Algorithms for three-dimensional morphology quantification of cracks in timber columns based on point cloud

The accurate quantification of crack sizes is particularly crucial for assessing the mechanical performance of naturally cracked timber columns. This paper focuses on two typical types of timber columns, cylindrical and square columns, and proposes a three-dimensional morphology quantification algorithm for cracks based on point cloud processing. This method involves a series of operations including morphology fitting, filtering, and clustering to accurately extract the crack point clouds from the original data, followed by a detailed quantification of the crack sizes through grille processing. It has been verified that the quantification results obtained by this algorithm have an error of less than 15 % compared to the size measured manually. This demonstrates the accuracy of this algorithm in quantifying the three-dimensional sizes of natural cracks. It provides a reference for damage detection in timber structures and offers more accurate physical parameters for further assessments of the mechanical performance of naturally cracked timber components.