Riverbed: A Novel User-Steered Image Segmentation Method Based on Optimum Boundary Tracking
Contribuinte(s) |
UNIVERSIDADE DE SÃO PAULO |
---|---|
Data(s) |
21/10/2013
21/10/2013
2012
|
Resumo |
This paper presents an optimum user-steered boundary tracking approach for image segmentation, which simulates the behavior of water flowing through a riverbed. The riverbed approach was devised using the image foresting transform with a never-exploited connectivity function. We analyze its properties in the derived image graphs and discuss its theoretical relation with other popular methods such as live wire and graph cuts. Several experiments show that riverbed can significantly reduce the number of user interactions (anchor points), as compared to live wire for objects with complex shapes. This paper also includes a discussion about how to combine different methods in order to take advantage of their complementary strengths. Fundacao de Amparo a Pesquisa do Estado de Sao Paulo (FAPESP) [09/16428-4, 07/52015-0, 09/11908-8, 11/01434-9] Fundacao de Amparo a Pesquisa do Estado de Sao Paulo (FAPESP) Conselho Nacional de Desenvolvimento Cientiacute Conselho Nacional de Desenvolvimento Cientiacute fico e Tecnologico (CNPq) [481556/2009-5, 201732/2007-6, 302617/2007-8] fico e Tecnologico (CNPq) |
Identificador |
IEEE TRANSACTIONS ON IMAGE PROCESSING, PISCATAWAY, v. 21, n. 6, supl. 4, Part 1, pp. 3042-3052, JUN, 2012 1057-7149 http://www.producao.usp.br/handle/BDPI/35254 10.1109/TIP.2012.2188034 |
Idioma(s) |
eng |
Publicador |
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC PISCATAWAY |
Relação |
IEEE TRANSACTIONS ON IMAGE PROCESSING |
Direitos |
restrictedAccess Copyright IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC |
Palavras-Chave | #GRAPH-CUT SEGMENTATION #GRAPH SEARCH ALGORITHMS #IMAGE FORESTING TRANSFORM (IFT) #WATERSHED TRANSFORM #RELATIVE FUZZY CONNECTEDNESS #LIVE-WIRE SEGMENTATION #ACTIVE SHAPE MODELS #INTERACTIVE SEGMENTATION #GRAPH CUTS #ALGORITHMS #COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE #ENGINEERING, ELECTRICAL & ELECTRONIC |
Tipo |
article original article publishedVersion |