Abstract
In this paper, we extend the basic edge-weighted centroidal Voronoi tessellation model (EWCVT) for image segmentation to a new advanced model, namely fuzzy and harmonic EWCVT model. This extended model introduces a fuzzy and harmonic form of clustering energy by combining the image intensity with cluster boundary information. Compared with the classic CVT and EWCVT methods, the fuzzy and harmonic EWCVT algorithm can not only overcome the sensitivity to the initialization and noise, but also improve the accuracy of clustering results, as verified in several biomedical images. © 2014 Springer International Publishing.
Author supplied keywords
Cite
CITATION STYLE
Hu, K., & Zhang, Y. J. (2014). Extended edge-weighted centroidal Voronoi tessellation for image segmentation. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8641 LNCS, pp. 164–175). Springer Verlag. https://doi.org/10.1007/978-3-319-09994-1_15
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
