A multiscale approach to automatic and unsupervised retinal vessel segmentation using Self-Organizing Maps
- Autori: Lupascu, C.; Tegolo, D.
- Anno di pubblicazione: 2016
- Tipologia: Contributo in atti di convegno pubblicato in volume
- OA Link: http://hdl.handle.net/10447/226107
Abstract
In this paper an automatic unsupervised method for retinal vessel segmentation is described. Self-Organizing Map, modified Fuzzy C-Means, STAPLE algorithms and majority voting strategy were adopted to identify a segmentation of the retinal vessels. The performance of the proposed method was evaluated on the DRIVE database.