Scholarlink

<p>Fine modeling is a crucial step in the research of historical architectural design, but traditional methods often face data accuracy and processing efficiency issues. A terrestrial laser scanning (TLS) point cloud data processing and fine-grained modeling method based on an improved C-means algorithm is introduced to address this challenge. Firstly, high-precision TLS point cloud data is obtained through laser ranging and time measurement systems. Then, the improved C-means algorithm is used to classify, cluster, and extract features from the data, achieving refined modeling. The results showed that this algorithm achieved convergence in the 900th round, with an accuracy rate of up to 94.2%. When the parameters changed within a certain range, the average similarity of the new historical building dataset was mainly between 79 and 83%. The average similarity gradually increased to 84–96% as the parameters changed. The introduction of the improved C-means algorithm not only improved data processing efficiency and model accuracy but also achieved refined modeling of historical architectural design. This has profound significance for protecting, researching, and reusing historic buildings and provides new technical support and research directions for related fields.</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

TLS point cloud data processing and refined modeling of historical architectural design based on improved C-means algorithm

  • 期刊论文

  • 发表日期：

    2026年1月30日

查看全文
  • Yu Gao,
  • Chunhuan Guo,
  • Longlong Yang,
  • Jinlan Tan

摘要

Fine modeling is a crucial step in the research of historical architectural design, but traditional methods often face data accuracy and processing efficiency issues. A terrestrial laser scanning (TLS) point cloud data processing and fine-grained modeling method based on an improved C-means algorithm is introduced to address this challenge. Firstly, high-precision TLS point cloud data is obtained through laser ranging and time measurement systems. Then, the improved C-means algorithm is used to classify, cluster, and extract features from the data, achieving refined modeling. The results showed that this algorithm achieved convergence in the 900th round, with an accuracy rate of up to 94.2%. When the parameters changed within a certain range, the average similarity of the new historical building dataset was mainly between 79 and 83%. The average similarity gradually increased to 84–96% as the parameters changed. The introduction of the improved C-means algorithm not only improved data processing efficiency and model accuracy but also achieved refined modeling of historical architectural design. This has profound significance for protecting, researching, and reusing historic buildings and provides new technical support and research directions for related fields.