Other formats:
BibTeX
LaTeX
RIS
@inproceedings{1762543, author = {Melnikova, Aleksandra and Matula, Petr}, address = {Nice}, booktitle = {The IEEE International Symposium on Biomedical Imaging}, doi = {http://dx.doi.org/10.1109/ISBI48211.2021.9433867}, keywords = {Segmentation fusion; Reference annotation; Cell annotation; Shape}, howpublished = {elektronická verze "online"}, language = {eng}, location = {Nice}, isbn = {978-1-6654-1246-9}, pages = {204-207}, publisher = {IEEE}, title = {Topology Preserving Segmentation Fusion for Cells with Complex Shapes}, year = {2021} }
TY - JOUR ID - 1762543 AU - Melnikova, Aleksandra - Matula, Petr PY - 2021 TI - Topology Preserving Segmentation Fusion for Cells with Complex Shapes PB - IEEE CY - Nice SN - 9781665412469 KW - Segmentation fusion KW - Reference annotation KW - Cell annotation KW - Shape N2 - We present an algorithm to fuse simply connected segmentation masks of complex and variable shapes that often appear in cell imaging. The algorithm is designed to preserve topology of the input masks and to faithfully represent their protrusions. It works in three main phases: (1) the detection of geodesic ends that correspond to the protrusions, (2) optimal matching of the geodesic ends, and (3) contour averaging of corresponding boundary segments. We show that our algorithm overcomes commonly used pixel-wise fusion algorithms (namely majority voting, SIMPLE, STAPLE, and topology-preserving STAPLE), as well as recently published geometric median shapes in terms of the visual quality of results as well as better representation of protrusions. We demonstrate the performance of our method based on synthetic images as well as real images from the cell segmentation benchmark datasets. ER -
MELNIKOVA, Aleksandra and Petr MATULA. Topology Preserving Segmentation Fusion for Cells with Complex Shapes. Online. In \textit{The IEEE International Symposium on Biomedical Imaging}. Nice: IEEE, 2021, p.~204-207. ISBN~978-1-6654-1246-9. Available from: https://dx.doi.org/10.1109/ISBI48211.2021.9433867.
|