Dictionary Based Segmentation in Volumes
We present a method for supervised volumetric segmentation based on a dictionary of small cubes composed of pairs of intensity and label cubes. Intensity cubes are small image volumes where each voxel contains an image intensity. Label cubes are volumes with voxel-wise probabilities for a given label. The segmentation process is done by matching a cube from the volume, of the same size as the dictionary intensity cubes, to the most similar intensity dictionary cube, and from the associated label cube we get voxel-wise label probabilities. Probabilities from overlapping cubes are averaged and hereby we obtain a robust label probability encoding. The dictionary is computed from labeled volumetric image data based on weighted clustering. We experimentally demonstrate our method using two data sets from material science – a phantom data set of a solid oxide fuel cell simulation for detecting three phases and their interfaces, and a tomogram of a glass fiber composite used in wind turbine blades for detecting individual glass fibers.
KeywordsVolume segmentation Materials images X-ray tomography Learning dictionaries Glass fiber segmentation
Unable to display preview. Download preview PDF.
- 1.Aldasoro, C.C.R., Bhalerao, A.: Volumetric texture segmentation by discriminant feature selection and multiresolution classification. IEEE TMI 26(1), 1–14 (2007)Google Scholar
- 2.Blot, L., Zwiggelaar, R.: Synthesis and analysis of solid texture: application in medical imaging, pp. 9–12 (2002)Google Scholar
- 5.Dahl, A.L., Larsen, R.: Learning dictionaries of discriminative image patches. In: 22nd BMVC (2011)Google Scholar
- 6.Government, T.D.: The danish climate policy plan - towards a low carbon society. Technical report, Dahish Energy Agency (2013)Google Scholar
- 7.Hansen, J.Z., Brøndsted, P., Jacobsen, T.K.: The effects of fibre architecture on fatigue life-time of composite materials. Ph.D. thesis, Technical University of Denmark, Risø National Laboratory for Sustainable Energy (2013)Google Scholar
- 13.Nijssen, R.P.L.: Fatigue life prediction and strength degradation of wind turbine rotor blade composites. Contractor Report SAND2006-7810P, Sandia National Laboratories, Albuquerque, NM (2006)Google Scholar
- 14.Randen, T., Monsen, E., Signer, C., Abrahamsen, A., Hansen, J.O., Sæter, T., Schlaf, J., Sønneland, L., et al.: Three-dimensional texture attributes for seismic data analysis. In: 70th Annual International Meeting, Society of Exploration Geophysics Expanded Abstracts, pp. 668–671 (2000)Google Scholar
- 17.Ushizima, D., Parkinson, D., Nico, P., Ajo-Franklin, J., MacDowell, A., Kocar, B., Bethel, W., Sethian, J.: Statistical segmentation and porosity quantification of 3d x-ray microtomography. In: SPIE Optical Engineering Applications, pp. 813502–813502. International Society for Optics and Photonics (2011)Google Scholar
- 18.Waggoner, J., Zhou, Y., Simmons, J., De Graef, M., Wang, S.: 3d materials image segmentation by 2d propagation: a graph-cut approach considering homomorphism. IEEE TIP 22(12), 5282–5293 (2013)Google Scholar