Please use this identifier to cite or link to this item:
|Type:||Artigo de evento|
|Title:||Combining Hierarchical Watershed Metrics And Normalized Cut For Image Segmentation|
De Carvalho M.A.G.
|Abstract:||Combining partitioning techniques became a promising approach in order to implement image segmentation and has produced good results on different applications. Some works has studied the graph spectrum as a partitioning tool, by means of the so-called Normalized Cut method (NCut) as a final partitioning process for images modeled by different types of graphs. This work explores the Watershed Transform as a modeling tool, using different criteria of the hierarchical Watershed (Area, Volume and Dynamics), in order to convert an image into a graph, followed by the use of the NCut to perform the final segmentation. The main goal is to compare the image segmentation results obtained from graphs modeled by Watershed Transform with others relevant results, using images from Berkeley Database. © 2012 Taylor & Francis Group.|
|Appears in Collections:||Unicamp - Artigos e Outros Documentos|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.