error of the occupation trajectory is reduced, the geometric consistency between the trajectory line and the terrain in sand table display is improved, and high-precision support is provided for space-time tracing in visualization.

Description

Technical Field

[0001] This invention relates to the field of land analysis technology, and in particular to a method for locating and displaying dynamic change detection results of illegal encroachment. Background Technology

[0002] In fields such as natural resource protection and land supervision, monitoring and evidence collection of encroachment has always been a core challenge. With the development of remote sensing monitoring, UAV aerial surveying and spatial behavior modeling technologies, the spatiotemporal identification capabilities of encroachment have been continuously enhanced. However, the relevant results are mostly presented in the form of images, curves or two-dimensional trajectories, lacking support for three-dimensional expression and dynamic backtracking of the encroachment process under real terrain, which makes it difficult to meet the actual needs such as the visualization of responsibility.

[0003] Currently, encroachment trajectories are mostly projected in a planar manner, ignoring the impact of terrain undulations on the behavioral path. This is particularly problematic in hilly, mining, or steep areas, leading to projection distortion, path overlap, or misjudgment, affecting the authenticity and readability of the spatiotemporal chain of the behavior. Furthermore, encroachment area and time information are often presented in text or chart form, lacking a visually quantifiable model at a unified spatial scale. This hinders law enforcement personnel from intuitively understanding and judging the intensity, evolution, and legal consequences of encroachment. In addition, existing visualization solutions generally lack automatic binding with responsible parties and interactive review mechanisms, resulting in a disconnect between encroachment information and responsible parties. This leads to low efficiency in tracing responsibility, broken chains of evidence, and difficulty in meeting the practical needs of rigorous scenarios such as review, land arbitration, or evidence presentation. Summary of the Invention

[0004] This invention provides a method for locating and displaying dynamic change detection results of illegal encroachment. It is a three-dimensional visualization method that integrates terrain data, behavior trajectory, and responsibility information, and supports time backtracking and responsibility linkage. It can not only realistically reflect the spatial evolution characteristics of the encroachment process, but also build an auxiliary display system that is traceable, interactive, and quantifiable, thereby improving the digital support capability of supervision.

[0005] A method for locating and displaying dynamic change detection results of illegal appropriation includes the following steps:

[0006] S1. Dynamic encroachment pattern inversion: Based on continuous time-series remote sensing images of the target area, extract the geometric deformation trajectory and expansion direction of the changed area to generate an encroachment behavior pattern vector;

[0007] S2. Intelligent association of responsible entities: The vector of the trespassing behavior pattern is matched spatiotemporally with a preset ownership database to output a list of responsible entities bound together;

[0008] S3. Location and Display of Encroachment Behavior: Overlay the encroachment behavior pattern vector and the list of responsible entities on the digital elevation model, and render and generate an interactive encroachment behavior location simulation sandbox.

[0009] Optionally, S1 further includes extracting the edge contours of the changing regions in the continuous time-series remote sensing images, superimposing the edge contours of each time window onto the same coordinate system, and generating a spatiotemporal overlay contour map.

[0010] Optionally, the extraction of the geometric deformation trajectory and expansion direction includes performing the following steps on the spatiotemporal overlay contour map:

[0011] Calculate the centroid offset vector of the profiles of adjacent time windows;

[0012] Extract the principal axis deflection angle of the smallest bounding rectangle of the outline.

[0013] Optionally, the encroachment behavior pattern vector is generated based on the centroid offset vector and the principal axis deflection angle, constructed in a time series.

[0014] Optionally, S2 includes performing spatiotemporal feature decoding on the trespassing behavior pattern vector and extracting key features, including:

[0015] Expansion direction angle: Take the weighted average direction of the centroid offset vectors of the last three time windows;

[0016] Behavior starting coordinates: Take the starting coordinates of the first non-zero offset vector;

[0017] Active encroachment period: The time window in which the intensity of change continuously increases.

[0018] Optionally, S2 further includes performing spatial filtering in the ownership database based on the expansion direction angle and the behavior start coordinates. Specifically, it includes constructing a funnel area with the behavior start coordinates as the vertex to expand the range of the expansion direction angle, and filtering a set of candidate plots within the funnel area whose land use is "construction without approval" or "protection red line".

[0019] Optionally, the spatiotemporal matching in S2 specifically includes performing the following for each candidate plot in the candidate plot set:

[0020] Spatial conditions: Calculate the straight-line distance between the candidate plot and the geometric center of the encroached area. Set a dynamic distance threshold based on the plot area. If the straight-line distance is less than the dynamic distance threshold, the candidate plot is considered to be within the main path of illegal expansion in space and meets the spatial matching conditions.

[0021] Time condition: If the overlap between the validity period of the ownership of the candidate land parcel and the period of active encroachment is greater than the overlap threshold, then the candidate land parcel is considered to be under effective supervision during the period when the illegal act occurred, thus meeting the time matching condition;

[0022] The verified candidate land parcel responsible entities are sorted in ascending order by the straight-line distance between the candidate land parcel and the geometric center of the encroached area, generating a list of responsible entities to be bound together.

[0023] Optionally, S3 specifically includes:

[0024] S31, Based on the digital elevation model, extract the surface curvature of the encroached area, and deform the geometric deformation trajectory in the encroachment behavior pattern vector according to the surface curvature to generate a terrain-adaptive trajectory line.

[0025] S32, along the terrain adaptation trajectory line, key frames are equidistantly placed between the starting point of the invasion and the current endpoint. With each key frame as the center, a semi-transparent cube is generated according to the invasion area at that moment, thus completing the construction of the invasion cube.

[0026] S33, A responsible entity information board is hung above the encroachment cube. The responsible entity information board includes the responsible entity from the responsible entity binding list, the ownership number, and the violation clause.

[0027] Optionally, the parameter settings for the encroaching cube include the cube's side length transparency;

[0028] The cube's side length is calculated by taking the square root of the encroached area corresponding to the current keyframe, reflecting the spatial scale of the encroached area at that moment. At the same time, the square root result is adjusted using a side length scaling factor to finally determine the cube's side length.

[0029] Transparency is calculated based on the time difference between the keyframe timestamp and the current system date, determining the time span since the event. The longer the time, the higher the transparency, achieving a visual decay effect of "the newer the time, the clearer it is; the older the time, the more it fades out," highlighting the display weight of recent intrusion behavior.

[0030] Optionally, S3 further includes dynamic interactive responses, specifically including:

[0031] Responding to the click cube action: Play back the invasion process animation along the terrain-adaptive trajectory line;

[0032] Responding to drag-and-drop information panel operations: The 3D scene is rotated to a panoramic view of the plot associated with the responsible person.

[0033] The beneficial effects of this invention are:

[0034] This invention, by introducing digital elevation models and surface curvature calculations, achieves dynamic fitting and spatial correction of encroachment behavior trajectories under three-dimensional terrain conditions. Compared with the traditional method of directly "pressing" the trajectory onto the ground or two-dimensional map, this method uses a surface curvature compensation algorithm to fine-tune the vertical position of the cube, ensuring that the spatial positioning accuracy error of the encroachment trajectory is reduced in complex terrain areas such as slopes and mines. This improves the geometric consistency between the trajectory line and the terrain in the sand table display, and provides high-precision support for spatiotemporal tracing in visualization.

[0035] This invention abstracts encroachment as a cube arranged in a time sequence, mapping the side length of the cube to the square root of the encroached area, thus expressing the intensity of encroachment through spatial volume. Simultaneously, it introduces an algorithm that decreases transparency over time, making recent actions more prominent while historical actions gradually fade out, creating a clear hierarchy and highlighting key visual elements. By embedding minute-level timestamps on the top surface of the cube, it achieves precise traceability of encroachment in the time dimension and visual presentation of evidence time points, enhancing the credibility of the integrated quantitative and evidentiary expression.

[0036] This invention integrates the binding information of responsible parties into a sand table display. A responsibility information board is displayed floating above each newly updated cube, clearly identifying the responsible person, ownership number, and violation clause. Users can refresh the binding list by clicking, and can replay the encroachment process or quickly rotate the view to locate the responsible area. Specifically, when there is multiple responsible parties, a hierarchical display of primary and secondary responsibility boards effectively avoids information obscuring and ambiguity. This mechanism connects the "behavioral trajectory" to "responsibility attribution," providing relevant institutions and natural resource regulatory departments with an operable and reliable digital evidence carrier and visual presentation method. Attached Figure Description

[0037] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0038] Figure 1 This is a schematic diagram of the method flow according to an embodiment of the present invention;

[0039] Figure 2 This is a schematic diagram of the spatiotemporal matching process according to an embodiment of the present invention. Detailed Implementation

[0040] The present invention will now be described in detail with reference to the accompanying drawings and specific embodiments. For some well-known technologies, those skilled in the art may also use other alternative methods to implement the invention. Moreover, the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0041] like Figures 1-2 As shown, a method for locating and displaying dynamic change detection results for illegal appropriation includes the following steps:

[0042] S1. Dynamic encroachment pattern inversion: Based on continuous time-series remote sensing images of the target area, extract the geometric deformation trajectory and expansion direction of the changed area to generate an encroachment behavior pattern vector.

[0043] S11, Temporal Contour Overlay Analysis: Edge extraction is performed on the changing areas in continuous temporal remote sensing images. The edge contours of each time window are uniformly projected onto the same coordinate system to generate a spatiotemporal overlay contour map. The spatiotemporal overlay contour map is used to describe the trend of regional boundary changes over time.

[0044] S12, Deformation Trajectory Quantization: In the spatiotemporal overlay contour map, the following two operations are performed sequentially for each time window:

[0045] S121. Centroid offset vector calculation: For the contours of the same changing region in a remote sensing image across two adjacent time windows (day t and day t+1), extract their geometric centroid coordinates. The centroid offset vector is defined as the vector difference between these two centroid coordinates. ;

[0046] in, For the first The horizontal and vertical coordinates of the centroid of the changing contour within a time window. For the first The centroid offset vector of each window compared to the previous window. Indicates the first The offset of the centroid of each time window relative to the previous window. Indicates the first The offset of the centroid ordinate of each time window relative to the previous window;

[0047] This vector describes the spatial movement trend of the changed area. If the offset vectors of multiple consecutive time windows are in the same direction, it indicates that the changed area is continuously expanding in a certain direction, exhibiting a clear "progressive" nature. In illegal encroachment, this continuous offset direction often points to a specific plot of land (such as a protected area or farmland), which can serve as indirect evidence of "purposeful encroachment." The centroid coordinates themselves are calculated from the average position of pixels within the contour boundary and are commonly used in image time series analysis to describe the evolution trajectory of a region.

[0048] S122. Calculation of principal spindle deflection angle (based on second-order moment):

[0049] ;

[0050] in, Indicates the outline of the changed region about The second central moment of the axis, Indicates the outline of the changed region about The second central moment of the axis, The mixed second central moments of the profile are represented. This indicates the deflection angle of the principal axis of the profile relative to the horizontal axis.

[0051] The principal axis deflection angle is used to characterize the "stretching direction" of a region's shape, such as the extension direction of a dirt road or dumping mark. By monitoring changes in the deflection angle over adjacent time periods, it is possible to identify any "abrupt changes" in behavior. If the angle abrupt change is small (e.g., daily variation <5°), it indicates a stable construction path, usually a legitimate activity; if the angle abrupt change is drastic (e.g., daily variation >15°), it indicates intentional adjustment of direction, possibly a "surprise" illegal encroachment to evade supervision. The second-order moment method is one of the commonly used standard methods in image shape recognition, possessing rotational invariance and statistical stability, and is suitable for contour morphology analysis in remote sensing images.

[0052] S13, Behavioral Pattern Vector Generation: Based on the aforementioned temporal centroid offset vector and principal axis deflection angle, construct a dynamic pattern representation vector for the encroachment behavior. : ;in, This represents the number of effective change time windows, with a vector dimension of [missing information]. Each It contains two components (x / y), each This is a scalar angle value.

[0053] For example: Illegal dumping by construction waste trucks was discovered during monitoring in a river protection area.

[0054] 1. Contour overlay analysis results (S11):

[0055] Day 1 outline center of gravity is principal axis angle (parallel bank);

[0056] The center of gravity of the outline on day 2 is principal axis angle (Suddenly turns towards the river channel);

[0057] 2. Deformation trajectory quantification results (S12):

[0058] The direction points towards the core area of ​​the river channel;

[0059] It was determined to be a surprise attack.

[0060] 3. Behavioral pattern vector generation (S13):

[0061] ; to vector Input S2, and the system will match the permitted construction area of ​​the dump truck company. It will find that the actual area does not match the actual area, and the company will be identified as the responsible party.

[0062] S2. Intelligent association of responsible entities: The vector of the encroachment behavior pattern is matched spatiotemporally with the preset ownership database to output a list of responsible entities bound together.

[0063] S21, Spatiotemporal Feature Decoding: From the intrusion behavior pattern vector Extract the following key features:

[0064] S211, Expansion Direction Angle : Take the centroid offset vector of the last three time windows The weighted average direction (weights based on time proximity):

[0065] ;in, These are weighting coefficients, corresponding to the third, second, and first time windows from the end, respectively. , represents a weighted strengthening of the direction of the latest time window, and arctan is the arctangent function, used to convert vector direction into angle. These are the horizontal components of the centroid offset vector in the third, second, and first time windows from the end, respectively. These are the vertical components of the centroid offset vector in the third, second, and first time windows from the end, respectively.

[0066] This expression is used to calculate the expansion direction angle. This means that the weighted average of the center-of-gravity shift vectors of the encroachment behavior over the three most recent time windows is used to extract the main direction of advancement of the changed area. The weighting factors in the expression... The weights, corresponding to the third-to-last, second-to-last, and first time windows respectively, increase in value, reflecting the emphasis on recent trends. By weighted summing of the lateral (dx) and longitudinal (dy) displacements and using the result as a parameter of the arctangent function, a direction angle relative to the horizontal axis can be obtained. This is used to characterize the overall expansion direction of the entire change process. Illegal encroachment often exhibits clear directional characteristics; for example, dumping of construction waste gradually advances from the road edge towards the core of farmland, or nighttime forced demolition rapidly expands from the boundary to the target building. Therefore, the offset direction of a single time slice may fluctuate or be disturbed. Using the weighted average direction of the three most recent changes can effectively eliminate random errors and more stably restore the dominant directional trend of the behavior. Calculate the expansion direction angle. To construct a legally oriented "funnel area" for subsequent ownership matching, that is, based on the direction angle of expansion. Constructing a spatial selection angle range allows the binding of responsible parties to no longer rely on simple geometric overlap, but rather on spatial evidence of subjective intent to "intentionally point" to the target, which helps to form a more convincing chain of responsibility.

[0067] S212, Beginning coordinates : Refers to the first non-zero offset vector The starting point, i.e., the centroid coordinates of this time window: This indicates that in the invasion behavior pattern vector, the first time window in which spatial change occurs, i.e., the moment when the first centroid offset vector is not zero, is found, and the centroid coordinates of the region corresponding to that moment are taken as the starting coordinates of the behavior. In other words, it is used to mark the location where the illegal encroachment first actually moves or expands. Illegal encroachment often does not break out simultaneously in multiple areas, but rather originates from a single spatial point and gradually expands. Therefore, identifying the first offset coordinate point not only helps to trace the source of the behavior, but also provides a reference starting point for calculating the subsequent expansion direction angle. Firstly, it provides the geometric vertex for constructing a spatial funnel for ownership screening; secondly, by comparing it with the boundaries of land parcels in the ownership database, it helps to determine whether the behavior originated within a specific land parcel, thus helping to infer whether the responsible party "subjectively participated" in initiating the encroachment. This positioning method provides a crucial time-space evidence anchor for subsequent accountability.

[0068] S213, Active Period of Invasion : Refers to the centroid offset magnitude in the behavior pattern vector. The time window of continuous growth: This indicates that in the behavior pattern vector, the magnitudes (i.e., the amplitude of spatial movement) of the centroid offset vector are compared sequentially across different time windows. When these magnitudes show a continuous increasing trend over a certain period, that period is considered an "active encroachment period," denoted as... In other words, it identifies a critical period in which illegal encroachment is gradually escalating and expanding at an increasingly rapid pace. Actual encroachment is usually not instantaneous, but rather evolves from small-scale, tentative intrusions into sustained, ongoing actions, particularly evident in nighttime demolitions and dumping of construction waste. The increasing magnitude of the center of gravity offset reflects the dynamic enhancement of regional deformation, objectively characterizing the "active phase" of the encroachment.

[0069] This provides the main time period of the encroachment for subsequent ownership time matching, thereby determining whether the ownership of the target plot is valid during this period; it also defines the key time nodes in the chain of evidence to avoid misjudgments caused by "occasional behavior" or "short-term sudden changes".

[0070] S22, Ownership Space Funnel Filtering: Based on and Perform the following space filtering operation in the ownership database:

[0071] Starting coordinates As the vertex, construct the subtended angle as An equiangular sector area (ownership screening funnel area);

[0072] Within the funnel-shaped area, select all plots of land marked with "construction without prior approval" or "ecological protection red line". .

[0073] The core idea of ​​S22 above is to construct a "spatial screening funnel" with directional and usage filtering functions to achieve precise initial screening of potentially illegally occupied plots. Specifically, the starting coordinates of the encroachment are used as a spatial anchor point, and a fan-shaped area is constructed within a 10-degree angle to the left and right of the expansion direction angle as the center. This fan-shaped area is the "ownership screening funnel area," representing the potential path range of illegal expansion. Subsequently, the ownership database is searched within this funnel area, selecting only highly sensitive plots with uses of "construction without approval" or "ecological protection red line" as the candidate set. Illegal encroachment often has direction and target, and does not exhibit completely random diffusion. Compared to traditional methods that delineate plots using a fixed radius, the introduction of the "expansion direction angle" feature can more realistically reflect the spatial advancement trend of illegal activities, thereby accurately locking down potentially victimized plots with a "direction" in spatial scope. Meanwhile, land use, as a crucial attribute for determining the sensitivity of infringement, can be filtered through preset use tags to effectively exclude general construction land or areas with approved construction permits, thereby improving the targeting and judicial applicability of the identification. Firstly, this significantly narrows the matching range, reduces false matches, and improves binding efficiency; secondly, it enhances the spatial orientation of liability determination, making the binding results more convincing; and thirdly, it provides a high-quality candidate land parcel foundation for subsequent spatiotemporal validity verification, improving the accuracy of the entire identification chain.

[0074] S23, Spatiotemporal validity verification: For each candidate land parcel Perform the following verification:

[0075] S231, Spatial Condition: Calculate the distance between its geometric center and the center of the currently occupied area. ,satisfy:

[0076] ;in, Indicates land parcel The European-style distance from the center of the occupied area Indicates land parcel area, This represents an empirically set distance ratio coefficient. Based on the expansion path derived from the encroachment behavior pattern vector and the geometric center of the current changed area, the Euclidean distance to the center of each candidate plot is calculated. A dynamic distance threshold is then set, taking into account the plot's area. Only when the plot distance is less than this threshold is the plot considered spatially related to the encroachment behavior. This ensures the spatial rationality of binding the responsible party and reflects the requirements of the "spatial matching" section.

[0077] S232, Time Condition: Determine the land parcel The validity period of ownership and the active period of encroachment The degree of time overlap, calculating the number of overlapping days. Requirements must be met: The active encroachment period, i.e. the time period during which the changing behavior is continuously enhanced, is extracted from the behavioral pattern vector and cross-compared with the validity period of the ownership recorded in the ownership database of the candidate land parcel. Only when there is a large degree of time overlap between the two and the overlap ratio exceeds a set threshold (70%), is the land parcel determined to be within the validity period of the encroachment behavior in the time dimension. This process achieves the requirement of "time matching" and further improves the timeliness and credibility of the binding.

[0078] Ultimately, only for plots that simultaneously meet both spatial and temporal conditions will the information of the responsible parties be extracted and compiled into a "list of responsible entities."

[0079] S24, Binding of Responsible Parties: For land parcels that have passed spatiotemporal verification, extract the information of their responsible parties and bind them according to... Sort by size from smallest to largest to generate a list of responsible parties to be bound together:

[0080] ;

[0081] in, This represents the sorted list of responsible parties, including the responsible person, distance, and overlapping time information. Indicates land parcel The legally responsible person (from the ownership database), Indicates the distance between the land parcel and the center of the encroached area. This indicates the number of days that overlap between the land ownership validity period and the active encroachment period. This list will be used in the sand table display module in S3, where the primary responsible party... The smallest (smallest) information will be highlighted first.

[0082] The ownership database is the core data support for binding the responsible parties of illegal encroachment in this invention. It is essentially a structured collection of land information, recording key information such as the spatial location, land use, ownership time, and identity of the responsible party for all land parcels within a specific area. The design of this database revolves around three judicial adaptation principles: "clear spatial location, traceable ownership responsibility, and identifiable use attributes." This ensures that its data content can be accurately compared with the geometric trajectory, expansion direction, and active time period of the encroachment, thereby achieving automatic matching and binding of the responsible parties.

[0083] First, at the spatial level, each plot in the database records complete boundary coordinates (usually stored in polygon form) and geometric center coordinates. This allows the system to construct a spatial filtering "funnel" after obtaining the expansion direction angle and the starting point of the action, filtering out suspicious plots that overlap with it. Simultaneously, the database also records the area information of each plot, used to set a threshold for spatial distance matching, ensuring that the binding results have spatial legal credibility.

[0084] Secondly, at the land use attribute level, the land use field in the database can identify the legal use nature of the land parcel, such as "construction without prior approval," "ecological protection red line," and "general construction land." This allows the system to further filter out legal construction areas based on spatial overlap, retaining only target land parcels with potential illegal risks.

[0085] Secondly, at the temporal level, the ownership database clearly records the start and end dates of ownership for each plot of land. The system calculates the overlap of these dates with the active periods of encroachment to determine whether the plot was within the effective regulatory period when the illegal act occurred. Only plots that spatially overlap and have a temporal overlap rate exceeding a set threshold (e.g., 70%) will be included in the scope of the responsible party's binding.

[0086] Finally, at the level of subject identification, the database provides the name of the responsible person and their unique identifier (such as ID card number or unified social credit code) for each plot of land, which enables the direct generation of a list of responsible persons that can be cited in judicial proceedings after screening, and supports information exchange with the administrative penalty system or natural resource supervision platform.

[0087] In summary, this ownership database not only provides complete land parcel spatial boundaries, but also integrates multi-dimensional information such as land use, ownership time, and the identity of the responsible party. It can accurately verify the correlation of illegal encroachment at the spatial, temporal, and behavioral levels, and serves as the basic support for achieving intelligent binding of responsible parties.

[0088] Table 1. Schematic diagram of the ownership database structure

[0089]

[0090] S3. Location and Display of Encroachment Behavior: Overlay the encroachment behavior pattern vector and the list of responsible parties onto the digital elevation model, and render an interactive encroachment behavior location simulation sandbox. In the sandbox:

[0091] The occupied area is marked with its volume and timestamp using a semi-transparent red cube.

[0092] The associated responsible entity information is displayed floating above the corresponding cube.

[0093] S3 specifically includes:

[0094] S31, Topographic Coupled Modeling: Extracting the surface curvature of the encroached area based on the digital elevation model (DEM). The geometric deformation trajectory in the encroachment behavior pattern vector is adapted to the surface curvature to generate a terrain-adaptive trajectory line, which serves as the basic path for subsequent 3D spatial construction. Specifically:

[0095] S311 uses digital elevation model (DEM) raster data in GeoTIFF format, with the following data structure:

[0096] Lateral longitude sampling points ;

[0097] Longitudinal latitude sampling points ;

[0098] Elevation value of each point .

[0099] S312, Curvature Definition and Calculation: The Gaussian curvature and principal curvature approximation model are used, and calculations are performed based on local second derivatives. Assume a continuous elevation function is already obtained. The curvature of the Earth's surface is defined as:

[0100] ;in, , Represents the first-order partial derivative. It represents the second-order partial derivative; it can be quickly calculated in DEM raster images using a third-order Laplacian convolution template.

[0101] S313, Regional-level processing flow: For each point of encroachment behavior trajectory... around it or Elevation is extracted within a pixel window; the curvature value of the corresponding point is calculated by approximating the derivative using the second-order finite difference method. Output the surface curvature sequence corresponding to the trajectory points: .

[0102] The specific terrain-fitting deformation of the geometric deformation trajectory is as follows:

[0103] S314, Original Trajectory Definition: A set of three-dimensional trajectory points constituted by keyframes in the trespass behavior pattern vector:

[0104] ;in, The horizontal coordinate is... This is the default ground height or the default value for the 2D trajectory.

[0105] S315, Terrain Fitting: Each point is vertically offset according to its corresponding terrain height in the DEM, and a curvature compensation term is introduced. ;in: , This represents the original ground elevation extracted from the DEM. Indicates the first Each trajectory point corresponds to a surface curvature. The average radius of the Earth This indicates the amount of cubic subsidence compensation caused by surface curvature (used to improve the fitting accuracy of steep slopes).

[0106] S315, finally yielding a set of terrain-adapted trajectory line points. Use this trajectory line as a spatial path reference for constructing the encroaching cube.

[0107] S32, Constructing the Invasion Cube: Along the terrain adaptation trajectory line, keyframe nodes are equidistantly placed between the invasion start point and the current endpoint;

[0108] A visualization cube is constructed centered on each keyframe, with the following parameters:

[0109] Formula for calculating the side length of a cube: ;in, Let be the side length of the cube. This represents the encroached area at the corresponding moment of the keyframe. This is the scaling factor for the side length, empirically set to... ;

[0110] Transparency calculation formula: ;in, The cube's transparency (range: 0~1). The timestamp of the current keyframe. The above formula represents the current system date and indicates that the event will gradually fade out over time, reflecting its age. A timestamp label is embedded in the center of the top face of each cube, in the format of "YYYY-MM-DDHH:MM", which is used to accurately express the time of occurrence.

[0111] The relationship between area and side length is a square root relationship; if we directly take... Using a side length of 0.7 would make the cube appear too large in large areas, hindering visual distinction and scene layout. 0.7 is a reasonably balanced scaling factor in practice, balancing the expressive power of the cube's volume with the overall compactness of the scene, preventing the cube from obscuring surrounding information. In actual testing... 1000m When taking an interval, take The cube volume at that time is relatively consistent with the actual perceived scale of encroachment, possessing good visual representation and legal evidentiary value. Therefore, 0.7 is an empirical value selected after engineering calibration and visual verification.

[0112] S33, Floating Responsible Entity Binding: A responsible entity information board is suspended directly above the newly constructed cube;

[0113] The information board content includes:

[0114] Name of the responsible person or organization (from the list of responsible entities bound by S2);

[0115] Land or mineral rights number and corresponding legal violation clauses;

[0116] A clickable real-time update button supports dynamic refreshing of the responsible entity's information.

[0117] S34, Click the cube to trigger the animation of the encroachment behavior to be played back in the order of the terrain-adaptive trajectory lines, which is used to show the encroachment and expansion process in space and time;

[0118] Drag and drop information boards: Link control scene view rotation, automatically locate to the panoramic view of the area corresponding to the responsible person, realize spatial visual traceability of the main responsible party.

[0119] The specific scheme for equidistantly deploying keyframe nodes along the terrain-adaptive trajectory line between the intrusion starting point and the current endpoint is as follows:

[0120] Trajectory data source: Input is the set of terrain-adapted trajectory lines and points generated in the previous step.

[0121] Each point has three-dimensional coordinates. This reflects the terrain conformation pattern after the encroachment trajectory has evolved over time.

[0122] Calculate the total arc length of the trajectory: Traverse the entire point set, accumulate the Euclidean distance segment by segment to obtain the total trajectory length. :

[0123] The distance is calculated using three-dimensional Euclidean distance to ensure the true length along the spatial curve.

[0124] Set equal step size and resampling: Set the desired spatial spacing between keyframes. (e.g., 5 meters or 10 meters), then the number of nodes to be sampled can be calculated. ; Proceed sequentially from the starting point according to the total arc length, until the cumulative length reaches Insert a new keyframe node at that location. This indicates rounding down to the nearest integer.

[0125] Interpolation keyframe position (for non-integer points falling within a segment): when the cumulative arc length falls exactly between two points. In between:

[0126] Use linear interpolation or spline interpolation to generate new points within the segment proportionally as keyframes:

[0127] ;in, This indicates the keyframe node position (3D coordinates) obtained by interpolation. This indicates the starting trajectory node (3D coordinates) of the current interpolation segment. This represents the endpoint trajectory node (3D coordinates) of the current interpolation segment. This represents the interpolation scaling factor, indicating the relative position (0~1) of the target point within this segment. , This indicates the current target arc length position (i.e., the cumulative length of the trajectory corresponding to the keyframe to be inserted). This indicates the cumulative arc length starting from the current segment. This represents the cumulative arc length at the end of the current trajectory segment.

[0128] The final output is a new set of keyframes: Each The spatial position is obtained by uniformly sampling along the original terrain adaptation trajectory line and used as the center point of the subsequent cube.

[0129] Isometric keyframe placement ensures consistent display of subsequent cube voxelization, avoiding dense foregrounds and sparse backgrounds or visual jumps; it can realistically reflect the uniform spatial progression of encroachment behavior, facilitating subsequent animation playback and responsibility association; it supports subsequent interpolation smoothing or uniform time tag distribution, improving the stability and readability of the 3D sandbox interaction.

[0130] This invention encompasses any substitutions, modifications, equivalent methods, and solutions made within the spirit and scope of this invention. To provide the public with a thorough understanding of this invention, specific details are described in detail in the following preferred embodiments; however, those skilled in the art will fully understand the invention even without these details. Furthermore, to avoid unnecessary misunderstanding of the essence of this invention, well-known methods, processes, procedures, components, and circuits are not described in detail.

[0131] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for locating and displaying dynamic change detection results of illegal encroachment, characterized in that, Includes the following steps: S1. Based on continuous temporal remote sensing images of the target area, extract the geometric deformation trajectory and expansion direction of the changed area to generate an encroachment behavior pattern vector; S1 further includes extracting the edge contours of the changing regions in the continuous time-series remote sensing images, and superimposing the edge contours of each time window onto the same coordinate system to generate a spatiotemporal superimposed contour map. The extraction of the geometric deformation trajectory and expansion direction is performed in the spatiotemporal overlay contour map: Calculate the centroid offset vector of the profiles of adjacent time windows; Extract the principal axis deflection angle of the smallest bounding rectangle of the contour; The encroachment behavior pattern vector is constructed and generated based on the centroid offset vector and the principal axis deflection angle, according to a time series. S2. Perform spatiotemporal matching of the encroachment behavior pattern vector with a preset ownership database, and output a list of responsible entities bound together; S3. Overlay the encroachment behavior pattern vector and the list of responsible entities on the digital elevation model, and render and generate an interactive encroachment behavior location simulation sandbox; S3 specifically includes: S31, Based on the digital elevation model, extract the surface curvature of the encroached area, and deform the geometric deformation trajectory in the encroachment behavior pattern vector according to the surface curvature to generate a terrain-adaptive trajectory line. S32, along the terrain adaptation trajectory line, key frames are equidistantly placed between the starting point of the invasion and the current endpoint. With each key frame as the center, a semi-transparent cube is generated according to the invasion area at that moment, thus completing the construction of the invasion cube. S33, A responsible entity information board is hung above the encroachment cube. The responsible entity information board includes the responsible entity from the responsible entity binding list, the ownership number, and the violation clause.

2. The method for locating and displaying dynamic change detection results for illegal encroachment as described in claim 1, characterized in that, S2 includes performing spatiotemporal feature decoding on the encroachment behavior pattern vector and extracting key features, which include: Expansion direction angle: Take the weighted average direction of the centroid offset vectors of the last three time windows; Behavior starting coordinates: Take the starting coordinates of the first non-zero offset vector; Active encroachment period: The time window in which the intensity of change continuously increases.

3. The method for locating and displaying dynamic change detection results for illegal encroachment as described in claim 2, characterized in that, The S2 further includes performing spatial filtering in the ownership database based on the expansion direction angle and the behavior start coordinates. Specifically, it includes constructing a funnel area with the behavior start coordinates as the vertex to expand the range of the expansion direction angle, and filtering a set of candidate plots within the funnel area whose land use is "construction without approval" or "protection red line".

4. The method for locating and displaying dynamic change detection results for illegal encroachment as described in claim 3, characterized in that, The spatiotemporal matching in S2 specifically includes performing the following on each candidate plot in the candidate plot set: Spatial conditions: Calculate the straight-line distance between the candidate plot and the geometric center of the encroached area. Set a dynamic distance threshold based on the plot area. If the straight-line distance is less than the dynamic distance threshold, the candidate plot is considered to be within the main path of illegal expansion in space and meets the spatial matching conditions. Time condition: If the overlap between the validity period of the ownership of the candidate land parcel and the period of active encroachment is greater than the overlap threshold, then the candidate land parcel is considered to be under effective supervision during the period when the illegal act occurred, thus meeting the time matching condition; The verified candidate land parcel responsible entities are sorted in ascending order by the straight-line distance between the candidate land parcel and the geometric center of the encroached area, generating a list of responsible entities to be bound together.

5. The method for locating and displaying dynamic change detection results for illegal encroachment as described in claim 1, characterized in that, The parameters for invading the cube include the cube's side length and transparency; The cube's side length is calculated by taking the square root of the encroached area corresponding to the current keyframe, reflecting the spatial scale of the encroached area at that moment. At the same time, the square root result is adjusted using a side length scaling factor to finally determine the cube's side length. Transparency is calculated based on the time difference between the keyframe timestamp and the current system date, determining the time span since the event. The longer the time, the higher the transparency, achieving a visual decay effect of "newer events are clearer, older events fade out," highlighting the display weight of recent intrusion behaviors.

6. The method for locating and displaying dynamic change detection results for illegal encroachment as described in claim 5, characterized in that, The S3 also includes dynamic interactive responses, specifically including: Responding to the click cube action: Play back the invasion process animation along the terrain-adaptive trajectory line; Responding to drag-and-drop information panel operations: The 3D scene is rotated to a panoramic view of the plot associated with the responsible person.

Images