A Comparative Study among Pattern Classifiers in Interactive Image Segmentation

Author

Spina, Thiago V. ; Montoya-Zegarra, Javier A. ; Andrijauskas, Fábio ; Faria, Fábio A. ; Zampieri, Carlos E A ; Pinto-Caceres, S.M. ; de Carvalho, T.J. ; Falcao, Alexandre X.

Author_Institution

Inst. of Comput., Univ. of Campinas (UNICAMP), Campinas, Brazil

fYear

2009

fDate

11-15 Oct. 2009

Firstpage

268

Lastpage

275

Abstract

Edition of natural images usually asks for considerable user involvement, being segmentation one of the main challenges. This paper describes an unified graph-based framework for fast, precise and accurate interactive image segmentation. The method divides segmentation into object recognition, enhancement and extraction. Recognition is done by the user when markers are selected inside and outside the object. Enhancement increases the dissimilarities between object and background and Extraction separates them. Enhancement is done by a fuzzy pixel classifier and it has a great impact in the number of markers required for extraction. In view of minimizing user involvement, we focus this paper on a comparative study among popular classifiers for enhancement, conducting experiments with several natural images and seven users.

Keywords

feature extraction; fuzzy set theory; image classification; image segmentation; object recognition; fuzzy pixel classifier; interactive image segmentation; natural images edition; object enhancement; object extraction; object recognition; pattern classifiers; unified graph based framework; user involvement; Artificial neural networks; Classification tree analysis; Computer graphics; Fuzzy neural networks; Image processing; Image segmentation; Object recognition; Process design; Support vector machine classification; Support vector machines; Pattern classifiers; fuzzy classification; graph-based image segmentation; image foresting transform;

fLanguage

English

Publisher

ieee

Conference_Titel

Computer Graphics and Image Processing (SIBGRAPI), 2009 XXII Brazilian Symposium on

Conference_Location

Rio de Janiero

ISSN

1550-1834

Print_ISBN

978-1-4244-4978-1

Electronic_ISBN

1550-1834

Type

conf

DOI

10.1109/SIBGRAPI.2009.45

Filename

5395194