摘要：Core Ideas A new ImageJ plugin, TopCap, automatically captures soil surface complexity from CT images. TopCap can quantify the immediate subsurface structure, highlighting soil crusting and sealing. Crust thickness varies under different soil textures following similar rainfall. The surface of a material such as soil, as characterized by its topology and roughness, typically has a profound effect on its functional behavior. While nondestructive imaging techniques such as X‐ray computed tomography (CT) have been used extensively in recent years to characterize the internal architecture of soil, less attention has been paid to the morphology of the soil surface, possibly because other techniques such as scanning electron microscopy and atomic force microscopy are viewed as more appropriate. However, X‐ray CT exploration of the surface of a soil also permits analyses immediately below its surface and beyond into the sample, contingent on its thickness. This provides important information such as how a connected structure might permit solute infiltration or gaseous diffusion through the surface and beyond into the subsurface matrix. A previous limitation to this approach had been the inability to segment and quantify the actual three‐dimensional structural complexity at the surface, rather than a predefined geometrically simplistic volume immediately below it. To overcome this, we formulated TopCap, a novel algorithm that operates with ImageJ as a plugin and automatically captures the actual three‐dimensional surface morphology, segments the pore structure within the acquired volume, and provides a series of incisive morphological measurements of the associated porous architecture. TopCap provides rapid, automated analysis of the immediate surface of materials and beyond, and while developed in the context of soil, is applicable to any three‐dimensional image volume.
关键词：2D; two-dimensional; 3D; three-dimensional; CT; computed tomography; GUI; graphical user interface; ROI; region of interest; SIOX; Simple Interactive Object Extraction.
