Computational Science Technical Note CSTN-176

CSTN Home

Analysing Spinodal Decomposition using Image Morphology with Thinning, Edge Detection and Graph Methods

K. A. Hawick

Archived: 2013

Abstract

Edge detection and morphological thinning techniques can be applied to images of cross-sections of metallic alloys and other materials systems to analyse and characterise the spatial structures. This is particularly useful if it can be applied to a time-sequence of images to characterise the temporal aging of phase separating mixtures. We simulate alloy mixtures using the Cahn-Hilliard partial differential equation and use feature recognition techniques to transform the spatially growing domains into spatial networks which can be analysed as graphs using component labelling and size histogramming. We show that growth of large-scale spinodally decomposing structures in the Cahn-Hilliard system corresponds to a reduction in the number of boundary loops but a growth in the length of such loops.

Keywords: digital image processing; morphological thinning; spinodal decomposition; image representations; graph topology.

Full Document Text: PDF version.

Citation Information: BiBTeX database for CSTN Notes.

BiBTeX reference:

@INPROCEEDINGS{CSTN-176,
        author = {K. A. Hawick},
        title = {Analysing Spinodal Decomposition using Image Morphology with Thinning,
                Edge Detection and Graph Methods},
        booktitle = {Proc. IASTED International Conference on Signal and Image Processing
                (SIP 2013)},
        year = {2013},
        number = {CSTN-176},
        pages = {804-040},
        address = {Banff, Canada},
        month = {17-19 July},
        publisher = {IASTED},
        institution = {Computer Science, Massey University, Auckland, New Zealand},
        keywords = {digital image processing; morphological thinning; spinodal decomposition;
                image representations; graph topology.},
        owner = {kahawick},
        timestamp = {2013.03.24}
}


[ CSTN Index | CSTN BiBTeX ]