Among pattern recognition and classification methods, geometric methods are more particularly developped. Firstly an edge vectorization technique based on "high" curvature points has been studied. Summarized shapes can then be compared to reference shapes using shape parameters, symmetry coefficients and metrics in order to establish a shape classification.
CITATION STYLE
Labouré, M. J., Fillère, I., Jourlin, M., & Becker, J. M. (1996). Geometric shape recognition and classification. Microscopy Microanalysis Microstructures, 7(5–6), 447–451. https://doi.org/10.1051/mmm:1996100
Mendeley helps you to discover research relevant for your work.