A method and system for automatically generating a room object based on a text label

By using a text-based identification method to filter specific layers in CAD drawings and parse the geometric and attribute data of Tianzheng wall objects, the problem of high dependence on drawing quality, low recognition accuracy, and low efficiency in existing technologies is solved. This achieves efficient and accurate generation of room objects, which is suitable for BIM model conversion in the field of architectural design.

CN122174302APending Publication Date: 2026-06-09BEIJING TIANZHENG SOFTWARE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
BEIJING TIANZHENG SOFTWARE CO LTD
Filing Date
2026-02-10
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

Existing technologies for generating BIM model room objects from CAD drawings suffer from several problems, including high dependence on drawing quality, inability to intelligently process Tianzheng objects, difficulty in connecting specific layer lines with Tianzheng wall lines, low accuracy in identifying closed wall areas, and a disconnect between generation and merging. These issues lead to recognition failures or fragmentation and are inefficient.

Method used

By using a text-based identification method, specific layers of CAD drawings are filtered, the geometric and attribute data of Tianzheng wall objects are parsed, line segment filtering and endpoint snapping are performed, intersection, break, merge and topology reconstruction are carried out, closed polygon regions are detected, and room objects with name, outline and area attributes are generated.

Benefits of technology

It achieves highly robust handling of auxiliary lines on non-standard layers during the design process, improves the success rate and accuracy of automated processing, and enhances the conversion efficiency from CAD drawings to BIM room models, making it particularly suitable for large-scale projects.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122174302A_ABST
    Figure CN122174302A_ABST
Patent Text Reader

Abstract

This application discloses a method and system for automatically generating room objects based on text identifiers. The method includes: filtering specific layers containing room names and Tianzheng wall layers; filtering line segments based on length, parallelism, and direction to obtain supplementary line segments that can participate in closure; parsing the geometric and attribute data of Tianzheng walls; snapping the supplementary line segments to the wall endpoints within tolerance and deleting redundancy to form a continuous edge set; performing intersection, breakage, merging, and topological reconstruction on the edge set; detecting all closed polygons and calculating their areas; mapping the closed areas containing text identifiers to the text content to generate Tianzheng room objects containing names, outlines, and areas; and automatically labeling the room names on the drawing. This invention achieves one-click automation of layer filtering, line segment merging, closure detection, text binding, and room generation, is compatible with non-standard layer auxiliary lines, and reduces the requirements for drawing integrity.
Need to check novelty before this filing date? Find Prior Art