BYTE
Recongnition and separation of discrete objects with in complex 3D voxelized structures
Computers & Geosciences - Special issue on three-dimensional reconstruction, modelling and visualization of geologic materials
Improved segmentation of meteorite micro-CT images using local histograms
Computers & Geosciences
Hi-index | 0.00 |
Increased computer speed has helped to proliferate tomography throughout the geosciences. Although best known from the medical CATscan, tomography is also a powerful tool for geological problems. Whereas qualitative tomography is widely used, quantitative data from tomograms are not so commonly available. Presented here are two programs, Electron Tomography Segmentation, surface Area, and Volume (ETSAV) and Electron Tomography CUTter (ETCut), that were written to obtain quantitative volume and surface-area data from tomograms. Based on a specified threshold value, ETSAV distinguishes objects within a tomogram and then returns volume and surface-area data for each object. If object separation is unsatisfactory using a single threshold, ETCut is called to digitally separate objects that appear connected in the tomogram, allowing ETSAV to perform its functions. By knowing the nature of the object of interest, error can be corrected to, on average, less than 5% for a single object; combined volumes of many particles of different sizes may result in lesser errors. A section of the Brenham pallasite meteorite was examined using the programs. Olivine comprises 67.2% of the volume (50.4%wt%), followed by FeNi metal at 23.1% (39.3%wt%), troilite at 4.4% (4.8%wt%), and schreibersite at 3.4% (5.5%wt%). In all, 56.1% and 44.7% of the troilite and schreibersite surface areas, respectively, border olivine, suggesting that they wet olivine. Since the programs can be applied to any tomogram, from any field, there are many potential applications.