Digital analysis of the size distribution of vascular bundles in bamboo culm

Abstract Understanding the structural characteristics of vascular bundles is fundamental to improving the performance of bamboo-based materials. Here, an image-processing-based intelligent model for bamboo culm cross-sections was developed, implementing bamboo contour extraction, vascular bundle identification, polar-coordinate construction, and morphological feature extraction. The model precisely estimated the vascular bundles’ position, length, width, aspect ratio, fiber-sheath area, and deflection angle. It demonstrated high accuracy and stability, supporting batch analysis of the geometric dimensions of cross-sectional vascular bundles. Radially, bundles at the inner skin were nearly circular, while those in the middle part of the culm wall were elongated and larger. At the outer skin, size decreased and tended toward circularity, with greater individual variability. The deflection angle decreased monotonically from the inner skin to the outer skin. Circumferentially, the area and length showed minimal variations. The widths at the inner skin and in the middle section change little, whereas the outer skin was wider near the long-axis and short-axis and thinner in inter-axial regions. The aspect ratio displayed an inverse relationship to the width, with periodic and symmetrical variations. These results provide data support for the structural analysis of bamboo materials as well as for the design and application of biomimetic materials.