CN-122023702-A - Cadastral three-dimensional modeling method and cadastral three-dimensional modeling system based on point cloud data

CN122023702ACN 122023702 ACN122023702 ACN 122023702ACN-122023702-A

Abstract

The invention relates to the technical field of cadastral three-dimensional modeling, in particular to a cadastral three-dimensional modeling method and a cadastral three-dimensional modeling system based on point cloud data, which comprises the following steps of acquiring ground surface points, boundary points and house outline points based on an airborne laser radar, and constructing a point cloud adjacency relation structure, forming a boundary line direction sequence according to the projection direction of the connected points, generating a religion boundary path structure according to steering engagement, and constructing a religion three-dimensional space entity by combining the ground surface elevation and the boundary point coordinates. According to the invention, an adjacent foundation is established around boundary points and contour points in a point cloud, a connection path is screened by combining direction sorting and steering continuous relation, a connection sequence is limited along a path extension process, then a skeleton structure is merged according to the corresponding relation between the boundary path and the contour points, and a space node relation is organized by combining elevation information, so that boundary enclosing, skeleton connection and space configuration form uniform association, and corresponding improvement is provided for the situations of path misconnection, connection interruption and enclosing looseness.

Inventors

HU YUQING
YANG ZHEN
SONG DANNI
YANG TAO
HE FEI
LIU QIYU
TIAN YUCHUN
HUANG JIAN

Assignees

重庆市规划和自然资源调查监测院

Dates

Publication Date: 20260512
Application Date: 20260413

Claims (9)

1. The cadastral three-dimensional modeling method based on the point cloud data is characterized by comprising the following steps of: S1, acquiring distribution relations of ground surface points, boundary points and house outline points based on an airborne laser radar, calculating a corresponding relation between echo point coordinates and point numbers, screening adjacent point number combinations, and adjusting an adjacent recording order to obtain a point cloud adjacent relation structure; S2, analyzing the projection direction change of the connected points based on the point cloud adjacency relation structure, calculating a direction angle sequencing result, comparing the steering relation of the connected edges, screening a steering continuous edge sequence, and adjusting the arrangement sequence of the connected edges to obtain a boundary line direction sequence; S3, analyzing a connecting edge turning connection relation based on the boundary line direction sequence, judging the initial position of a boundary point, calculating the direction change of the adjacent edge in path extension, screening a continuous path sequence, and adjusting the path closing sequence to obtain a land boundary path structure; S4, based on the land parcel boundary path structure, analyzing the corresponding relation between the boundary sequence and the outline points of the house, calculating the connection relation between the boundary line segments and the outline points, comparing the connection conditions of adjacent line segments, and adjusting the merging sequence of the line segments to obtain cadastral skeleton connection data; And S5, acquiring the elevation of the ground surface and coordinates of the boundary points based on the cadastral framework connection data, calculating the spatial position corresponding relation of the nodes, comparing the enclosing relation of the boundary line and the contour line, and screening the node combination sequence to obtain the religion three-dimensional spatial entity.
2. The cadastral three-dimensional modeling method based on point cloud data according to claim 1, wherein the point cloud adjacency structure comprises a point number index, a neighbor point number list and a connection order identifier, the boundary direction sequence comprises a start azimuth mark, a turn azimuth mark and a boundary arrangement mark, the religion boundary path structure comprises a boundary start-stop identifier, a closure path unit and a religion attribution identifier, the cadastral skeleton connection data comprises a skeleton node identifier, a skeleton connection edge identifier and a contour hanging identifier, and the religion three-dimensional space entity comprises a religion bottom surface, a religion side surface and a religion closure body.
3. The cadastral three-dimensional modeling method based on point cloud data according to claim 1, wherein the step of obtaining the point cloud adjacency structure specifically comprises: s111, analyzing the unified reference relation of the coordinates of the ground surface point, the boundary point and the outline point of the house based on the airborne laser radar, judging the corresponding state of the echo point number and the space coordinate, screening the repeated relation of the number and the dislocation relation of the coordinates, and adjusting the corresponding relation of the abnormal records to obtain a point number coordinate association table; S112, analyzing the spatial adjacent relation of each point based on the point number coordinate association table, calculating the position proximity degree between the point numbers to form candidate combinations, comparing the connection points of the candidate point numbers with the circumferential order of the peripheral point numbers, and screening the crossing connection relation to obtain an adjacent point number combination set; and S113, analyzing the connection sequence relation of each point based on the adjacent point number combination set, calculating an adjacent point number azimuth sequencing structure, comparing the connection front and back receiving states of the same point and multiple connections, and adjusting the reverse connection relation and the staggered connection relation to obtain a point cloud adjacent relation structure.
4. The cadastral three-dimensional modeling method based on point cloud data according to claim 1, wherein the step of obtaining the boundary line direction sequence specifically comprises: s211, acquiring a coordinate projection relation of the connected points based on the point cloud adjacent relation structure, checking the landing positions of the connection start points and the terminal points, calculating the deflection order of the connection direction relative to the reference azimuth, and comparing the front-rear azimuth engagement relation of the same connection group to obtain a direction angle arrangement sequence; s212, analyzing the azimuth change relation of adjacent connecting edges of the same point based on the direction angle arrangement sequence, calculating the initial pointing rotation sequence of the tail end pointing direction of the front connecting edge and the rear connecting edge, comparing the steering bearing relation of the connecting edges, and screening continuous edge groups of the forward connecting relation to obtain a continuous steering edge set; S213, analyzing the front-back adjacent relation of each connecting edge based on the continuous turning edge set, calculating the extending arrangement order of the edge groups along the boundary, comparing the corresponding relation of the connection directions of the first edge and the tail edge, and adjusting the arrangement positions of the intermittent edge and the reverse edge to obtain the boundary line direction sequence.
5. The cadastral three-dimensional modeling method based on point cloud data according to claim 1, wherein the step of obtaining the religious boundary path structure specifically comprises: s311, analyzing the head-tail pointing relation of each connecting edge based on the boundary line direction sequence, checking the front-rear pointing connection state of the connecting point, calculating the edge steering sequence at the connecting point, comparing the distribution relation of the adjacent edges of the points which are not included in the path boundary line, and judging the corresponding point number of the path starting position to obtain a path starting point set; S312, analyzing the connection relation between the current edge and the subsequent edge in the extending process of the path based on the path starting point set, calculating the direction change sequence of the adjacent connection edges, comparing the continuous state of each candidate edge turning, and screening the group of the connection edges along the current path to obtain a path edge connection sequence; S313, analyzing the adjacent relation between the head point number and the tail point number of the path based on the path edge connection sequence, calculating the corresponding relation between the head point direction and the tail edge tieback direction, comparing the closed connection states, and adjusting the arrangement order of the path point numbers to obtain the religion boundary path structure.
6. The cadastral three-dimensional modeling method based on point cloud data according to claim 1, wherein the cadastral skeleton connection data obtaining step specifically comprises the following steps: s411, based on the religion boundary path structure, searching the end point coordinates of each line segment in the boundary path, matching the space position of the outline point outside the house, calculating the relation between the distance difference from the end point of the line segment to the outline point and the direction included angle, and comparing the space adjacent corresponding states of each line segment and the outline point to obtain a boundary outline matching coefficient; S412, calculating the connection direction difference value of the adjacent boundary line segments based on the boundary contour matching coefficient, comparing the connection continuous states of the adjacent line segments, and adopting the formula: ; Obtaining segment connection degree, screening connection relation continuous segment group to obtain segment connection sequence, The connection degree of the finger line segments represents the connection continuity index of two adjacent line segments in the geometric direction continuity and the matching degree with the outline points of the house, Refers to the direction included angle of two adjacent line segments at the shared endpoint, The term "circumferential rate constant" refers to the rate constant, Finger number Matching coefficients of the boundary line segments and the outline points of the house, Finger number Boundary contour matching coefficients of the boundary line segments; s413, based on the line segment connection sequence, checking the connection sequence relation before and after the line segments, comparing the space continuous state after the line segments are merged, and adjusting the line segment arrangement sequence and the connection merging structure to obtain cadastral skeleton connection data.
7. The cadastral three-dimensional modeling method based on point cloud data according to claim 1, wherein the step of obtaining the religious three-dimensional space entity specifically comprises: s511, checking the planar positions of points of corresponding points of skeleton nodes based on the cadastral skeleton connection data, matching the relationship of the corresponding table heights Cheng Lawei of each node, comparing the projection positions of the nodes with the elevation states, and arranging the spatial order to obtain a node spatial registration set; s512, analyzing the connection position relation of the boundary line and the contour line based on the node space registration set, calculating the connection relation of the endpoints of each line segment, comparing the closed connection state of the boundary line with the surrounding relation of the contour line, and screening node combinations to obtain a node connection sequence; And S513, analyzing the front-back connection sequence of the nodes based on the node connection sequence, calculating the space connection extension between the nodes, comparing the closing states of the head node and the tail node, and adjusting the node arrangement sequence to obtain the religion three-dimensional space entity.
8. The cadastral three-dimensional modeling method based on point cloud data according to claim 1, wherein the echo point coordinates refer to space coordinates corresponding to point cloud points formed by laser pulse contacting an earth surface, a building and an enclosing wall target and returning, and the projection direction change refers to a rotation change relationship between different connection directions after mapping the connection directions corresponding to the connection point coordinates to a horizontal plane.
9. A cadastral three-dimensional modeling system based on point cloud data, characterized in that the system is used for implementing the cadastral three-dimensional modeling method based on point cloud data according to any one of claims 1 to 8, the system comprising: The adjacency construction module acquires the distribution relation of ground surface points, boundary points and house outline points based on the airborne laser radar, calculates the corresponding relation between echo point coordinates and point numbers, screens adjacent point number combinations, and adjusts adjacency recording order to obtain a point cloud adjacency relation structure; the direction ordering module analyzes the projection direction change of the connected points based on the point cloud adjacency relation structure, calculates a direction angle ordering result, compares the steering relation of the connected edges, screens a steering continuous edge sequence, and adjusts the arrangement sequence of the connected edges to obtain a boundary line direction sequence; the path construction module analyzes the steering connection relation of the connecting edge based on the boundary line direction sequence, judges the initial position of the boundary point, calculates the direction change of the adjacent edge in the path extension, screens the continuous path sequence, and adjusts the path closing sequence to obtain a land boundary path structure; The framework generation module analyzes the corresponding relation between the boundary sequence and the outline point of the house based on the land boundary path structure, calculates the connection relation between the boundary line segment and the outline point, compares the connection condition of adjacent line segments, and adjusts the merging sequence of the line segments to obtain cadastral framework connection data; The entity construction module acquires the elevation of the ground table and coordinates of boundary points based on the cadastral framework connection data, calculates the corresponding relation of the space positions of the nodes, compares the enclosing relation of boundary lines and contour lines, and screens the node combination sequence to obtain the religion three-dimensional space entity.

Description

Cadastral three-dimensional modeling method and cadastral three-dimensional modeling system based on point cloud data Technical Field The invention relates to the technical field of cadastral three-dimensional modeling, in particular to a cadastral three-dimensional modeling method and system based on point cloud data. Background The technical field of cadastral three-dimensional modeling relates to a related technology for three-dimensional expression and management of land and accessories thereof by utilizing spatial data, and mainly comprises the steps of spatial modeling and data organization of elements such as land block boundary buildings, terrains and the like so as to support applications such as real estate registration homeland resource management, city planning and the like, and is widely applied to industries such as real estate management digital city construction, space information service and the like. The traditional cadastral three-dimensional modeling refers to geometric modeling and spatial expression of plots and attachments through measurement data image data or existing geographic information data, and discrete spatial information is converted into a continuous three-dimensional model by adopting a photogrammetry technology geographic information modeling technology or a computer graphics processing method, so that the spatial expression and management of cadastral information are realized. The existing cadastral three-dimensional modeling relies on measurement data and image data to carry out geometric expression, data processing focuses on point splicing and surface reconstruction, unified constraint of point-to-point abutting relationship, direction sequence and boundary enclosing relationship is lacking in the space organization process, adjacent point connection in point cloud depends on local position relationship, when land-parcel juncture, building turning and terrain change areas are faced, dislocation splicing and direction jumping are easy to occur on connection paths, continuity of boundary paths is difficult to maintain, a fracture phenomenon exists on a framework structure, enclosing relationship is difficult to stably establish, and boundary misconnection, structural separation and entity construction incompleteness are caused in the space expression process. Disclosure of Invention The invention aims to solve the defects in the prior art, and provides a cadastral three-dimensional modeling method and system based on point cloud data. In order to achieve the above purpose, the invention adopts the following technical scheme that the cadastral three-dimensional modeling method based on point cloud data comprises the following steps: S1, acquiring distribution relations of ground surface points, boundary points and house outline points based on an airborne laser radar, calculating a corresponding relation between echo point coordinates and point numbers, screening adjacent point number combinations, and adjusting an adjacent recording order to obtain a point cloud adjacent relation structure; S2, analyzing the projection direction change of the connected points based on the point cloud adjacency relation structure, calculating a direction angle sequencing result, comparing the steering relation of the connected edges, screening a steering continuous edge sequence, and adjusting the arrangement sequence of the connected edges to obtain a boundary line direction sequence; S3, analyzing a connecting edge turning connection relation based on the boundary line direction sequence, judging the initial position of a boundary point, calculating the direction change of the adjacent edge in path extension, screening a continuous path sequence, and adjusting the path closing sequence to obtain a land boundary path structure; S4, based on the land parcel boundary path structure, analyzing the corresponding relation between the boundary sequence and the outline points of the house, calculating the connection relation between the boundary line segments and the outline points, comparing the connection conditions of adjacent line segments, and adjusting the merging sequence of the line segments to obtain cadastral skeleton connection data; And S5, acquiring the elevation of the ground surface and coordinates of the boundary points based on the cadastral framework connection data, calculating the spatial position corresponding relation of the nodes, comparing the enclosing relation of the boundary line and the contour line, and screening the node combination sequence to obtain the religion three-dimensional spatial entity. The invention improves, the point cloud adjacency relation structure includes point index, adjacent point list and connection order mark, the boundary line direction sequence includes initial azimuth mark, steering order mark and boundary line arrangement mark, the land parcel boundary path structure includes boundary start-stop mark, enclosing path unit and land parcel