A method for detecting changes in geographic information mapping based on spatiotemporal data

Abstract

The application discloses a kind of geographic information surveying and mapping change detection methods based on space-time data, it is related to geographic information system technical field, including: obtaining historical object version, current surveying and mapping result and collection batch data, unified coordinate and time reference, split object comparison fragment and establish space-time index;With no change registration and the reference point of coordinate residual difference not exceeding allowable value is corrected historical object comparison fragment, generates fragment corresponding relationship;According to the change vector difference of collection batch edge two sides, length, direction and continuous change range, record collection batch position difference, or generate geographic object change candidate record and feedback measurement task record;Compare feedback change vector, current cycle change vector and surveying and mapping allowable error, determine change result and update object geometry, version, valid time, transaction time and space-time index.This method can distinguish collection batch position difference and geographic object change, and keep the association of change result and object version and space-time index.

Description

Technical Field

[0001] This invention relates to the field of geographic information system technology, and in particular to a geographic information mapping change detection method based on spatiotemporal data. Background Technology

[0002] Geographic information mapping change detection falls under the field of geographic information system data management and spatial analysis, and is commonly used in land surveys, urban renewal, road and pipeline maintenance, and natural resource monitoring. Conventional methods typically convert vector mapping results from different periods to a unified coordinate reference system, establish object correspondences based on object identifiers, spatial locations, geometric overlap relationships, and topological relationships, and then identify the addition, disappearance, movement, or boundary changes of geographic objects through point distances, line offsets, areal overlay differences, and attribute changes, and store the detection results in a geographic information database.

[0003] However, existing methods still have two limitations: first, it is difficult to distinguish between local coordinate deviations at the junctions of data collection batches and changes in object boundaries; second, the granularity of the association between change records and object versions, validity periods, and spatial indexes is limited. Existing technologies mainly suffer from misjudgment of junction changes, accompanied by insufficient association with version indexes. This invention belongs to the field of spatiotemporal data processing in geographic information systems, and solves the problem of misjudgment of changes caused by local deviations across batches. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a geographic information mapping change detection method based on spatiotemporal data, which solves the problem of misjudgment of geographic object boundary changes caused by local coordinate deviations at the junction of cross-collection batches.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This invention provides a method for detecting changes in geographic information mapping based on spatiotemporal data. The method includes: acquiring historical geographic information object version data, current period mapping result data, batch data, geographic object change registration data, and mapping quality control data; unifying the coordinate reference system and time benchmark; segmenting current geographic objects and historical geographic objects according to the coverage of the batch data; generating object comparison segment records and spatiotemporal indexes; based on the object comparison segment records and spatiotemporal indexes, filtering reference points with no change registrations and coordinate residuals not exceeding allowable values; calculating position correction amounts based on the difference between the current and historical coordinates of the reference points; correcting historical object comparison segments; and generating a correspondence between current and historical object comparison segments. Through correction... The system compares historical object fragments and their corresponding relationships. At the edge location of the data collection batch, it calculates the change vector of the current boundary point relative to the corrected historical boundary point. Based on the difference in change vectors on both sides of the edge, the length of the change vector, the direction of the vector, and the range of continuous change, it records the location differences of the data collection batch, or generates candidate records of geographic object changes and feedback measurement task records. Based on the candidate records of geographic object changes and feedback measurement task records, it calculates the feedback change vector. It compares the feedback change vector with the current periodic change vector and the corresponding allowable surveying error, and generates confirmed change, unconfirmed change, or pending verification results. Based on the confirmed change results, it generates the formal geographic object geometry and updates the geographic object version, validity period, transaction time, spatial index, and time index.

[0007] As a preferred embodiment of the geographic information mapping change detection method based on spatiotemporal data described in this invention, the unified coordinate reference system and time base, the current geographic object and historical geographic object are divided according to the coverage of the collection batch, and the object comparison segment record and spatiotemporal index are generated by: converting the coordinates of the version data of historical geographic information objects and the current periodic mapping results data to the target coordinate reference system, converting the collection time to the unified time base, deleting repeated continuous nodes in the line object, checking the closure state, self-intersection state and ring direction of the outer and inner rings of the surface object, and intersecting the geometry of the current geographic object and the geometry of the historical geographic object that have passed the geometric check with the coverage of the collection batch, respectively, and writing the object identifier, version number, collection batch identifier, original object boundary mark and collection batch segmentation boundary mark into the object comparison segment obtained by the intersection.

[0008] As a preferred embodiment of the geographic information mapping change detection method based on spatiotemporal data described in this invention, the generation of object comparison fragment records and spatiotemporal indexes includes: filtering historical object versions from the query results whose object version confirmation status is confirmed as valid, whose valid start time is no later than the current collection time, and whose valid end time is empty or later than the current collection time; when only one historical object version exists in the filtering results, generating a candidate relationship between the current object comparison fragment and the historical object version; when two or more historical object versions exist in the filtering results, generating a historical object version time conflict record, and determining the historical object candidate version; when no confirmed valid version has a valid time range covering the current collection time, sorting confirmed valid historical object versions earlier than the current collection time in ascending order according to the time interval between the current collection time and the valid end time of each historical object version; when the minimum time interval corresponds to only one historical object version, reading the first-ranked historical object version and recording the time interval; when the minimum time interval corresponds to two or more historical object versions, generating a historical object version time conflict record; when only historical object versions later than the current collection time are found, generating a historical object version missing record.

[0009] As a preferred embodiment of the geographic information mapping change detection method based on spatiotemporal data described in this invention, the following steps are included: The selection of reference points without change registration and whose coordinate residuals do not exceed the allowable value includes: extracting control points or geographic feature points with the same identifier in the historical and current periods from the same collection batch; deleting points that have been migrated, demolished, rebuilt, or whose coordinates have been remeasured within the collection time interval; for each geographic feature point, temporarily deleting the geographic feature point, calculating the current predicted coordinates of the geographic feature point using the inverse distance weighting of the coordinate differences of reference points other than the geographic feature point, and deleting geographic feature points whose distance between the current predicted coordinates and the current measured coordinates exceeds the allowable value of the position correction residual; when the number of retained points reaches the minimum number of reference points and the convex hull formed by the retained points covers the position correction calculation position, the retained points are written into the reference point record.

[0010] As a preferred embodiment of the geographic information mapping change detection method based on spatiotemporal data described in this invention, the step of calculating the position correction amount based on the difference between the current coordinates and historical coordinates of the reference point and correcting the historical object comparison segment includes: calculating the planar coordinate difference between the current coordinates and historical coordinates of each reference point; calculating the planar distance from the historical point object coordinates, historical line object nodes, or historical surface object boundary sampling points to each reference point; by outputting the position correction amount of the corresponding coordinates, nodes, or boundary sampling points, translating the historical object comparison segment according to the position correction amount, and storing the translated historical object comparison segment in a temporary spatial data table and an R-tree spatial index.

[0011] As a preferred embodiment of the geographic information mapping change detection method based on spatiotemporal data described in this invention, the step of generating the correspondence between the current object comparison segment and the historical object comparison segment includes: reading the current and historical allowable plane errors in the mapping quality control data, the maximum value of the coordinate transformation residual in the coordinate transformation check record, and the maximum remaining plane distance in the segment position correction record; adding the current allowable plane error, the historical allowable plane error, the maximum value of the coordinate transformation residual, and the maximum remaining plane distance to obtain the object candidate correspondence retrieval extension distance; and generating a correspondence based on the minimum bounding rectangle of the current object comparison segment by expanding outward according to the object candidate correspondence retrieval extension distance. For each candidate object corresponding to a search range, the historical object comparison segments that intersect with the modified geometric data and the candidate object corresponding to the search range are queried using the modified geometric R-tree spatial index. From the query results, historical object comparison segments with the same object classification code, the same geometric type, and the same set of classification codes for connected objects or shared edge objects are retained. A one-to-one relationship is recorded when one current object comparison segment corresponds to one historical object comparison segment; a one-to-many relationship is recorded when one current object comparison segment corresponds to two or more historical object comparison segments; a many-to-one relationship is recorded when two or more current object comparison segments correspond to one historical object comparison segment; and a many-to-many relationship is recorded when two or more current object comparison segments correspond to two or more historical object comparison segments.

[0012] As a preferred embodiment of the geographic information mapping change detection method based on spatiotemporal data described in this invention, the step of calculating the change vector of the current boundary point relative to the corrected historical boundary point at the edge position of the collection batch includes: for two collection batches whose coverage boundaries intersect and are associated with the same current temporary object, extracting object comparison segments with the same historical geographic object identifier, historical version number, boundary ring identifier, and boundary node order; outputting the change vector difference at the edge position of the collection batch; for surface objects, extracting boundary sampling points from the intersection of the original outer ring or inner ring with the coverage boundary of the collection batch; and for line objects, using the segmentation endpoints of the collection batch as boundary sampling points instead of as the actual endpoints of the line objects.

[0013] As a preferred embodiment of the geographic information mapping change detection method based on spatiotemporal data described in this invention, the step of recording the location differences of the collection batches or generating candidate records of geographic object changes and feedback measurement task records based on the difference in change vectors on both sides of the edge, the length of the change vectors, the vector direction, and the continuous change range includes: for line object comparison segments or area object comparison segments formed by different collection batches and having edge relationships between collection batches, adding the historical original object boundary sampling position coordinates to the corresponding position correction vector, and then subtracting the corrected historical original object boundary sampling position coordinates from the current original object boundary corresponding position coordinates to obtain the change vector of the boundary sampling position; the difference in change vectors at the edge positions of the collection batches is not greater than the edge allowable value, and the length of the change vectors on both sides of the edge is... When the difference in the position of the data collection batch is not greater than the sum of the allowable planar error of the historical object version and the allowable planar error of the corresponding current period survey result, the location difference of the data collection batch is recorded; when the difference in the change vector of the edge position of the data collection batch is not greater than the allowable edge value, the length of the change vector on both sides of the edge is greater than the sum of the corresponding errors, the projections of the two change vectors on the historical boundary normal direction have the same sign, and the cumulative length of the continuous change boundary segment of the line object is not less than the minimum representable length or the area of ​​the change region of the surface object is not less than the minimum representable area, a candidate record of geographic object change is generated; when the difference in the change vector is greater than the allowable edge value, the length of the change vector on only one side of the edge is greater than the sum of the corresponding errors, the projections of the two change vectors on the historical boundary normal direction have opposite signs, or the object segment maintains a many-to-many relationship, a feedback measurement task record is generated.

[0014] As a preferred embodiment of the geographic information mapping change detection method based on spatiotemporal data described in this invention, the step of calculating the feedback change vector based on candidate records of geographic object changes and feedback measurement task records, and comparing the feedback change vector with the current periodic change vector and the corresponding mapping allowable error to generate a confirmed change, an unconfirmed change, or a result pending verification includes: collecting feedback boundary coordinates and feedback reference point coordinates based on the feedback measurement task records, calculating the plane coordinate difference between the feedback coordinates of the feedback reference point and the historical coordinates, calculating the feedback position correction using a distance inverse weighting method, and subtracting the sum of the historical boundary coordinates and the feedback position correction from the feedback boundary coordinates to obtain the feedback change vector; the difference length between the feedback change vector and the current periodic change vector is not large. If the sum of the allowable planar error of the current period's surveying results and the allowable planar error of the feedback measurement, the length of the feedback change vector is greater than the sum of the allowable planar error of the historical object version and the allowable planar error of the feedback measurement, and the dot product of the two change vectors of the point object is greater than zero, or the projections of the two change vectors of the line or surface object onto the historical boundary normal direction have the same sign, then record and confirm the change. If the length of the difference between the feedback change vector and the change vector of the current period is not greater than the sum of the previous errors, and the length of the feedback change vector is not greater than the sum of the allowable planar error of the historical object version and the allowable planar error of the feedback measurement, then record and do not confirm the change. If the length of the difference between the feedback change vector and the change vector of the current period is greater than the sum of the previous errors, or the direction of the change vector does not meet the corresponding direction condition, then record and wait for verification.

[0015] As a preferred embodiment of the geographic information mapping change detection method based on spatiotemporal data described in this invention, the step of generating formal geographic object geometry based on the confirmed change result and updating the geographic object version, validity time, transaction time, spatial index, and time index includes: for records whose change determination status is confirmed change and associated with valid feedback measurement data, when the allowable plane error of the feedback measurement is less than the allowable plane error of the current period's mapping results, the feedback boundary coordinates are used; when the allowable plane error of the feedback measurement is not less than the allowable plane error of the current period's mapping results, the coordinates of the current period's mapping results are used; point object coordinates are generated, or a sequence of line object nodes or a ring of polygon object boundaries is generated according to the order of historical object nodes; when two polygon objects sharing a line segment with the change boundary both record confirmed changes, the versions of the two polygon objects are updated in the same database transaction; when the object position, boundary, or attribute changes, the geographic object identifier is retained and a version number is added; when one historical object corresponds to two or more current objects, the object segmentation relationship is recorded; when two or more historical objects correspond to one current object, the object merging relationship is recorded; the change occurrence time is written into the validity time field, the database write time is written into the transaction time field, and the spatial index and time index are updated according to the formal geographic object geometry and validity time.

[0016] The beneficial effects of this invention are as follows: by collecting batch object location correction records and object segment correspondences, the change vector difference between the two sides of the edge of the collection batch is calculated, and candidate records of geographic object changes are generated by combining vector length, direction and continuous change range, so as to classify the location differences of the collection batch and the changes of object boundaries and reduce the misjudgment of edge location changes; by generating confirmed change results and formal geographic object geometry through feedback measurement task records, the geographic object version, effective time, transaction time, spatial index and time index are updated synchronously, so as to realize the corresponding storage of change results and object spatiotemporal versions, and achieve the effect of refining version index association. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a geographic information mapping change detection method based on spatiotemporal data.

[0019] Figure 2 A flowchart for generating spatiotemporal baseline records for geographic information mapping, comparison fragment records of collected batches of objects, and spatiotemporal indexes for geographic information.

[0020] Figure 3 A flowchart is generated for stabilizing reference point records, fragment position correction records, and initial correspondence records for object fragments.

[0021] Figure 4 This document outlines the process for collecting batch boundary difference determination records, geographic object change candidate records, and confirmation of change results.

[0022] Figure 5 To collect the error rate of change judgment and the root mean square error curve of change position under the superposition of batch position disturbance and actual change.

[0023] Figure 6 A graph showing the average response time for spatiotemporal version queries under different numbers of geographic object version records. Detailed Implementation

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0025] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0026] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0027] Reference Figures 1-6 This is one embodiment of the present invention, which provides a geographic information mapping change detection method based on spatiotemporal data, comprising the following steps: S1. Obtain historical geographic information object version data, current period surveying and mapping result data, collection batch data, geographic object change registration data and surveying and mapping quality control data, unify coordinate reference system and time benchmark, divide current geographic objects and historical geographic objects according to the coverage of collection batch, and generate object comparison fragment records and spatiotemporal indexes.

[0028] The process involves acquiring historical geographic information object version data, current periodic surveying and mapping result data, collection batch data, geographic object change registration data, surveying and mapping quality control data, and auxiliary surveying and mapping data for the target surveying and mapping area. Source registration, coordinate transformation, time normalization, and geometric checks are performed on all types of data. Current surveying and mapping objects and historical object candidate versions are organized according to geographic grids, collection batch coverage, and object version validity time. Current surveying and mapping objects spanning multiple collection batches are divided into object comparison segments, generating a geographic information surveying and mapping spatiotemporal basic record, collection batch object comparison segment record, and geographic information spatiotemporal index. The historical object candidate version retrieval extension distance is determined for each object comparison segment based on object type code and mapping scale. The process reads the current permissible plane error registered in the surveying and mapping task quality control file, the historical permissible plane error registered in the corresponding historical object version's result quality record, and the plane residuals of each common control point or independent check point in the coordinate transformation check record. The maximum plane residual is used as the maximum coordinate transformation residual. The current permissible plane error, the historical permissible plane error, and the maximum coordinate transformation residual are added together to obtain the historical object candidate version retrieval extension distance. When the source and target coordinate reference systems are identical and no coordinate transformation has been performed, the maximum value of the coordinate transformation residual is recorded as zero. If the calculated result is less than the minimum coordinate storage resolution of the target coordinate reference system, the minimum coordinate storage resolution is used as the extended distance for historical object candidate version retrieval. If the maximum value of the coordinate transformation residual exceeds the allowable value, or if any error data is missing, historical object candidate version retrieval data is generated to be supplemented, and the corresponding object comparison segment is not automatically queried.

[0029] The target mapping area is determined based on the task boundary registered in the mapping task file. The task boundary is represented by closed surface geometric data under the target coordinate reference system, and the mapping task number, task sub-region number, administrative division code, mapping scale, planned data acquisition time range, geographic grid version number, and result reception time are recorded. When a mapping task contains more than two spatially discontinuous areas, task sub-region numbers are generated separately, while retaining the same mapping task number. If the task boundary is not closed, self-intersecting, or has duplicate boundary lines, a task boundary anomaly record is generated, and the corresponding area is not included in the current cycle's mapping object retrieval.

[0030] Historical geographic information object version data is read from the geographic information system spatial database, recording the geographic object number, object type code, point, line, and polygon spatial geometric data, attribute data, topological relationships, object version number, effective start time, effective end time, data write time, data source number, and version confirmation status. The object type code is obtained according to the geographic feature classification and coding rules adopted by the geographic information system. The version confirmation status is divided into confirmed valid, transitional, pending verification, and discontinued. Object versions with a confirmed valid status participate in the historical object candidate version query, while other object versions retain their original status records.

[0031] The current period's surveying and mapping results use vector geographic objects that have completed mapping processing as spatial comparison objects, recording the current surveying object's temporary number, object type code, spatial geometric data, surveying attribute data, source file number, source element number, result quality status, and source acquisition batch number. When the current surveying object does not have a corresponding relationship with existing geographic objects, a temporary number for the current surveying object is generated based on the source acquisition batch number, source file number, and source element number; if it has inherited the number of an existing geographic object, both the existing geographic object number and the current surveying object's temporary number are saved. The former is used for historical version queries, while the latter identifies the current object in the current surveying and mapping results.

[0032] Geographic object change registration data records the change registration number, geographic object number, change type, change occurrence time or time range, change spatial range, registration source, and registration status. Surveying and mapping quality control data records historical permissible plane error, current permissible plane error, permissible coordinate transformation residual value, minimum representable length, minimum representable area, boundary sampling interval, minimum number of reference points, minimum number of reference points for verification, and permissible position correction residual value. All values ​​are registered according to the surveying and mapping task technical design documents, map scale, and result accuracy level.

[0033] For each current surveying object, a source relationship record is generated, including the current surveying object's temporary number, source file number, source element number, source acquisition batch number, source geometric range, and source relationship status. When the current surveying object is generated from a single acquisition batch file, the acquisition batch number is directly registered; when it is formed by merging two or more acquisition batches, the valid mapping range of each source file is recorded separately; when acquisition batch information is missing, a batch attribution candidate record is generated based on the source file, actual acquisition time, and spatial coverage. If two or more candidate acquisition batches meet the attribution criteria and cannot be excluded by source file or actual acquisition time, the batch attribution status is marked as pending verification.

[0034] Auxiliary mapping data utilizes orthophotos, oblique photogrammetry results, laser point clouds, GPS measurement records, real-time dynamic differential positioning measurement records, and control point measurement records. Image and point cloud data records include data number, spatial coverage, acquisition time, spatial resolution or point cloud density, measurement accuracy, quality status, and storage location. Measurement records include measurement point number, measurement time, plane coordinates, elevation, solution status, measurement accuracy, and control network number. Auxiliary mapping data is associated with the current mapping object according to the source acquisition batch and spatial intersection range, and does not directly replace the spatial geometric data of the object comparison segment.

[0035] The batch number registered in the surveying operation record is used as the primary data acquisition batch. If the operation record does not register a batch number, the initial batch is divided according to the number of flights, the time period of a single mobile surveying operation, or the time period of a single on-site surveying operation. When the combination of surveying equipment, control network number, coordinate calculation parameters, or result quality inspection record changes within the initial batch, a new acquisition batch number is generated. The acquisition batch data is written with the acquisition batch number, task sub-region number, surveying equipment number, control network number, actual acquisition start and end time, acquisition coverage area, source coordinate reference system, allowable plane error, allowable elevation error, ground resolution, result quality status, and source file number.

[0036] The acquisition coverage area of ​​UAV mapping data is formed by merging the image projection areas that have passed quality inspection; the acquisition coverage area of ​​mobile measurement data is formed based on the travel trajectory, sensor observation direction, and measurement distance registered in the mission file; the acquisition coverage area of ​​GNSS measurement, real-time dynamic differential positioning measurement, total station measurement, and manual mapping data is determined based on the boundaries of the on-site sub-region and the distribution of measurement points. The acquisition coverage area is represented by closed surface geometric data, and the generation basis, source file number, and coverage area version number are recorded. If the coverage area is not closed, self-intersecting, or does not intersect with the mission sub-region, an acquisition coverage area anomaly record is generated.

[0037] The system reads the coordinate reference system information from historical geographic information object version data, current period surveying and mapping results data, and auxiliary surveying and mapping data to obtain the geodetic datum, projection method, central meridian, zone information, coordinate units, elevation datum, and coordinate system identifier. When the source and target coordinate reference systems are consistent, the original coordinates are retained; when they are inconsistent, coordinate transformation is performed based on the translation, rotation, scale, projection, and elevation correction parameters registered in the surveying and mapping task file. Before transformation, the original spatial geometric data and the original coordinate reference system identifier are saved; after transformation, the spatial geometric data under the target coordinate reference system, the coordinate transformation parameter number, and the transformation time are saved.

[0038] Read common control points with known coordinates in both the source and target coordinate reference systems, and calculate the abscissa residual, ordinate residual, and elevation residual. If the number or distribution of common control points does not meet the surveying quality control data, perform a transformation check by reading independent check points; if neither common control points nor independent check points meet the check conditions, generate a coordinate transformation check incomplete record. If coordinate transformation parameters are missing or coordinate transformation residuals exceed allowable values, register a coordinate transformation pending review status, and the corresponding data is not written to the current surveying object record.

[0039] The actual start and end times of data collection, image imaging time, point cloud acquisition time, survey point measurement time, object version validity time, and data writing time are converted to the unified time format used by the Geographic Information System (GIS), while preserving the original time, original time zone, and time precision. When the source data only has the collection date, the start and end times of that day are recorded as the collection time range, and the time precision is marked as date-level; if the collection date is missing, a missing collection time record is generated. When the acquisition time of auxiliary surveying data is not within the current surveying object's acquisition time range, the time interval between the two is recorded.

[0040] Perform geometric checks on current-cycle vector geographic objects that have undergone coordinate transformation and time normalization. Point objects are checked for coordinate fields, coordinate value ranges, and their spatial relationship with task boundaries and acquisition coverage. Line objects are checked for node order, consecutive repeating nodes, zero-length line segments, endpoints, and self-intersections. Polygon objects are checked for boundary closure, consecutive repeating nodes, self-intersections, inner and outer loop relationships, and polygon geometry non-empty states. Consecutive repeating nodes are merged according to the minimum coordinate storage resolution of the target coordinate reference system, and the original geometric data and node processing records are saved. Zero-length line segments, unclosed polygons, self-intersecting polygons, and objects whose inner and outer loop relationships cannot be determined are written into the mapping object geometric anomaly record. If the current mapping object is within the task boundary but not within the coverage area of ​​any acquisition batch, a current mapping object acquisition coverage missing record is generated.

[0041] The geographic grid is generated using map sheet division rules or spatial tile rules registered in the geographic information system, and records the geographic grid number, closed grid boundary, grid level, and grid version number. When a point object is located on a common grid boundary, its assigned grid is determined according to the geographic grid coding order, and the intersecting grid number is recorded. When line objects and polygon objects cross geographic grids, the spatial range of their intersection with each geographic grid is recorded separately, and no new geographic object number is generated.

[0042] In the spatial database of a Geographic Information System (GIS), an R-tree spatial index is built based on the minimum bounding rectangle of the current mapping object, historical object versions, control points, topological nodes, and auxiliary mapping data. A B-tree temporal index is built based on the collection start and end times, the effective start and end times of the object version, and the data write time. The spatiotemporal association index fields consist of the geographic grid number, the current mapping object's temporary number, the collection batch number, the object comparison segment number, and the object version number. When performing a historical object retrieval, the R-tree spatial index is first used to query the geographic grids that intersect with the spatial extent of the current mapping object, and then the B-tree temporal index is used to filter historical object versions before the current object's collection time.

[0043] When the current surveying object carries an existing geographic object number, the historical object version is queried according to the existing geographic object number, and the current object type code is compared with the historical object type code; if the types are inconsistent, a conflict record between the object number and the object type is generated. When the current surveying object does not carry an existing geographic object number, the minimum bounding rectangle of the current surveying object is expanded according to the historical allowable plane error, the current allowable plane error, and the coordinate transformation residual, and the historical object version that intersects with the expanded range and has the same object type code is queried.

[0044] When the current survey object does not carry an existing geographic object number, a buffer is generated outward based on the smallest bounding rectangle of the current object comparison segment, according to the historical object candidate version retrieval extension distance. An R-tree spatial index is used to query historical object versions that intersect with the buffer and have the same object type code. Then, a B-tree time index is used to filter historical object versions with a valid start time no later than the current acquisition time and a valid end time that is empty or later than the current acquisition time. If only one historical object version exists in the filtering results, that historical object version is added to the historical object candidate version record. If two or more historical object versions exist in the filtering results, a historical object version time conflict record is generated, and none of the historical object versions is directly selected. If no historical object version matching the time condition is obtained, a historical object version missing record is generated.

[0045] The valid time range of historical object versions is processed by including the valid start time but excluding the valid end time. An empty valid end time indicates that the corresponding historical object version remains valid. First, it checks whether the transition time of the object version change overlaps with the actual acquisition time range of the current survey object. If there is overlap, a conflict record between the object acquisition time and the version change time is generated, and the candidate historical object version is not determined. If there is no overlap, historical object versions with a confirmed valid status, a valid start time no later than the actual acquisition start time of the current survey object, and a valid end time that is empty or later than the actual acquisition start time of the current survey object are selected. If only one historical object version exists in the selection results, that historical object version is added to the historical object candidate version record. If two or more historical object versions exist in the selection results, a historical object version time conflict record is generated, and no candidate version is determined from the relevant historical object versions.

[0046] If no valid version with a valid time range covering the actual start time of the current survey object is found, query for valid historical object versions with a valid end time no later than the actual start time of the current survey object. Calculate the time interval between the actual start time of the current survey object and the valid end time of each historical object version, and sort them in ascending order of time interval. If the minimum time interval corresponds to only one historical object version, write the top-ranked historical object version into the historical object candidate version record and record the corresponding time interval; if the minimum time interval corresponds to two or more historical object versions, generate a historical object version time conflict record. If only historical object versions with a valid start time later than the actual start time of the current survey object are found, or if no historical object versions are found, generate a historical object version missing record.

[0047] Object comparison segments are divided based on the current mapping object source relationship records. When the current mapping object originates from a single acquisition batch, the object is registered as an object comparison segment. When the object is formed by merging the results of two or more acquisition batches, the spatial intersection of the current object's spatial geometry with the effective mapping range of each source file is calculated, and each spatial intersection is registered as an object comparison segment. When there is no clear source batch, object comparison segments to be verified are generated according to the batch affiliation candidate records.

[0048] Point objects are not geometrically segmented; the acquisition batch is determined based on the source measurement record and the actual measurement time. If the batch cannot be determined, the batch affiliation is retained as pending verification. When a line object intersects with the valid mapping area of ​​the source file more than twice, the intersection point is used as the segment endpoint, and segment numbers are generated according to the original line object node order. When a polygon object intersects with the valid mapping area of ​​the source file to form two or more independent polygons, object comparison segments are generated separately. The object comparison segment retains the current temporary number of the surveyed object and writes the object comparison segment number, acquisition batch number, actual acquisition time range, occupied geographic grid number, and source file number into it.

[0049] When two acquisition batches form the current mapping object in the same area, their respective temporary mapping object numbers are retained, and a candidate relationship record for overlapping objects in the acquisition batches is generated. When there is a spatial gap between the coverage areas of the acquisition batches, the geometry of the objects located within the spatial gap is written into the acquisition coverage missing record. When the object comparison segment is lower than the minimum representable length or minimum representable area, the segment geometry is retained, and a small-range object segment record is generated at the boundary of the acquisition batch; it is not directly merged into other object comparison segments.

[0050] For object comparison segments whose coverage boundaries intersect and belong to the same current mapping object, a record of the edge connection relationship for the acquisition batch is established. The edge connection position for line objects is recorded as the common endpoint, and the edge connection position for polygon objects is recorded as the boundary segment formed by the effective mapping range of the acquisition batch and the object geometry, with the start point, end point, and node order of the boundary segment written in. When object comparison segments have spatial overlap at their edge connection positions, an overlapping edge connection status is recorded; when there are areas with missing acquisition coverage, an intermittent edge connection status is recorded. The dividing boundary formed by the effective mapping range of the acquisition batch is only written into the edge connection relationship record of the acquisition batch and is not recorded as the original boundary of the geographic object.

[0051] The spatial geometry of historical object candidate versions is divided into equal-range segments according to the effective mapping range of the acquisition batch corresponding to the current object comparison segment, generating historical object candidate segments. Relationship records between the current object comparison segment and historical object candidate segments are generated based on the acquisition batch number, geographic grid number, and acquisition time. Each relationship record contains one current object comparison segment number and one historical object candidate segment number. When a current object comparison segment corresponds to more than two historical object candidate segments, separate relationship records are generated; the final correspondence between object segments is not determined in the current processing stage.

[0052] Auxiliary mapping data association records are generated based on the acquisition batch number, acquisition time, and spatial intersection range. These records include the current object comparison segment number, auxiliary mapping data type, auxiliary mapping data number, source acquisition batch number, spatial intersection range, acquisition time, time interval, quality status, and storage location. Data that fails the quality check retains its source registration but is not written to the auxiliary mapping data association record. When the spatial coverage does not cover the object comparison segment or the acquisition time is missing, the coverage loss status and time loss status are recorded respectively.

[0053] After data organization, a geographic information mapping spatiotemporal basic record, a batch object comparison segment record, and a geographic information spatiotemporal index are generated. The geographic information mapping spatiotemporal basic record includes the mapping task number, task sub-region number, target coordinate reference system, geographic grid version number, batch number, acquisition time range, coordinate transformation status, geometric check status, historical object candidate version record number, auxiliary mapping data association record number, and anomaly record number. The batch object comparison segment record includes the current mapping object temporary number, object comparison segment number, object type code, batch number, segment sequence number, segment spatial geometric data, source file number, actual acquisition time range, geographic grid number, edge object comparison segment number, edge location, edge status, and segment processing status. The geographic information spatiotemporal index stores the spatial index identifier, time index identifier, and the association between the geographic grid number, the current mapping object temporary number, the batch number, the object comparison segment number, and the object version number. Data with coordinate transformation pending verification, missing acquisition time, failed geometric check, missing acquisition coverage, or historical version time conflicts are recorded as corresponding anomaly records and are not written to the batch object comparison segment record.

[0054] S2. Based on the object comparison segment record and spatiotemporal index, filter reference points that have no change registration and whose coordinate residuals do not exceed the allowable value. Calculate the position correction amount based on the difference between the current coordinates and historical coordinates of the reference points, correct the historical object comparison segments, and generate the correspondence between the current object comparison segment and the historical object comparison segment.

[0055] Read the geographic information mapping spatiotemporal basic records, collection batch object comparison segment records, geographic information spatiotemporal index, historical object candidate version records, current object comparison segment-historical object candidate segment relationship records, auxiliary mapping data association records, geographic object change registration data, and mapping quality control data. Extract reference point candidates according to the collection batch number and reference point retrieval range. Perform data source, version time, object change, coordinate difference, spatial distribution, and topological relationship verification in sequence. Calculate the position correction vector, write the corrected geometric data of historical object candidate segments into a temporary spatial data table and establish a spatial index. Generate stable reference point records, segment position correction records, and initial correspondence records of object segments.

[0056] The system reads the coordinate transformation status, acquisition time status, geometric check status, acquisition coverage status, and fragment processing status from the batch object comparison fragment records, processing only the current object comparison fragments where all statuses are valid. If there are batch attribution pending verification, historical version time conflicts, geometric anomalies, or missing acquisition coverage, a fragment position correction supplement record is generated. Before calculating the feedback change vector, if the fragment position correction record is in a supplementary state due to insufficient stable reference points, convex hull not covering the position to be corrected, exceeding the limit for one-point-at-a-time verification, or correcting geometric anomalies, stable reference points are supplemented based on the feedback reference point coordinates. The position correction vector, historical object candidate fragment correction geometric data, initial corresponding record of the object fragment, current period actual change vector, and geographic object change candidate record are regenerated. If the recalculation still fails, it is marked as pending verification.

[0057] Surveying quality control data includes object type code, map scale, current allowable plane error, historical allowable plane error, minimum representable length, minimum representable area, boundary sampling interval, minimum number of reference points, minimum number of reference points for item-by-item verification, spatial distribution conditions of reference points, allowable values ​​for reference point verification, and allowable values ​​for position correction residuals. All parameters use plane distance units under the target coordinate reference system and are registered according to the surveying task quality control documents, surveying equipment calibration records, historical surveying results quality records, and coordinate transformation inspection records.

[0058] Based on the current object comparison segment number, the geographic grid number, collection batch number, actual collection time range, spatial geometric data of the current segment, adjacent geographic grid numbers, and auxiliary mapping data association record number are read from the geographic information spatiotemporal index. The geographic grid occupied by the current object comparison segment is merged with the first-layer adjacent geographic grids sharing the boundary line segment or boundary point, and then intersected with the boundary of the task sub-region to generate the reference point retrieval range. It does not continue to expand to the second-layer adjacent geographic grids.

[0059] The R-tree spatial index is used to query the current geographic objects, survey control points, and geographic object topology nodes belonging to the same acquisition batch within the reference point retrieval range. When the reference point candidate carries a reference point number or an existing geographic object number, the corresponding number and the B-tree time index are used to query the confirmed valid historical object version before the actual acquisition time. When no corresponding number is carried, a buffer is generated based on the current reference point coordinates and the historical object candidate version retrieval extension distance. The location of historical points of the same type is queried, and historical versions that meet the acquisition time condition are selected. The historical object candidate version retrieval extension distance is the sum of the current allowable plane error, the historical allowable plane error, and the maximum value of the coordinate transformation residual.

[0060] The query results extract the locations of survey control points, road intersections, pipeline fixed nodes, bridge fixed connection points, and permanent structures with independent feature point numbers. For survey control points, the relationship between current and historical coordinates is established based on the control point number; for road intersections, the relationship is established based on the set of road node numbers and connecting road numbers; for pipeline fixed nodes, the relationship is established based on the pipeline node number and connecting pipeline number; and for bridge fixed connection points and permanent structure locations, the relationship is established based on the geographic object number, feature point number, and adjacent boundary segment number. Area object boundary nodes without independent feature point numbers are not considered as candidate reference points.

[0061] The reference point candidate record includes the reference point number, reference point type, current and historical feature point coordinates, data source number, historical object version number, acquisition batch number, control network number, allowable plane error, coordinate transformation residual, attribute change status, version time status, and topological relationship status. When the same reference point number corresponds to multiple current or historical coordinates, a reference point number conflict record is generated. Current object comparison segments and other segments formed by dividing the same current surveying object are not used as position correction reference points.

[0062] Query the geographic object change registration data between the effective time of the historical object version and the current collection time. If the reference point has records of demolition, relocation, expansion, reconstruction or coordinate update, the historical version is in a transitional state, pending verification or version time conflict state, the data source quality check failed, the reference location is located in an area with missing collection coverage, or the current or historical coordinates are missing, write it into the reference point exclusion record or reference point version time conflict record.

[0063] The allowable values ​​for control point re-measurement are retrieved from the surveying task technical design document or the control network result quality inspection record, based on the control network level corresponding to the control network number. If no allowable value is directly registered, twice the square root of the sum of the squares of the historically registered coordinate plane position error and the current re-measurement coordinate plane position error is taken, and the smaller of this calculated value and the current allowable plane error is selected. If the control network level or any plane position error is missing, an abnormal control point status record is generated.

[0064] If a control point is valid, has no damaged or migrated markers, and the planar distance between the remeasured coordinates and the registered coordinates does not exceed the allowable value for control point remeasurement, it is written into the initial set of stable reference points. If the control point status is undetermined or the remeasurement result exceeds the limit, a control point status exception record is generated. If the length of the difference vector between the coordinates of two or more control points exceeds the sum of the allowable values ​​for the corresponding control point remeasurement, a control point coordinate conflict record is generated, and no position correction is performed on the corresponding current object comparison segment.

[0065] Perform topological relationship verification on road intersections, pipeline fixed nodes, bridge fixed connection points, and permanent structure feature points. For road intersections, compare the current and historical sets of connected road numbers and road connection directions; for pipeline fixed nodes, compare the sets of connected pipeline numbers and connection order; for bridge fixed connection points and permanent structure feature points, compare the numbers and arrangement order of adjacent boundary segments. If the topological relationships are consistent, write them into the coordinate difference verification set; if inconsistent, generate a reference point topological relationship conflict record.

[0066] Calculate the two-dimensional planar coordinate difference vector for each reference point in the coordinate difference verification set. When verifying each reference point, temporarily remove it and calculate the predicted coordinate difference based on the coordinate difference vectors of other reference points and their spatial distances. If the number of reference points participating in the prediction is lower than the minimum number of reference points required for verification, stop verifying the geographic object feature points and retain only valid re-measurement control points. If the planar distance between the actual coordinate difference vector and the predicted coordinate difference vector does not exceed the allowable value for reference point verification, add the reference point to the initial set of stable reference points; otherwise, generate a reference point coordinate difference conflict record. The allowable value for reference point verification is the sum of the current allowable planar error, the historical allowable planar error, and the maximum value of the coordinate transformation residual.

[0067] When coordinate differences conflict between measurement control points and geographic object feature points, retain the control points that are valid after remeasurement and exclude the corresponding geographic object feature points. Arrange the geographic object feature points in descending order of coordinate difference verification residuals, and remove feature points whose residuals exceed the allowable value for reference point verification, then re-verify. If residuals are the same, prioritize removing feature points with lower data source quality levels; if quality levels are the same, remove feature points with larger acquisition time intervals; if still indistinguishable, generate a segment position correction record to be supplemented.

[0068] A set of reference point coordinates is generated based on the historical coordinates in the initial set of stable reference points. Reference points with the same coordinates are merged to form a convex hull. If the number of merged reference points is lower than the minimum number of stable reference points, or if a closed convex hull with a non-zero area cannot be formed, a segment position correction record is generated to be supplemented. The minimum number of stable reference points is read from the surveying quality control data and is not less than the number required to form a closed convex hull with a non-zero area.

[0069] A set of locations to be corrected is generated for candidate fragments of historical objects. Point objects are written with historical point coordinates; line objects are written with all original line nodes, and temporary calculated locations are inserted when the distance between adjacent nodes exceeds the boundary sampling interval; surface objects are written with outer node nodes, inner node nodes, and temporary calculated locations inserted according to the boundary sampling interval. Each location to be corrected records the location number, historical object candidate fragment number, location source type, node sequence number, planar coordinates, and the corresponding current object comparison fragment number.

[0070] When the number of stable reference points reaches the minimum, the convex hull has a non-zero area, and all positions to be corrected are located inside or on the boundary of the convex hull, the spatial distribution of reference points is recorded as passed. Reference points that pass the verification of data source, version time, object changes, coordinate differences, topological relationships, quantity, and spatial distribution are written into the stable reference point record. If there are positions to be corrected outside the convex hull, the spatial distribution of reference points is recorded as failed, a fragment position correction supplementary record is generated, and segment position correction is not performed.

[0071] The fragment position correction supplement record is written with the current object comparison fragment number, historical object candidate fragment number, number of stable reference points, minimum number, convex hull geometry and check status, convex hull outside the convex hull position number to be corrected, coordinates, shortest distance to the convex hull boundary and data type to be supplemented, which is used to determine the feedback measurement range and the supplementary reference point position.

[0072] For the current object comparison segment where the spatial distribution of reference points is "passing", the position correction vector corresponding to the position to be corrected is calculated based on the two-dimensional planar coordinate difference vector of each stable reference point and the planar distance between the stable reference point and the position to be corrected: ; in, Indicates the first Position in the current object comparison segment The corresponding position correction vector; Indicates participation in the The current object comparison fragment position correction of the first The two-dimensional plane coordinate difference vector between a stable reference point and the current survey result and the historical object version is calculated by subtracting the historical coordinate from the current coordinate; Indicates the first A stable reference point relative to position Distance weights; Indicates the first A stable reference point and location Planar distance between them; Indicates participation in the The number of stable reference points for the current object's comparison fragment position correction; This represents the minimum coordinate storage resolution of the target coordinate reference system.

[0073] For point objects, the point coordinates are used as the positions to be corrected; for line objects, each line node is used as the position to be corrected; and for surface objects, both external and internal node points are used as the positions to be corrected. When the distance between adjacent nodes exceeds the boundary sampling interval, a temporary calculation position is inserted. This temporary calculation position is only used for correcting vector calculations and is not written into the original geometric data of the historical object.

[0074] The position correction vector is added to the corresponding coordinates of the candidate fragment of the historical object to generate corrected geometric data. The original geometric data, version number, and original coordinate reference system information of the historical object remain unchanged. A leave-one-out verification is performed on stable reference points. Each time, the current reference point is removed, and the position correction vector at the historical coordinates of that reference point is calculated based on the remaining reference points. The corrected historical coordinates are then compared with the current coordinates. If the number of remaining reference points reaches the minimum required level, the verified reference point is located within the convex hull formed by the remaining reference points, and the remaining planar distances of all reference points do not exceed the allowable value for position correction residuals, the fragment position correction is registered as passed; otherwise, the fragment position correction is registered as pending supplementation, and no initial corresponding record for the object fragment is generated.

[0075] Perform geometric checks on the corrected geometric data. Point objects are checked for coordinate fields and value ranges; line objects are checked for node order, zero-length line segments, and self-intersections; face objects are checked for boundary closure, self-intersections, inner and outer loop relationships, and face geometry non-empty states. If a check fails, a corrected geometric anomaly record is generated and updated to a pending supplementation status; if a check passes, the corrected geometric data, along with its historical object candidate fragment numbers, historical version numbers, current object comparison fragment numbers, fragment position correction record numbers, and correction status, are written to a temporary spatial data table.

[0076] A modified geometry R-tree spatial index is built based on the modified geometry field in the temporary spatial data table. A relationship is established in the geographic information spatiotemporal index between the current object comparison fragment number, the historical object candidate fragment number, the fragment position correction record number, and the modified geometry spatial index identifier. If index generation fails, an abnormal status is registered for the modified geometry index, and the corresponding data is not included in the candidate matching query.

[0077] The current object comparison fragment, after being corrected for its position, generates a search range corresponding to the object candidate fragment. The search range is based on the minimum bounding rectangle of the current object comparison fragment, and its extension distance is the sum of the current allowable planar error, the historical allowable planar error, the maximum value of the coordinate transformation residual, the maximum length of the position correction vector, and the minimum representable length. First, the historical object candidate fragments within the extended range are queried using the original geometry R-tree spatial index of the historical object. Corrected geometry is calculated for the supplementary candidates, and then the data intersecting with the current object comparison fragment is queried using the corrected geometry R-tree spatial index.

[0078] When supplementary retrieval yields candidate fragments of historical objects, a new relationship record is added between the current object comparison fragment and the candidate fragment of the historical object, and the source of the relationship is marked as "position correction supplementary retrieval." If the original candidate fragment no longer meets the retrieval criteria after correction, the original relationship record is retained and updated to "excluded after correction."

[0079] When the object type code of the current object comparison fragment is consistent with that of the historical object candidate fragment, perform geometric and topological relationship queries; when the type codes are inconsistent, query the geographic object change registration data. If there is an object type change record, mark it as an object type change pending verification; otherwise, mark it as an object type conflict.

[0080] For point objects, the planar distance between the corrected historical coordinates and the current coordinates is calculated. Candidate relationships are retained if the distance does not exceed the retrieval extension distance. For line objects, a buffer range is generated for the current line object. Candidate relationships are retained if the corrected historical line geometry intersects with the buffer range, or if both have the same connection node number. For polygon objects, the spatial intersection relationship between the corrected historical polygon geometry and the current polygon geometry is queried, and the numbers of adjacent, contained, and shared boundary objects are compared. Candidate relationships are retained if spatial intersection or shared topological objects exist; otherwise, they are updated and excluded after correction.

[0081] If the current mapping object carries an existing geographic object number, and the number, object type code, historical candidate version, and corrected spatial extent are all consistent, an initial corresponding record for number inheritance is generated. If it does not carry an existing geographic object number, an initial corresponding record for spatial candidates is generated based on the object type, corrected spatial relationships, and topological relationships. If the geometric relationship meets the conditions but the attribute categories are inconsistent, the attribute is marked as pending verification. If the connection, adjacency, inclusion, or shared boundary relationships are inconsistent, the topology is marked as pending verification.

[0082] When a candidate fragment of a historical object forms a candidate relationship with a comparison fragment of a current object, a one-to-one initial correspondence record is generated; when a candidate fragment of a historical object corresponds to more than two comparison fragments of current objects, a one-to-many initial correspondence record is generated; when more than two candidate fragments of historical objects correspond to a comparison fragment of a current object, a many-to-one initial correspondence record is generated; when multiple candidate fragments of historical objects form an intersecting candidate relationship with multiple comparison fragments of current objects, a many-to-many initial correspondence record is generated, and all candidate relationships are retained.

[0083] If no historical object candidate fragments are found for the current object comparison fragment, a record indicating "No historical candidate record for the current object" is generated. If the historical results and current data collection coverage are complete, quality checks pass, and the auxiliary surveying data or feedback measurement data are geometrically consistent with the current object, a new object candidate is generated; otherwise, it is marked as insufficient data. If the projections of the change vectors at each sampling location of a line object onto the normal direction of the historical line segment have the same sign and exceed the allowable error for actual change, a line object position movement candidate is generated. If the geometric change of a line object or area object does not exceed the allowable error for actual change but the business attribute field changes, an attribute change candidate is generated. After verification by auxiliary surveying data, geographic object change registration data, or feedback measurement data, new object candidates and attribute change candidates are recorded as either confirmed changes or pending verification.

[0084] After completing the query and classification, an initial corresponding record for the object fragment is generated. The record includes the current temporary number of the surveying object, the current object comparison fragment number, the historical geographic object number, the historical version number, the historical object candidate fragment number, the data collection batch number, the fragment position correction record number, the corrected geometric spatial index identifier, the initial corresponding type, the relationship source, the object type status, the attribute status, the topology status, the data collection coverage status, the auxiliary surveying data status, and the corresponding relationship processing status. The record also saves the following statuses: fragment position correction pending, object type conflict, object type change pending verification, attribute pending verification, topology pending verification, many-to-many candidate, no historical candidate for the current object, no current candidate for the historical object, insufficient data conditions, and corrected geometric anomaly status.

[0085] S3. By comparing the corrected historical object segments and the corresponding relationships between object segments, calculate the change vector of the current boundary point relative to the corrected historical boundary point at the edge position of the collection batch. Based on the difference in change vectors on both sides of the edge, the length of the change vector, the direction of the vector, and the range of continuous change, record the position difference of the collection batch, or generate candidate records of geographic object changes and feedback measurement task records.

[0086] Read fragment position correction records, object fragment initial correspondence records, batch object comparison fragment records, historical object candidate fragment correction geometric data, correction geometric R-tree spatial index, geographic information spatiotemporal index, batch overlapping object candidate relationship records, auxiliary surveying data association records, geographic object change registration data, and surveying quality control data. Query object comparison fragments with edge relationships according to the current surveying object temporary number, historical geographic object number, historical version number, batch number, and valid time. Calculate the change vector difference on both sides of the edge and generate batch position difference records, geographic object change candidate records, and feedback measurement task records.

[0087] The system reads the initial correspondence type, object type status, attribute status, topology status, acquisition coverage status, and correspondence processing status from the initial correspondence record of the object fragment. It proceeds to change calculation when the fragment position correction, correction geometry check, and correction geometry R-tree spatial index status are all valid; otherwise, it generates a data supplement record if there are issues such as position correction pending, correction geometry anomalies, historical version time conflicts, missing acquisition coverage, or vector result quality checks failing.

[0088] Based on the current temporary number of the surveyed object, query the comparison segments a and b of objects that are divided from the same object, collected from different batches, and have edge-connecting relationships. When the endpoint of the collection batch division of a line object is located on the original path, it is registered as the edge-connecting verification position and only participates in the line segment edge-connecting calculation; the collection batch division boundary inside the surface object only participates in the segment combination, and the intersection of the original inner and outer rings with the boundary of the collection batch coverage area is the edge-connecting verification position. When overlapping line segments are formed, the edge-connecting verification position is extracted according to the boundary sampling interval.

[0089] Read the actual start and end times of data collection on both sides of the boundary. If two data collection time ranges overlap, or the time interval does not exceed the allowable value for the same period's data collection interval and there is no geographic object change registration data, register the data collection time as comparable; if the time interval exceeds the limit or the data collection time is missing, generate a data collection time pending verification record. If there are construction, relocation, demolition, expansion, cadastral adjustment, or object type change records that intersect with the target object or boundary location within the data collection interval, register the change status during the data collection period and compare the change vector with the change registration spatial range.

[0090] If the current temporary mapping object numbers of the two object comparison segments are the same, the historical geographic object numbers and historical version numbers of the corresponding historical segments are the same, the edge verification positions come from the same boundary ring or line node sequence, and there is no unresolved many-to-many relationship, then calculate the edge change vector difference; otherwise, generate an edge object relationship to be verified record.

[0091] The original boundaries of candidate fragments of historical objects are sampled according to the boundary sampling interval. Line objects are sampled from the original starting point in node order and the ending point is retained; for surface objects, the inner and outer ring directions are unified and then sampled from the starting node of each ring. The segmentation boundaries and nodes of the acquisition batch are not written into the original boundary sampling record.

[0092] The tangent direction is calculated based on the historical boundary sampling position and its preceding and following positions. For non-inflection points, the direction perpendicular to the tangent is used as the normal; for inflection points, the direction bisects the angle between the two boundary lines is used as the normal. The external normal of the face object is determined according to the direction of the boundary node, and the original boundary intersection points of the current object are queried along the positive and negative directions of the normal. A single intersection point is recorded as the position corresponding to the current boundary; multiple intersection points are arranged in ascending order of normal distance, and the node order of the first intersection point and its corresponding preceding and following intersection points is verified. If the order still cannot be determined, a record of multiple boundary intersection points pending verification is generated; if there are no intersection points, a record of missing boundary position is generated.

[0093] At the joint verification location Read the historical boundary sampling positions on both sides and the corresponding current boundary positions, calculate the original displacement vectors of segments a and b before correction, and calculate the edge change vector difference by combining the segment position correction vector: ;

[0094] in, Indicates the location of the joint verification The planar difference between the actual change vectors on both sides; Indicates the location for edge verification; and These represent the comparison segments of the current objects on both sides of the edge; and These represent the original displacement vectors obtained by subtracting the historical boundary sampling coordinates from the current boundary coordinates on both sides. and These represent the position correction vectors of the comparison segments of the two objects at the edge verification position, respectively. and Represent the actual change vectors on both sides; symbols This represents the length of a two-dimensional planar vector.

[0095] The allowable plane error of the current surveying in the a-th and b-th acquisition batches, the maximum remaining plane distance in the position correction records of the two side segments, and the maximum value of the coordinate transformation residual are added together to obtain the allowable edge joint value. Edge joint change vector difference. The difference is directly compared with the allowable value for the joint, without deducting other errors. When the vector difference of the joint change does not exceed the allowable value, the continuous change status on both sides of the joint is recorded. When it exceeds the allowable value, a record of discontinuous change in the joint position is generated. The vector difference of the joint change and the allowable value for the joint are written into the joint difference judgment record of the acquisition batch.

[0096] For the first position corresponding to the valid current boundary The current object comparison fragment is read. Based on historical boundary sampling locations, the current boundary location, and segment location correction records, calculate the actual change vector: ; in, Indicates the first In the current object comparison segment, the first The actual change vector corresponding to each boundary sampling position, with the vector direction pointing from the corrected historical boundary sampling position to the current boundary corresponding position; Indicates the first candidate fragment of the historical object Original boundary sampling locations of each object; Indicates the original boundary of the current object and The corresponding position; Indicates the first The current object comparison fragment is located at the historical boundary sampling position. Position correction vector at; Indicates the corrected historical boundary sampling location; Indicates the segment number compared to the current object. Indicates the boundary sampling location number.

[0097] The sum of the historical allowable plane error, the current allowable plane error, the maximum value of the coordinate transformation residual, and the maximum remaining plane distance constitutes the actual allowable change error. If the actual change vector length does not exceed this allowable error, mark it within the allowable error range; otherwise, mark the change sampling position.

[0098] When the difference in the edge change vector does not exceed the allowable value for the edge and the lengths of both vectors do not exceed their respective allowable errors for actual changes, a record of position difference in the acquisition batch is generated; when the difference in the edge change vector does not exceed the allowable value for the edge, the lengths of both vectors exceed the allowable error and the normal projections have the same sign, a candidate for continuous change across acquisition batches is generated; when the difference in the edge change vector exceeds the allowable value for the edge, only one vector length exceeds the allowable error or the normal projections on both sides have opposite signs, a record of discontinuous edge position change and a feedback measurement task record are generated.

[0099] The surface object projects the actual change vector onto the unit outward normal of the historical sampling location. When the projection distance is greater than the allowable error of the actual change, the direction is marked outward; when it is less than the negative value of the allowable error, the direction is marked inward; when the distance is between the positive and negative errors, no direction marker is generated. When the change sampling location numbers are consecutive, the normal directions are the same, and there are no missing boundary positions or multiple intersection points to be verified, they are connected as a continuous change boundary segment; if the cumulative length reaches the minimum representable length, it is retained; otherwise, a local boundary to be verified record is generated.

[0100] Query the comparison segments between the current and historical objects based on the current temporary object number, historical geographic object number, and historical version number. When the distance between the endpoints of a line object does not exceed the allowable edge connection value, take the midpoint as the temporary edge connection coordinate and connect them according to the node sequence; when the distance between the boundary nodes of a polygon object does not exceed the allowable edge connection value, adjust the matching nodes to the midpoint, merge the polygon geometry, and delete the internal cutting boundaries.

[0101] After edge processing, the current object comparison fragments form the current object's combined geometric data, and the historical object candidate fragments corrected geometric data form the historical object's corrected combined geometric data. When the edge distance exceeds the limit, the node order does not correspond, the combined line objects intersect each other, or the combined surface objects are not closed, self-intersecting, or have incorrect inner and outer loop relationships, an object fragment combination verification record is generated, and object-level spatial difference calculation is not performed.

[0102] Spatial overlay is performed on the combined geometric data of the current object that passes the inspection and the corrected combined geometric data of the historical object. The portion of the current geometry that exceeds the historical geometry is designated as the added region, and the portion of the historical geometry that exceeds the current geometry is designated as the reduced region. A boundary error range is generated with the actual allowable error as its width. The portion falling into the boundary error range is subtracted from the added and reduced regions. When the remaining area reaches the minimum representable area, effective added region records and effective reduced region records are generated respectively.

[0103] Two-dimensional translation registration is performed on the corresponding boundary sampling positions in the combined geometry of the current object and historical objects. Candidate translation vectors and translation registration residuals are calculated based on each two-dimensional change vector. When the translation registration residuals of each valid boundary sampling position do not exceed the allowable error of the actual change, and the absolute value of the area difference between the two combined geometries does not exceed the minimum representable area, a candidate for moving the face object position is generated. If no candidate for moving the face object position is generated, a boundary expansion candidate is generated when a continuously changing boundary segment outward intersects with the boundary of an effective increasing region; a boundary contraction candidate is generated when a continuously changing boundary segment inward intersects with the boundary of an effective decreasing region; and a local shape change candidate is generated when both effective increasing and effective decreasing regions exist simultaneously.

[0104] Line object change judgment only uses the object's original endpoints. When the current line object's combined geometry extends beyond the historical line object's corrected combined geometry by the length of the extension along the original endpoint direction to the minimum representable length, a line object extension candidate is generated; when the end segment of the historical geometry extends beyond the current geometry to the minimum representable length, a line object shortening candidate is generated; when a historical line object continuously lacks corresponding boundary positions in the middle, and the acquisition coverage and current vector result quality status are valid, a line object break candidate is generated. Acquisition batch segmentation endpoints are not included in the judgment.

[0105] When the distance between the historical corrected coordinates and the current coordinates of a point object exceeds the allowable error for actual change, a position movement candidate is generated; if the distance does not exceed the allowable error but the attribute field changes, an attribute change candidate is generated. If the current point object has no historical candidates or the historical point object has no current candidates, the corresponding state is retained.

[0106] Read initial correspondence records for one-to-many, many-to-one, and many-to-many relationships. Fragments formed by splitting the same current mapping object's temporary number are not used as the basis for splitting. When multiple current mapping object temporary numbers meet the conditions for overlapping mapping, generate overlapping mapping object verification records; after excluding this relationship, when one historical object corresponds to multiple current objects with independent boundaries and temporary numbers, generate object splitting candidates; when multiple historical objects correspond to one current object with a closed boundary or a unified line connection relationship, generate object merging candidates.

[0107] For many-to-many initial correspondence records, the current object comparison segment and the historical object candidate segment are used as nodes, and the initial correspondence relationship of the object segments is used as the connecting edge to query the connected components of the graph. Combination, spatial overlay and topological relationship queries are performed on the current segment and historical segment in each connected component. If there are still multiple split, merge or local boundary change results, the many-to-many relationship is marked as pending verification.

[0108] Retrieves connection nodes, adjacency, containment, and shared boundary relationships. For road and pipeline candidate queries, the connection status between the original endpoint and road network nodes, pipeline nodes, valves, or ancillary facilities is retrieved. For plots, buildings, and water bodies candidate queries, the spatial relationships between effectively added and reduced areas and adjacent objects and shared boundaries are retrieved. When geometric changes are inconsistent with the topology query results, the topology change is marked as requiring verification.

[0109] Based on the spatial range of effectively added areas, effectively reduced areas, and continuously changing boundary segments, auxiliary mapping data is queried. Valid classification points are extracted from associated point cloud tiles, and classification codes and elevations are read to generate point cloud data records for the changed areas. If the point cloud quality check fails, classification is incomplete, or the number of valid points is insufficient, a point cloud data insufficiency status is recorded. The minimum number of valid points is determined by rounding up the product of the query range area and the minimum valid point density of the point cloud.

[0110] When issues arise such as insufficient stable reference points, need to supplement fragment position corrections, discontinuous changes in edge positions, missing corresponding boundary positions, need to verify multiple boundary intersections, need to verify acquisition time, need to verify object fragment combinations, need to verify topological changes, inconsistencies between auxiliary mapping data and geometric change results, need to verify overlapping mapping objects, or need to verify many-to-many relationships, a feedback measurement task record is generated. The feedback measurement spatial range covers candidate boundary changes, stable reference points on both sides of the edge, connection nodes, and areas with missing acquisition coverage. The allowable planar error for feedback measurement is less than the allowable error for actual changes.

[0111] The feedback measurement task record includes the surveying task number, geographic object number, current temporary surveying object number, historical version number, data acquisition batch number, edge location number, sampling location of the boundary to be measured, stable reference point number, connection node to be measured, feedback measurement spatial range, allowable plane error, supplementary data type, task triggering reason, and task execution status. Reachable points, original endpoints of objects, and surface boundary locations are measured using real-time dynamic differential positioning via a global navigation satellite system or a total station; continuous surface boundaries and inaccessible areas are locally mapped using unmanned aerial vehicles (UAVs).

[0112] After classifying the sources of change, a batch edge difference determination record and a geographic object change candidate record are generated. The batch edge difference determination record saves the comparison segment of objects on both sides of the edge, the batch, the edge location, the original displacement vector, the location correction vector, the edge change vector difference, the edge allowable value, the actual change vector, and the difference determination status. The geographic object change candidate record saves the geographic object number, historical version number, candidate change type, actual change vector, continuous change boundary segment, effective added area, effective reduced area, attribute change field, topological relationship record, auxiliary surveying data, and confirmation status.

[0113] Write continuously changing boundary segments, effective increase areas, and effective decrease areas into a temporary change data table. Establish a change candidate R-tree spatial index and a B-tree temporal index. Write the change candidate record number, geographic grid number, current mapping object temporary number, historical version number, and index identifier into the geographic information spatiotemporal index.

[0114] S4. Calculate the feedback change vector based on the candidate records of geographic object changes and the feedback measurement task records. Compare the feedback change vector with the current periodic change vector and the corresponding allowable surveying error to generate confirmed change, unconfirmed change, or pending verification results. Generate the formal geographic object geometry based on the confirmed change results and update the geographic object version, validity period, transaction time, spatial index, and time index.

[0115] The Geographic Information System (GIS) processor reads the edge difference determination record of the acquisition batch, the candidate record of geographic object changes, the feedback measurement task record, the record of comparison segments of objects in the acquisition batch, the record of segment position correction, the record of initial correspondence of object segments, the current period's surveying and mapping results data, the historical geographic information object version data, the geographic object change registration data, the surveying and mapping quality control data, and the auxiliary surveying and mapping data. It then matches these data according to the surveying and mapping task number, the geographic object number, the object version number, the acquisition batch number, the object comparison segment number, the candidate record number of changes, and the feedback measurement task number. Current object comparison segments of the same geographic object are grouped into the same processing set; when there are candidate object splitting or merging, historical geographic objects and current surveying and mapping objects with many-to-many correspondences are grouped into the same processing set.

[0116] The processing set is checked for object ID, version number, object classification code, coordinate reference system, acquisition time, geometric type, position correction status, object fragment correspondence, candidate geometry of changes, and surveying quality parameters. If necessary data is missing, stable reference points are insufficient, the spatial distribution of reference points does not meet the conditions, the position correction residual exceeds the limit, or the many-to-many relationship is not resolved, the processing set is marked as pending verification and will not be included in the formal object version writing.

[0117] The surveying quality control data includes the current allowable plane error, historical allowable plane error, feedback measurement allowable plane error, allowable coordinate transformation residual, allowable position correction residual, minimum number of stable reference points, minimum number of independent checkpoints, boundary sampling interval, allowable vector angle, minimum representable length, minimum representable area, allowable error for topological node connections, feedback measurement range extension distance, continuous area measurement area threshold, minimum effective point density of point cloud, and maximum number of remeasurements. Each parameter is written according to the surveying task technical design document, data production specifications, map scale, and result accuracy level. Missing parameters are marked as pending verification.

[0118] The system queries the feedback measurement task records associated with candidate changes to geographic objects. It then expands the feedback measurement range outwards from the minimum bounding rectangle of the candidate change geometry to generate the target measurement range. When a candidate change is located at the edge of a data collection batch, the target measurement range simultaneously covers the original object boundaries and stable reference points on both sides of the edge. The data collection batch's dividing boundary is not considered the actual boundary of the geographic object.

[0119] When a fixed solution from the Global Navigation Satellite System is obtained for point objects, line object endpoints, or short-distance boundary points, real-time dynamic differential positioning is used to acquire coordinates. When no fixed solution is available or obstructions exist, a total station is used to acquire coordinates. When the target measurement area of ​​a continuous planar boundary reaches the continuous planar measurement area threshold, UAV photogrammetry is used to acquire images, and boundary vertices are extracted from orthophotos or point clouds. When the residual of an independent checkpoint exceeds the allowable planar error of the feedback measurement, real-time dynamic differential positioning or a total station is used to re-acquire coordinates at the out-of-limit locations.

[0120] Transform the feedback measurement coordinates to the target coordinate reference system. For real-time dynamic differential positioning data, delete observations with non-fixed solutions, missing time values, or excessive coordinate dispersion. For total station data, delete observations with incomplete station points, backsight points, or instrument parameters. For UAV photogrammetry results, delete data with excessive residuals at independent checkpoints. If no valid coordinates are available, mark the feedback measurement data as invalid.

[0121] Extract valid control points, fixed road nodes, fixed pipeline nodes, fixed bridge points, and permanent building feature points from the feedback measurement range. Delete reference points that were relocated, demolished, rebuilt, or whose coordinates were remeasured during the feedback measurement period. Insufficient reference points are recorded if the number of stable reference points or independent checkpoints is below the minimum, or if the convex hull of the reference points does not cover the target's changed location.

[0122] Calculate the planar coordinate difference between the feedback measurement coordinates of valid reference points and the historical version coordinates. Weight the difference according to the inverse of the planar distance from the target location to each reference point to obtain the feedback position correction. When the correction residual of an independent checkpoint exceeds the allowable value, successively delete the non-control reference point with the largest residual and recalculate. If there are insufficient remaining reference points or coordinate conflicts occur among valid control points, mark the change judgment status as pending verification. Write the reference point number, coordinate difference, correction amount, correction residual, and processing status into the feedback measurement position correction record.

[0123] Sampling points are extracted from the feedback measurement boundary according to the boundary sampling interval. Historical original boundary sampling points are matched based on the historical geographic object number, historical object version number, boundary ring number, boundary segment number, node order, and cumulative distance along the boundary. Point objects are matched according to the geographic object number, line objects according to the historical node order and cumulative distance along the line, and polygon objects are matched with the outer and inner rings respectively. The endpoints and edges of the batch data collection are not considered actual endpoints or boundaries; when one-to-many or many-to-one correspondences of non-adjacent sampling points occur, the correspondence conflict is recorded.

[0124] Subtracting the sum of the historical original boundary sampling point coordinates and the feedback position correction from the feedback boundary sampling point coordinates yields the actual feedback change vector. When the historical geographic object number, historical object version number, boundary ring number, boundary segment number, and historical sampling position are all the same, compare the current period's actual change vector with the feedback actual change vector: ; in, This indicates a comparison fragment of the collected batch objects. The Middle The planar vector difference between the current period actual change vector and the feedback actual change vector at each boundary sampling position; This represents the actual change vector in the current cycle; Represents the actual change vector in the feedback, upper right corner. This indicates that the data source is feedback measurement; Indicates the allowable vector difference; This represents the total permissible planar error of the current period's surveying and mapping results, including the current period's acquisition error, boundary extraction error, coordinate transformation permissible error, and the permissible value of the acquisition batch location correction residual. This represents the total permissible plane error of the feedback measurement, including feedback data acquisition error, boundary extraction error, coordinate transformation permissible error, and feedback position correction residual permissible value; This indicates the remaining length of variation after deducting the allowable planar error of historical versions and feedback measurements; This represents the total permissible planar error for the historical geographic information object version; Indicates the sequence number of the comparison segment for the collected batch objects; Indicates the boundary sampling position number; symbol This represents the length of the plane vector in the target coordinate reference frame. , and All use the same length unit and are added according to the maximum permissible error.

[0125] First, compare the lengths of the current cycle change vector and the feedback change vector. If the lengths of both change vectors exceed their respective allowable errors, perform a direction consistency check: For point objects, compare the angle between the two change vectors; for line objects, determine the direction along the line and the normal direction based on the historical line segments. When moving purely along the normal, the two normal projections have the same sign and the projections along the line do not exceed the allowable error of the direction component. When moving purely along the line, the two projections along the line have the same sign and the normal projections do not exceed the allowable error of the direction component. When moving obliquely, the normal projections of the two change vectors have the same sign, and the projections along the line have the same sign, and the angle between the complete two-dimensional vectors does not exceed the allowable value of the vector angle; for surface objects, compare the angle between the two change vectors at the same boundary sampling position and their projections on the normal direction outside the historical boundary. Not greater than , When the change vectors are greater than zero and their directions are consistent, the corresponding sampling location is marked as a confirmed change. When the lengths of both change vectors do not exceed their respective allowable errors, the corresponding sampling location is marked as an unconfirmed change. When only one change vector exceeds its corresponding allowable error, the corresponding sampling location is marked as pending verification. When the lengths of both change vectors exceed their respective allowable errors, but... Greater than , If the value is not greater than zero or the direction is inconsistent, the corresponding sampling position will be marked as pending verification.

[0126] Point objects determine their direction based on the vector dot product and the allowable angle between vectors; line objects determine their line segment direction according to the historical node sequence, and compare the projections of the changing vectors onto the line segment normal direction and along the line direction; area objects determine their outward normals according to the historical boundary loop direction, and compare the signs of the projections of the changing vectors onto the outward normal direction. When the directions of the two edges connected by the boundary vertices are different, the direction of the sum of the unit outward normals of the two edges is taken; when the sum of the unit outward normals is zero, the judgment results of the sampling points inside the adjacent edge segments are read; if there are no valid judgment results inside the adjacent edge segments, the corresponding sampling position is marked as pending verification.

[0127] when Not greater than and When the value is not greater than zero, the corresponding sampling location is marked as not exceeding the allowable surveying error. Greater than ,or Not greater than , If the result is greater than zero but the two change vectors are not in the same direction, it is marked as a measurement result conflict. When a measurement result conflict occurs, the reference points are re-selected and the position correction is calculated; if all reference points are valid but the conflict still exists, the measurement range is expanded according to the feedback measurement range extension distance and the data is collected again. If a conflict still exists after reaching the maximum number of remeasurements, the change judgment status is marked as pending verification.

[0128] The formal object geometry is generated based on confirmed changes in sampling points, sampling points within the allowable surveying error range, and the valid geometry of the current cycle. When the allowable plane error of the feedback measurement is less than the allowable plane error of the current cycle's surveying results, the feedback coordinates are used for confirmed changes, while the coordinates of the current cycle's surveying results are used for other locations. Line objects connect changed and unchanged line segments according to the historical node sequence; surface objects combine outer and inner loops respectively, delete internal edges formed by the segmentation of the collection batch, and check boundary closure, self-intersection, inner loop inclusion relationships, and boundary loop direction. Objects that fail the geometry check are marked as pending verification.

[0129] Query adjacent geographic objects whose geometry intersects, connects, or shares edges with formal objects. If the changed boundary is confirmed to be a shared edge of two polygon objects, associate the shared edge change with both objects; if both objects confirm the same shared edge change, update the object version in the same database transaction; otherwise, keep them pending verification. Generate records for additions, disappearances, location movements, boundary expansions, boundary contractions, shape changes, line segment extensions, line segment shortenings, line segment breaks, attribute changes, object splits, or object merging based on the formal object geometry and object correspondence. If a historical object does not match the current object, an object disappearance record is only generated if the current period's valid mapping area completely covers the historical object geometry and there are no missing data areas.

[0130] When the location, boundaries, or attributes of an object change, the geographic object identifier is retained and a version number is added; a new geographic object identifier is assigned to a new object; when an object disappears, the historical valid version is closed; when objects are split or merged, the source object version is closed, and the source object, target object, change type, change time, and spatial intersection range are recorded.

[0131] When a specific time of change occurs, that time is used as the effective end time of the old version and the effective start time of the new version. When no specific time is available, the time interval for the change occurrence is formed by the last confirmation of no change and the first confirmation of change. When a version is written to the database, the transaction start time is recorded. When data correction occurs, the original transaction record is closed and a replacement record is inserted.

[0132] Before a database transaction begins, check the geometry of the formal objects, adjacency relationships, shared edge relationships, connectivity relationships, containment relationships, intersection relationships, object version numbers, and valid time order. If the checks pass, write the geographic object version record, object splitting or merging relationship record, topology relationship record, change detection result record, and index association record within the same transaction. If any record write fails, roll back the current transaction.

[0133] For the new version's official geometric calculation of the minimum bounding rectangle, the geographic object identifier, version number, validity time, and minimum bounding rectangle are written into the R-tree spatial index. The B-tree index is updated according to the geographic object identifier, version number, validity time, and transaction time, and the geographic grid-geographic object version-validity time association record is updated. Geographic objects that are pending verification or whose changes are not confirmed are not replaced with the current valid version.

[0134] After completing the database transaction, the system generates geographic information mapping change detection results, geographic object version update records, and spatiotemporal index update records, which respectively save the change type, change geometry, change occurrence time, data source, change determination status, differences between the old and new versions, and newly added, modified, or closed index items.

[0135] To verify whether fragment position correction can preserve real changes and distinguish positional disturbances in the acquisition batches, two acquisition batches with a boundary relationship were set up in the target coordinate reference system, and stable reference points were placed on both sides of the boundary. The positional disturbance of the acquisition batch consisted of random directional translation, spatial linear offset, and continuously varying offset, with a disturbance amplitude ranging from 0 meters to 0.5 meters, and adjacent amplitudes differing by 0.05 meters. Each disturbance amplitude was repeated thirty times, generating one hundred geographic objects each time. Forty-eight boundary sampling positions were extracted for each object, and measurement errors were superimposed on the current boundary coordinates and the coordinates of the stable reference points, respectively. The allowable error for actual change was set to 0.11 meters.

[0136] The simulation uses four data scenarios. Scenario 1 does not set actual changes to geographic objects; instead, it overlays different sampling batch position perturbations on both sides of the edge to calculate the misclassification rate of sampling positions without actual changes. Scenario 2 also does not set actual changes to geographic objects, but overlays a global coordinate translation of the same direction and length onto the current period's surface object and its surrounding stable reference points to calculate the misclassification rate of the overall batch translation of the surface object. Scenario 3 overlays a true extension vector along the outward normal direction of the historical boundary within the continuous boundary range of the surface object, along with sampling batch position perturbations. Scenario 4 overlays a true lateral movement vector perpendicular to the historical line segment direction onto the line object, along with sampling batch position perturbations. The length of the true change vector is between 0.14 meters and 0.32 meters, and the true change vector only incorporates the current boundary coordinates of the geographic object, excluding the coordinates of stable reference points.

[0137] The processor reads the current and historical coordinates of stable reference points, calculates the planar coordinate differences between each stable reference point, and then calculates the position correction vector for the boundary sampling positions using a distance-inverse weighted method. The processor adds the position correction vector to the historical boundary sampling position coordinates and subtracts the corrected historical boundary sampling position coordinates from the current boundary position coordinates to obtain the actual change vector. The simulation program matches the actual change vector with the pre-recorded true change vector one by one, recording the number of sampling positions included in the statistical analysis, the number of incorrectly identified sampling positions, the number of missed sampling positions, the length of the true change vector, the projected length of the position-corrected change vector, and the root mean square error of the change position.

[0138] Figure 5 In subplot (a), the solid dot represents the misjudgment rate of sampling positions without real changes, the dashed square dot represents the misjudgment rate of overall batch translation of the surface object, the dotted-dash triangle represents the missed detection rate of real boundary expansion, and the dotted-diamond dot represents the missed detection rate of lateral movement of the real line object. In subplot (b), the solid dot represents the root mean square error of the position of change in real boundary expansion, and the dashed square dot represents the root mean square error of the position of change in lateral movement of the real line object. The data points on each curve are the mean of thirty repeated trials, and the gray area around the curve represents a range of one standard deviation.

[0139] Depend on Figure 5 It is evident that when the amplitude of the batch position disturbance increases to 0.5 meters, the misjudgment rate for sampling positions without real changes is 0.59%, the misjudgment rate for overall batch translation of surface objects is 0.02%, the missed detection rate for real boundary expansion is 1.40%, and the missed detection rate for lateral movement of real line objects is 1.13%. The root mean square errors of the positions of real boundary expansion and lateral movement of real line objects are 0.044 meters and 0.042 meters, respectively. The retention rates of the real change vectors in the corresponding data records are approximately 99.8% and 100%, respectively, indicating that the real change vectors added to the geographic object boundary are still retained after position correction, and the coordinate translations jointly generated by the stable reference point and the geographic object are recorded as batch position disturbances.

[0140] To verify the impact of the spatiotemporal index of geographic information on object version queries, 100,000, 200,000, 400,000, 600,000, 800,000, and 1,000,000 geographic object version records were established according to the same geographic grid distribution, object type ratio, and valid time distribution, respectively. Both databases had R-tree spatial indexes and B-tree temporal indexes. One database contained a record linking geographic grid, geographic object version, and valid time, while the other did not. Each query randomly determined the geographic grid, spatial query range, and query time point, returning geographic object versions within the valid time period within that spatial range and at that query time point. Thirty repeated trials were performed for each record size, with 10,000 queries executed each time.

[0141] The database query log records the number of geographic object version records, index processing method, query start and end times, response time, number of candidate version records filtered by entry time, and number of returned versions for each iteration. The mean and standard deviation of the query response time are calculated based on the query response time of each repeated trial, thus obtaining... Figure 6 . Figure 6 The solid black line and circular markers in the diagram represent the spatiotemporal index of geographic information in this invention; the gray dashed line and square markers represent the comparison processing for records not written into the geographic grid-geographic object version-valid time association record; and the light-colored area represents plus or minus one standard deviation. Figure 6 As can be seen, as the number of geographic object version records increases, the query response time corresponding to the black solid line increases less than that of the gray dashed line. This indicates that the associated records between geographic grid, geographic object version, and valid time can narrow down the range of candidate versions in spatial query results and support spatiotemporal condition queries for object versions.

[0142] In summary, this invention collects batch object location correction records and object segment correspondences, calculates the change vector difference between the two sides of the edge of the collection batch, and generates candidate records of geographic object changes by combining vector length, direction, and continuous change range. This achieves the classification of location differences and object boundary changes in the collection batch, reducing misjudgments of edge location changes. By generating confirmed change results and formal geographic object geometry through feedback measurement task records, it synchronously updates geographic object version, validity time, transaction time, spatial index, and time index, realizing the corresponding storage of change results and object spatiotemporal versions, and achieving the effect of refining version index association.

[0143] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for detecting changes in geographic information mapping based on spatiotemporal data, characterized in that, include: Acquire historical geographic information object version data, current period surveying and mapping result data, collection batch data, geographic object change registration data and surveying and mapping quality control data, unify coordinate reference system and time benchmark, divide current geographic objects and historical geographic objects according to the coverage of collection batch, and generate object comparison fragment records and spatiotemporal indexes; Based on object comparison fragment records and spatiotemporal indexes, reference points with no change registrations and coordinate residuals not exceeding allowable values ​​are filtered out. The position correction amount is calculated based on the difference between the current coordinates and historical coordinates of the reference points, the historical object comparison fragments are corrected, and the correspondence between the current object comparison fragments and historical object comparison fragments is generated. By comparing the corrected historical object fragments and the corresponding relationships between object fragments, the change vector of the current boundary point relative to the corrected historical boundary point is calculated at the edge position of the collection batch. Based on the difference of change vectors on both sides of the edge, the length of the change vector, the direction of the vector and the range of continuous change, the position difference of the collection batch is recorded, or candidate records of geographic object changes and feedback measurement task records are generated. The feedback change vector is calculated based on the candidate records of geographic object changes and the feedback measurement task records. The feedback change vector is compared with the current periodic change vector and the corresponding surveying allowable error to generate confirmed change, unconfirmed change, or pending verification results. Based on the confirmed change results, the formal geographic object geometry is generated, and the geographic object version, validity period, transaction period, spatial index, and time index are updated.

2. The geographic information mapping change detection method based on spatiotemporal data as described in claim 1, characterized in that, The unified coordinate reference system and time base are used to segment current geographic objects and historical geographic objects according to the coverage of the data collection batch, generating object comparison fragment records and spatiotemporal indexes, including: The coordinates of historical geographic information object version data and current period surveying results data are converted to the target coordinate reference system, the collection time is converted to a unified time base, repeated continuous nodes in line objects are deleted, the closure status, self-intersection status and ring direction of the outer and inner rings of the surface objects are checked, and the current geographic object geometry and historical geographic object geometry that have passed the geometric check are intersected with the collection batch coverage area respectively. The object identification, version number, collection batch identification, original object boundary marker and collection batch segmentation boundary marker are written into the object comparison segment obtained by the intersection.

3. The geographic information mapping change detection method based on spatiotemporal data as described in claim 1 or 2, characterized in that, The generated object comparison fragment record and spatiotemporal index include: Filter the historical object versions from the query results. If the filtering results contain only one historical object version, generate candidate relationships between the current object comparison segment and the historical object version. When the filtering results have more than two historical object versions, a historical object version time conflict record is generated, and the candidate version of the historical object is uncertain. If there is no valid version with a valid time range covering the current collection time, the valid historical object versions earlier than the current collection time are sorted in ascending order according to the time interval between the current collection time and the valid end time of each historical object version. If the minimum time interval corresponds to only one historical object version, the first historical object version in the sorted list is read and the time interval is recorded. If the minimum time interval corresponds to more than two historical object versions, a historical object version time conflict record is generated. If only historical object versions later than the current collection time are found, a record of missing historical object versions is generated.

4. The geographic information mapping change detection method based on spatiotemporal data as described in claim 3, characterized in that, The reference points that have no record of changes and whose coordinate residuals do not exceed the allowable value include: Extract control points or geographic feature points with the same identifier in the historical period and the current period from the same collection batch, and delete points that have been migrated, demolished, rebuilt or whose coordinates have been remeasured within the collection time interval; For each geographic feature point, temporarily delete the geographic feature point, use the coordinate difference of reference points other than the geographic feature point to perform distance inverse weighting, calculate the current predicted coordinates of the geographic feature point, and delete geographic feature points whose distance between the current predicted coordinates and the current measured coordinates exceeds the allowable value of the position correction residual. When the number of retained points reaches the minimum number of reference points and the position of the convex hull covering the location of the retained points is calculated, the retained points are written into the reference point record.

5. The geographic information mapping change detection method based on spatiotemporal data as described in claim 4, characterized in that, The step of calculating the position correction based on the difference between the current coordinates and historical coordinates of the reference point, and correcting the historical object comparison segment, includes: Calculate the difference between the current coordinates and historical coordinates of each reference point in the plane; calculate the plane distance from the historical point object coordinates, historical line object nodes, or historical surface object boundary sampling points to each reference point. Based on the coordinate difference and distance weight of each reference point, calculate the position correction amount of the corresponding coordinates, nodes or boundary sampling points, translate the historical object comparison segment according to the position correction amount, and store the translated historical object comparison segment into a temporary spatial data table and an R tree spatial index.

6. The geographic information mapping change detection method based on spatiotemporal data as described in claim 5, characterized in that, The process of generating the correspondence between the current object comparison fragment and the historical object comparison fragment includes: Read the current and historical allowable plane errors from the surveying quality control data, the maximum value of the coordinate transformation residual in the coordinate transformation check record, and the maximum remaining plane distance in the segment position correction record. Add the current allowable plane error, the historical allowable plane error, the maximum value of the coordinate transformation residual, and the maximum remaining plane distance to obtain the object candidate corresponding retrieval extension distance. Based on the minimum bounding rectangle of the current object comparison segment, expand outward according to the object candidate corresponding retrieval extension distance to generate the object candidate corresponding retrieval range. Query the historical object comparison segments that intersect the modified geometric data and the object candidate corresponding retrieval range through the modified geometric R-tree spatial index. From the query results, retain historical object comparison segments that have the same object classification code, the same geometric type, and the same set of classification codes for connected objects or shared edge objects. A one-to-one relationship is recorded when one current object comparison fragment corresponds to one historical object comparison fragment; a one-to-many relationship is recorded when one current object comparison fragment corresponds to two or more historical object comparison fragments; a many-to-one relationship is recorded when two or more current object comparison fragments correspond to one historical object comparison fragment; and a many-to-many relationship is recorded when two or more current object comparison fragments correspond to two or more historical object comparison fragments.

7. The geographic information mapping change detection method based on spatiotemporal data as described in claim 6, characterized in that, The step of calculating the change vector of the current boundary point relative to the corrected historical boundary point at the edge position of the batch acquisition includes: For two collection batches whose coverage boundaries intersect and are associated with the same current temporary object, extract object comparison segments with the same historical geographic object identifier, historical version number, boundary ring identifier, and boundary node order; Output the vector difference of the change position of the sampling batch. For surface objects, extract the boundary sampling points from the intersection of the original outer or inner ring and the boundary of the coverage area of ​​the sampling batch. For line objects, use the sampling batch split endpoints as boundary sampling points instead of the actual endpoints of the line objects.

8. The geographic information mapping change detection method based on spatiotemporal data as described in claim 7, characterized in that, The process of recording the location differences of the collection batches, or generating candidate records of geographic object changes and feedback measurement task records based on the difference in change vectors on both sides of the junction, the length of the change vectors, the direction of the vectors, and the range of continuous change, includes: For line object comparison segments or surface object comparison segments formed by different acquisition batches and having acquisition batch edge relationships, the historical original object boundary sampling position coordinates are added to the corresponding position correction vector, and then the corrected historical original object boundary sampling position coordinates are subtracted from the current original object boundary corresponding position coordinates to obtain the boundary sampling position change vector. When the difference in the change vector of the edge position of the batch is not greater than the allowable value of the edge, and the length of the change vector on both sides of the edge is not greater than the sum of the allowable plane error of the historical object version and the allowable plane error of the corresponding current period survey result, the difference in the position of the batch is recorded. When the difference in the change vector at the edge of the batch is not greater than the allowable value for the edge, the lengths of the change vectors on both sides of the edge are greater than the sum of the corresponding errors, the projections of the two change vectors in the direction of the historical boundary normal have the same sign, and the cumulative length of the continuous change boundary segment of the line object is not less than the minimum representable length or the area of ​​the change region of the surface object is not less than the minimum representable area, a candidate record of geographic object change is generated. Feedback measurement task records are generated when the difference between the change vectors is greater than the allowable value for the junction, the length of the change vector on only one side of the junction is greater than the sum of the corresponding errors, the projections of the two change vectors in the direction of the historical boundary normal have opposite signs, or the object segments maintain a many-to-many relationship.

9. The geographic information mapping change detection method based on spatiotemporal data as described in claim 8, characterized in that, The step of calculating the feedback change vector based on candidate records of geographic object changes and feedback measurement task records, comparing the feedback change vector with the current periodic change vector and the corresponding allowable surveying error, and generating confirmed change, unconfirmed change, or pending verification results includes: Based on the feedback measurement task record, the feedback boundary coordinates and feedback reference point coordinates are collected. The planar coordinate difference between the feedback coordinates and historical coordinates of the feedback reference point is calculated. The feedback position correction is calculated using the distance reciprocal weighting method. The feedback change vector is obtained by subtracting the sum of the historical boundary coordinates and the feedback position correction from the feedback boundary coordinates. If the length of the difference between the feedback change vector and the current cycle change vector is not greater than the sum of the allowable plane error of the current cycle surveying results and the allowable plane error of the feedback measurement, and the length of the feedback change vector is greater than the sum of the allowable plane error of the historical object version and the allowable plane error of the feedback measurement, and the dot product of the two change vectors of the point object is greater than zero or the projections of the two change vectors of the line or surface object onto the historical boundary normal direction have the same sign, then the change is recorded and confirmed. If the length of the difference between the feedback change vector and the current period change vector is not greater than the sum of the previous errors, and the length of the feedback change vector is not greater than the sum of the allowable plane error of the historical object version and the allowable plane error of the feedback measurement, then the change is not confirmed. If the length of the difference between the feedback change vector and the current periodic change vector is greater than the sum of the previous errors, or if the direction of the change vector does not meet the corresponding direction condition, it is recorded as pending verification.

10. The geographic information mapping change detection method based on spatiotemporal data as described in claim 9, characterized in that, The step of generating formal geographic object geometry based on the confirmed change results, and updating the geographic object version, validity period, transaction time, spatial index, and time index includes: For records whose change determination status is confirmed and associated with valid feedback measurement data, the feedback boundary coordinates are used when the allowable plane error of the feedback measurement is less than the allowable plane error of the current period's surveying and mapping results, and the coordinates of the current period's surveying and mapping results are used when the allowable plane error of the feedback measurement is not less than the allowable plane error of the current period's surveying and mapping results. Generate point object coordinates, or generate a sequence of line object nodes or a circle object boundary ring according to the order of historical object nodes; when both circle objects that share a line segment with the changed boundary record and confirm the change, update the versions of the two circle objects in the same database transaction; When the location, boundaries, or attributes of an object change, the geographic object identifier is retained and a version number is added. When a historical object corresponds to two or more current objects, the object segmentation relationship is recorded. When two or more historical objects correspond to one current object, the object merging relationship is recorded. Write the time of change to the valid time field, write the database write time to the transaction time field, and update the spatial index and time index based on the formal geographic object geometry and valid time.

Images