<p>The risks of aging, damage, and disappearance of ancient buildings are becoming increasingly severe. Three-dimensional digital technology is increasingly crucial for their protection, restoration, and research. Addressing the low efficiency and poor accuracy of point cloud segmentation in the digital conservation of ancient wooden components, this paper proposes an innovative method that integrates traditional construction knowledge with modern point cloud processing techniques. First, based on construction techniques, we summarize a section acquisition method constrained by the rules of large timber joinery in ancient buildings. Second, we propose a method for extracting component segmentation parameters by fusing Euclidean clustering with construction knowledge, analyzing sectional point cloud data to obtain relevant parameters. Finally, by combining pass-through filtering and region-growing algorithms and utilizing the obtained parameters, this method achieves efficient and high-precision segmentation of components.</p>