Scholarlink

<p>Heritage Building Information Modeling (HBIM) for traditional timber architectural heritage faces bottlenecks, including low efficiency, limited accuracy, and constraints on applying and integrating architectural expertise. An intelligent framework integrates machine vision (YOLO) with an expertise-based metamodel for traditional timber architectural heritage. The metamodel specifies core data and adaptive algorithms for automated HBIM; machine vision detects and acquires data from point-cloud images. This integrated approach provides a viable pathway to intelligent and automated HBIM for traditional timber architectural heritage and was validated on a Dong Drum Tower: for plan views, detection F1 Score 0.928 and Precision 0.969; for section views, detection F1 Score 0.873 and Precision 0.897. Automatically generated models achieved about 99% completeness, 99% accuracy, and efficiency improved by 95%. The method replaces strict geometric constraints with expertise-based construction logic, providing a new solution for data-driven intelligent HBIM suited to the numerous and typologically complex traditional timber architectural heritage.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Intelligent Hbim modeling of traditional timber architectural heritage, integrating machine vision and architectural expertise

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月17日

查看全文
  • Yi Deng,
  • Qianju Li,
  • Shihan Guo,
  • Ling Cai

摘要

Heritage Building Information Modeling (HBIM) for traditional timber architectural heritage faces bottlenecks, including low efficiency, limited accuracy, and constraints on applying and integrating architectural expertise. An intelligent framework integrates machine vision (YOLO) with an expertise-based metamodel for traditional timber architectural heritage. The metamodel specifies core data and adaptive algorithms for automated HBIM; machine vision detects and acquires data from point-cloud images. This integrated approach provides a viable pathway to intelligent and automated HBIM for traditional timber architectural heritage and was validated on a Dong Drum Tower: for plan views, detection F1 Score 0.928 and Precision 0.969; for section views, detection F1 Score 0.873 and Precision 0.897. Automatically generated models achieved about 99% completeness, 99% accuracy, and efficiency improved by 95%. The method replaces strict geometric constraints with expertise-based construction logic, providing a new solution for data-driven intelligent HBIM suited to the numerous and typologically complex traditional timber architectural heritage.