Vector Image Retrieval Methods Based on Fuzzy Patterns
Yevgeniya S. Sulema, Etienne E. Kerre, Oksana Shkurat · International Journal of Modern Education and Computer Science · 2020
In this work we present two methods of vector graphic objects retrieval based on a fuzzy description of their shapes.Both methods enable the retrieval of vector images resembling to a given fuzzy pattern.The basic method offers a geometrical interpretation of a fuzziness measure as a radius of a circle with the center in each vertex of a given candidate object.It enables the representation of uncertain information about a pattern object defined by its "fuzzy" vertices.The advanced method generalizes this approach by considering an ellipse instead of a circle.The basic method can be used for the comparison of polygons and other primitives in vector images.The advanced method can be used for complex shapes retrieval.To enable saving a "fuzzy" image as a file, the modification of the SVG format with a new attribute "fuzziness" is proposed for both methods.The advanced method practical implementation is illustrated by the retrieval of medical images, namely, heart computer tomography images.