Frozen-State Hierarchical Annealing
There is growing demand for methods to synthesize large images of porous media. Binary porous media generally contain structures with a wide range of scales. This poses difficulties for generating accurate samples using statistical techniques such as simulated annealing. Hierarchical methods have previously been found quite effective for such problems. In this paper, a frozen-state approach to hierarchical annealing is presented that offers over an order of magnitude reduction in computational complexity versus existing hierarchical techniques. Current limitations to this approach and areas of further research are discussed.
KeywordsPorous Medium Training Data Simulated Annealing Energy Function Training Image
Unable to display preview. Download preview PDF.
- 3.Alexander, S.K., Fieguth, P., Vrscay, E.R.: Hierarchical Annealing for Scientific Models. IEEE ICASSP 3, 33–36 (2004)Google Scholar
- 7.Hofmann, T., Puzicha, J., Buhmann, J.M.: Unsupervised texture segmentation in a deterministic annealing framework. IEEE PAMI 20(8), 803–818 (1998)Google Scholar
- 10.Liang, Z., Ioannidis, M.A., Chatzis, I.: Reconstruction of 3d porous media using simulated annealing. In: Bentley, et al. (eds.) Computational Methods in Water Resources XIII, Balkema, Rotterdam (2000)Google Scholar
- 11.Rozman, M.G., Utz, M.: Efficient reconstruction of multiphase morphologies from correlation functions. Physical Review E 63(066701) (2001)Google Scholar