CN-121982253-A - Three-dimensional modeling method and system based on mapping multi-source data

CN121982253ACN 121982253 ACN121982253 ACN 121982253ACN-121982253-A

Abstract

The invention relates to the field of data modeling, in particular to a three-dimensional modeling method and system based on mapping multi-source data; the method comprises the steps of analyzing acquired mapping multi-source data, dividing a target area to generate a three-dimensional disturbance hierarchical map, analyzing the three-dimensional disturbance hierarchical map, extracting disturbance level attributes and modeling complexity data, analyzing the disturbance level attributes and the modeling complexity data to generate a modeling subarea attribute set, constructing a modeling boundary relationship and a modeling priority based on the modeling subarea attribute set, analyzing the modeling boundary relationship by combining the modeling priority to generate a modeling control structure, analyzing the target area based on the modeling control structure to generate candidate point cloud data, processing the candidate point cloud data to generate a modeling candidate data set, and generating a three-dimensional modeling fusion result according to the modeling control structure and the modeling candidate data set. The invention can promote the continuity and the regional adaptability of the three-dimensional modeling.

Inventors

LI LUFEI

Assignees

榆林市测绘地理信息中心(榆林市自然资源档案馆)

Dates

Publication Date: 20260505
Application Date: 20260126

Claims (8)

1. The three-dimensional modeling method based on mapping multi-source data is characterized by comprising the following steps of: S1, analyzing the acquired mapping multisource data, dividing a target area and generating a three-dimensional disturbance layering map; S2, analyzing the three-dimensional disturbance hierarchical map, extracting disturbance level attributes and modeling complexity data, analyzing the disturbance level attributes and the modeling complexity data, and generating a modeling subarea attribute set; S3, analyzing the target area based on the modeling control structure to generate candidate point cloud data; and S4, carrying out three-dimensional modeling and region fusion based on the modeling control structure and the modeling candidate data set, and generating a three-dimensional modeling fusion result.
2. The three-dimensional modeling method based on mapping multisource data according to claim 1, wherein analyzing the acquired mapping multisource data, dividing a target area, and generating a three-dimensional disturbance layered map includes: Carrying out data format analysis on the mapping multi-source data, carrying out three-dimensional coordinate reference alignment processing on the data of different sources based on a unified space reference coordinate system, and expressing the mapping multi-source data as point location information in the same three-dimensional space; The method comprises the steps of respectively calculating three-dimensional point cloud continuity indexes, direction vector stability indexes and local height change gradients for each three-dimensional space grid unit after mapping, constructing a disturbance intensity field model for mirror image disturbance intensity estimation of all three-dimensional space grid units in a target area based on the three-dimensional point cloud continuity indexes, the direction vector stability indexes and the local height change gradients, respectively dividing grid units with comprehensive disturbance intensity values in different sections into a stable modeling subarea, a light disturbance modeling subarea, a medium disturbance modeling subarea and a strong disturbance modeling subarea according to the disturbance intensity field model, and generating a corresponding three-dimensional disturbance layered map.
3. The three-dimensional modeling method based on mapping multi-source data according to claim 2, wherein the process of analyzing the three-dimensional disturbance hierarchical map, extracting disturbance level attributes and modeling complexity data, analyzing the disturbance level attributes and the modeling complexity data, and generating a modeling sub-region attribute set comprises the following steps: Analyzing the three-dimensional disturbance layering map, respectively obtaining three-dimensional space grid number sets of a stable modeling subarea, a light disturbance modeling subarea, a medium disturbance modeling subarea and a strong disturbance modeling subarea, and extracting disturbance level attributes corresponding to each grid unit; Setting modeling response control parameters for each modeling subarea according to the modeling complexity data and disturbance level attributes of each modeling subarea, wherein the modeling response control parameters comprise a modeling triggering mode, a modeling starting time condition, a modeling allowed space range, a modeling pause condition and a modeling delay window length, creating corresponding modeling execution monitoring tables for all the modeling subareas, and combining the modeling response control parameters with the modeling execution monitoring tables to generate a modeling subarea attribute set.
4. The three-dimensional modeling method based on mapping multi-source data according to claim 3, wherein modeling boundary relation and modeling priority are built based on modeling sub-region attribute sets, the modeling boundary relation is analyzed in combination with the modeling priority, and a process of generating a modeling control structure is as follows: Reading a modeling subarea attribute set, and executing area boundary extraction processing on three-dimensional space grid units in the modeling subarea boundary according to the three-dimensional space range, disturbance level label and modeling task complexity data of each modeling subarea recorded in the attribute set; Calculating modeling priority of a three-dimensional space grid unit positioned in a modeling subarea boundary area based on disturbance level labels, point cloud continuity indexes, direction vector change conditions, local altitude change gradients and modeling task stacking depth in a modeling subarea attribute set; sequencing modeling task queues corresponding to each modeling subarea according to the modeling priority, and generating a modeling execution sequence associated with the disturbance level; and generating a modeling control structure based on the ordered modeling task queues and the modeling boundary relation.
5. The three-dimensional modeling method based on mapping multi-source data according to claim 4, wherein the process of analyzing the target area based on the modeling control structure to generate candidate point cloud data comprises: reading the space ranges of the light disturbance modeling subarea and the medium disturbance modeling subarea from the modeling control structure, acquiring three-dimensional point cloud data of the current period in the corresponding space range, and calling the point cloud data of the previous modeling period from the data cached by the historical modeling data as reference point cloud data; The method comprises the steps of carrying out identification of a space overlapping region on point cloud data and reference point cloud data input in a current period, calculating local surface normal vectors, local curvature vectors and point cloud density change rates of corresponding points in the space overlapping region, comparing the change amplitudes of main direction vectors of the same grid units in two continuous modeling periods, and marking the corresponding grid units as 'direction unstable grids' if the change amplitudes exceed a threshold value; And calculating local boundary line segment displacement, local corner offset and curved surface slope variation for the area which is not marked as the unstable grid, marking the corresponding point set as a 'structure inconsistent point set' if the local boundary line segment displacement, the local corner offset and the curved surface slope variation are not within the tolerance range, and giving a low modeling weight label to the data points marked as the unstable grid or the structure inconsistent point set to generate candidate point cloud data.
6. The three-dimensional modeling method based on mapping multi-source data of claim 5, wherein the processing of the candidate point cloud data to generate the modeled candidate data set comprises: The method comprises the steps of carrying out multidimensional feature extraction processing on candidate point cloud data, calculating feature parameters including local curvature change amplitude, direction vector continuity score, neighborhood height fluctuation factor and historical disturbance record quantity for each data point, constructing feature description vectors of each data point based on the extracted feature parameters, inputting the feature description vectors into a disturbance absorption evaluation rule set, and judging the geometric credibility level of each data point; and carrying out grading treatment on the candidate point cloud data according to the geometric credibility grade, and generating a modeling candidate data set by analyzing the stable point proportion of each data point in the corresponding grid neighborhood.
7. The method of three-dimensional modeling based on mapping multi-source data according to claim 6, wherein the process of performing three-dimensional modeling and region fusion based on the modeling control structure and the modeling candidate data set to generate a three-dimensional modeling fusion result comprises: The method comprises the steps of carrying out three-dimensional modeling on a stable modeling subarea in a priority mode to construct a three-dimensional grid structure of the stable area, carrying out three-dimensional modeling on a light disturbance modeling subarea after a modeling delay time window set by a modeling rhythm control structure is finished, carrying out a local weighted fitting algorithm under directional constraint by combining point data with high geometric reliability grade and medium geometric reliability grade in a modeling candidate data set, adopting a boundary adsorption strategy to ensure connection continuity between the modeling subarea and the stable modeling subarea, carrying out continuous boundary fusion processing on the medium disturbance modeling subarea based on normal vector consistency and curvature compatibility, carrying out three-dimensional modeling on three-dimensional modeling subareas respectively, fusing modeling results of all subareas, and generating a three-dimensional modeling fusion result with consistent integral area.
8. The three-dimensional modeling system based on mapping multi-source data is particularly applied to the three-dimensional modeling method based on mapping multi-source data according to any one of claims 1 to 7, and comprises a management center, wherein the management center is in communication connection with a data layering module, a data control module, a modeling candidate module and a modeling fusion module: the data layering module is used for analyzing the acquired mapping multisource data, dividing a target area and generating a three-dimensional disturbance layering map; The data control module is used for analyzing the three-dimensional disturbance layered map, extracting disturbance level attributes and modeling complexity data, analyzing the disturbance level attributes and the modeling complexity data to generate a modeling subarea attribute set, constructing a modeling boundary relationship and a modeling priority based on the modeling subarea attribute set, and analyzing the modeling boundary relationship by combining the modeling priority to generate a modeling control structure; The modeling candidate module is used for analyzing the target area based on the modeling control structure to generate candidate point cloud data; The modeling fusion module is used for carrying out three-dimensional modeling and region fusion based on the modeling control structure and the modeling candidate data set, and generating a three-dimensional modeling fusion result.

Description

Three-dimensional modeling method and system based on mapping multi-source data Technical Field The invention relates to the field of data modeling, in particular to a three-dimensional modeling method and system based on mapping multi-source data. Background In the fields of three-dimensional modeling, geographic information systems, digital twinning, infrastructure planning and the like, high-precision three-dimensional modeling based on mapping multi-source data (such as laser radar point clouds, remote sensing images, oblique photography, DEM and the like) has become a key support technology. The modeling method is generally based on the spatial analysis of multi-source heterogeneous data, and combines multidimensional spatial features with terrain structure information to realize the three-dimensional reconstruction of a target area. In particular, in the scenes of urban updating, disaster monitoring, traffic engineering, ecological assessment and the like, the three-dimensional form of the earth surface structure is accurately and efficiently reconstructed, and the three-dimensional form is a precondition for subsequent analysis, simulation and decision. The China patent with the publication number of CN116778105A discloses a multi-precision three-dimensional mapping data fusion modeling method, which comprises the steps of acquiring mapping data at mapping points, summarizing, establishing a first topographic survey data set, establishing a mapping condition set, generating a mapping condition coefficient Cxs, screening a mapping image by a first image quality coefficient when the mapping condition coefficient Cxs is not higher than a condition threshold value, screening or recombining a first image to be selected, determining a target image, establishing a two topographic survey data set, carrying out three-dimensional processing on a planar live-action map, fusing the two three-dimensional topographic models to form a model to be selected, screening an error area if fusion errors exist, and collecting topographic data at the mapping points to correct the model to be selected. In the prior art, the problem that modeling is over-fitted, boundary discontinuous or modeling calculation resource allocation is uneven easily occurs due to the lack of a dynamic identification and layering control mechanism for the disturbance degree in a modeling area, which leads to mixed modeling of a high disturbance area and a stable area and influences the accuracy and stability of an overall model, and the problem that continuous fusion among modeling results is difficult to realize after multi-subarea modeling, which often causes regional boundary faults, structural jump and the like, is solved. Disclosure of Invention The invention aims at solving the problems in the background technology and provides a three-dimensional modeling method and system based on mapping multi-source data. The technical scheme of the invention is that the three-dimensional modeling method based on mapping multi-source data comprises the following steps: S1, analyzing the acquired mapping multisource data, dividing a target area and generating a three-dimensional disturbance layering map; S2, analyzing the three-dimensional disturbance hierarchical map, extracting disturbance level attributes and modeling complexity data, analyzing the disturbance level attributes and the modeling complexity data, and generating a modeling subarea attribute set; S3, analyzing the target area based on the modeling control structure to generate candidate point cloud data; and S4, carrying out three-dimensional modeling and region fusion based on the modeling control structure and the modeling candidate data set, and generating a three-dimensional modeling fusion result. Preferably, the process of analyzing the acquired mapping multisource data, dividing the target area and generating the three-dimensional disturbance layered map includes: Carrying out data format analysis on the mapping multi-source data, carrying out three-dimensional coordinate reference alignment processing on the data of different sources based on a unified space reference coordinate system, and expressing the mapping multi-source data as point location information in the same three-dimensional space; The method comprises the steps of respectively calculating three-dimensional point cloud continuity indexes, direction vector stability indexes and local height change gradients for each three-dimensional space grid unit after mapping, constructing a disturbance intensity field model for mirror image disturbance intensity estimation of all three-dimensional space grid units in a target area based on the three-dimensional point cloud continuity indexes, the direction vector stability indexes and the local height change gradients, respectively dividing grid units with comprehensive disturbance intensity values in different sections into a stable modeling subarea, a light disturbance modeling s