CN-121639937-B - Batch generation method and system for semantic three-dimensional family room model

CN121639937BCN 121639937 BCN121639937 BCN 121639937BCN-121639937-B

Abstract

The application provides a semantic three-dimensional house model batch generation method and a semantic three-dimensional house model batch generation system, which relate to the technical field of mapping and three-dimensional modeling, wherein the method comprises the steps of extracting original building information and preprocessing the original building information; loading shp vector surface data of a single building contour, fusing the shp vector surface data with an enhanced building information table, outputting a house-level vector image layer, executing roof model preprocessing, judging whether the house-level vector image layer is a roof, executing a roof generation rule to generate a semantic three-dimensional house model if the house-level vector image layer is the roof, and reading a house-level vector image layer to execute a house model generation rule chain to generate a batch of main bodies if the house-level vector image layer is not the roof to obtain the semantic three-dimensional house model. The method solves the technical problem that in the prior art, the modeling efficiency of the three-dimensional house model is low due to the dependence on traditional manual modeling, data splitting and model semantic deletion, and the modeling efficiency is improved by carrying out deep fusion on house vectors and building information to generate the three-dimensional house models in batches.

Inventors

Lv Beijia
DONG CHENGWEI
XING CHEN
ZHANG WAN
WANG LIN
Cao zhonghao

Assignees

北京市测绘设计研究院

Dates

Publication Date: 20260508
Application Date: 20251210

Claims (7)

1. The semantic three-dimensional family room model batch generation method is characterized by comprising the following steps of: Extracting original building information, wherein the original building information comprises a real estate unit number, a building number, a room number part, a starting floor and a stopping floor, preprocessing the original building information, and outputting an enhanced building information table; loading shp vector surface data of a single building contour, fusing the shp vector surface data with the enhanced building information table, and outputting a house-level vector layer; executing roof model preprocessing, judging whether the house-level vector image layer is a roof, if so, executing roof generation rules, and generating a semantic three-dimensional house model; if the building roof is not the roof, reading a building model generation rule chain of the building-level vector image layer to perform batch main body generation, and obtaining a semantic three-dimensional building model; An output enhanced building information form comprising: Building a family serial number algorithm formula: ; for the skip floor and cross-floor house type conditions, a house number correction formula is constructed: ; Wherein, the Indicating the room number for the i-th room without regard to the jump, Indicating the number of rooms after the i-th room is corrected by considering the jump layer situation, n indicating the total number of rooms, m indicating the total number of grouping conditions, 、 The value of the jth room and the ith room on the kth grouping condition is represented by I, the indication function is represented by I, if the condition is satisfied by I=1, if the condition is not satisfied by I=0, E represents a room number column, G represents a room starting layer number column, and L represents a room ending layer number column; Constructing an algorithm formula of the number of households of each layer: , wherein, The floor where the ith room is located is represented as a total of a plurality of households, I represents an indication function, G represents a room starting layer sequence, and L represents a room ending layer sequence; adopting the house number algorithm formula and the house number correction formula to calculate the house number of the original building information to obtain a house number information set; calculating the number of households of the original building information based on the algorithm formula of the number of households of each floor to obtain an information set of the number of households of each floor; and outputting the enhanced building information table according to the family serial number information set and the family number information set of each layer.
2. The semantic three-dimensional room model batch generation method of claim 1, wherein outputting a room-level vector layer comprises: loading shp vector surface data of a single building contour and the enhanced building information table to execute one-to-many attribute table association operation to obtain building information associated house vector data; automatically copying an original building contour surface to generate a plurality of geometrically coincident surface elements, wherein each surface element is associated with a house record in the enhanced building information table; And carrying out information fusion on the basis of the building information-associated house vector data and the face elements, and outputting the house-level vector image layer.
3. The batch generation method of semantic three-dimensional room models according to claim 1, wherein reading the room-level vector layer to execute a room model generation rule chain to perform batch main body generation, and obtaining the semantic three-dimensional room models comprises: The household model generation rule chain comprises attribute reading, three-dimensional stretching and pre-segmentation, household model generation and jump household processing; And carrying out batch generation processing on the family-level vector image layer based on the family model generation rule chain to obtain the semantic three-dimensional family model.
4. The method for batch generation of semantic three-dimensional room models according to claim 3, wherein obtaining the semantic three-dimensional room models comprises: According to the house-level vector image layer, acquiring building contours and calculating the length of an X side and the length of a Z side; Acquiring a side length comparison result of the X side length and the Z side length, and dividing according to the longest side according to the side length comparison result to acquire a household number and each floor of household number; judging the room position based on the room serial number and each layer of the number of rooms to obtain room position information; Determining a segmentation rule according to the room position information, and adding a hierarchical segmentation line based on the segmentation rule to obtain the semantic three-dimensional room model.
5. The semantic three-dimensional room model batch generation method of claim 4, wherein obtaining room location information comprises: if the room number=1, the room position information is the first room; If the room number is more than or equal to 1 and the room number is smaller than the number of the rooms in each layer, the room position information is a middle room; if the room number=the number of each floor of rooms, the room position information is the last room.
6. The method for batch generation of semantically three-dimensional user room models of claim 4, wherein determining the segmentation rules comprises: If the room position information is the first room, dividing the division rule into the number of the rooms of each layer according to the longest edge, generating the rooms in the first part, and enabling the remaining 1-1 parts of the number of the rooms of each layer to be empty; If the room position information is a middle room, the division rule is that the room position information is divided into the number of each layer of rooms according to the longest edge, the number-1 of the front room is empty, the number of the first room generates rooms, and the number of the remaining rooms of each layer is empty; if the room position information is the last room, the division rule is to divide the room position information into the number of the rooms of each layer according to the longest edge, the number of the rooms of the first layer is 1 to 1, and the last room is generated.
7. A semantic three-dimensional room model batch generation system, characterized by the steps for implementing the semantic three-dimensional room model batch generation method according to any one of claims 1 to 6, comprising: The data preparation module is used for extracting original building information, wherein the original building information comprises a real estate unit number, a building number, a room number part, a starting floor and a stopping floor, preprocessing the original building information and outputting an enhanced building information table; the geometric model generating module is used for loading shp vector surface data of a single building contour, fusing the shp vector surface data with the enhanced building information table and outputting a house-level vector image layer; The house-level vector layer judging module is used for executing roof model preprocessing, judging whether the house-level vector layer is a roof or not, if so, executing a roof generation rule and generating a semantic three-dimensional house model; And the batch processing module is used for reading the household-level vector image layer to execute a household model generation rule chain to generate batch main bodies if the household-level vector image layer is not a roof, so as to obtain a semantic three-dimensional household model.

Description

Batch generation method and system for semantic three-dimensional family room model Technical Field The application relates to the technical field of mapping and three-dimensional modeling, in particular to a semantic three-dimensional room model batch generation method and system. Background With the rapid development of the fields of smart cities, digital twinning, real estate management, city planning, emergency response and the like, higher requirements are put forward on the refinement and automatic modeling of three-dimensional models of urban buildings. Especially for residential buildings, not only an external three-dimensional building form is required to be constructed, but also an internal house model corresponding to building information is required to be generated so as to support house-level space query, analysis and visualization application. At present, the generation of a three-dimensional building and house model mainly depends on a traditional manual modeling mode, wall body, door and window and house segmentation are manually drawn according to two-dimensional drawings or field measurement data, and attribute information is manually input. However, manual modeling is time-consuming and labor-consuming, the efficiency requirement of urban large-scale modeling cannot be met, the efficiency is extremely low, the urban large-scale modeling requirement is difficult to deal with, and because the geometric vector data of the building and the building attribute information are usually stored in a system, errors are easily introduced in manual association, the generated model is inconsistent with objective reality, and the accuracy and reliability of the model are severely restricted. Meanwhile, the prior art has serious defects of automatic processing capability of non-standard house types such as skip floors, duplex floors and the like, lacks an effective recognition and modeling mechanism, and still needs a large amount of manual intervention for later repair, thereby further influencing the modeling efficiency of the three-dimensional house model. In summary, in the prior art, the problem of low modeling efficiency of the three-dimensional indoor model is caused by the dependence on traditional manual modeling, data splitting and model semantic deletion. Disclosure of Invention The application aims to provide a batch generation method and a batch generation system for a semantic three-dimensional room model, which are used for solving the technical problem that the three-dimensional room model modeling efficiency is low due to the fact that traditional manual modeling, data splitting and model semantic deletion are relied on in the prior art. In order to achieve the purpose, the application provides a semantic three-dimensional room model batch generation method and system. The application provides a semantic three-dimensional room model batch generation method which is realized by a semantic three-dimensional room model batch generation system, wherein the semantic three-dimensional room model batch generation method comprises the steps of extracting original building information, preprocessing the original building information, outputting an enhanced building information table, loading shp vector surface data of a single building contour and the enhanced building information table to fuse, outputting a room-level vector image layer, executing roof model preprocessing, judging whether the room-level vector image layer is a roof, executing a roof generation rule, generating a semantic three-dimensional room model, and if not, reading a room-level vector image layer, executing a room-level vector image layer generation rule chain, and carrying out batch main body generation to obtain the semantic three-dimensional room model. Optionally, a family room sequence number algorithm formula is constructed: And for the skip layer and cross-layer house type conditions, constructing a house serial number correction formula: wherein, the method comprises the steps of, Indicating the room number for the i-th room without regard to the jump,Indicating the number of rooms after the i-th room is corrected by considering the jump layer situation, n indicating the total number of rooms, m indicating the total number of grouping conditions,、The j-th room and the I-th room are represented by the values of the kth grouping condition, I represents the indication function, i=1 if the condition is satisfied, i=0 if the condition is not satisfied, e represents the room number column, G represents the room start layer number column, and L represents the room end layer number column. Constructing an algorithm formula of the number of households of each layer: , wherein, The method comprises the steps of representing that a floor where an ith room is located has a plurality of households, wherein I represents an indication function, G represents a starting floor number sequence, L represents a terminating floor number