Enhancing handwritten word segmentation by employing local spatial features

Simistira, F; Papavassiliou, V; Stafylakis, T; Katsouros, V

dc.contributor.author	Simistira, F	en
dc.contributor.author	Papavassiliou, V	en
dc.contributor.author	Stafylakis, T	en
dc.contributor.author	Katsouros, V	en
dc.date.accessioned	2014-03-01T02:53:16Z
dc.date.available	2014-03-01T02:53:16Z
dc.date.issued	2011	en
dc.identifier.issn	15205363	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/36199
dc.subject	document image processing	en
dc.subject	handwritten word segmentation	en
dc.subject	support vector machines	en
dc.subject.other	Connected component	en
dc.subject.other	Data sets	en
dc.subject.other	Document image processing	en
dc.subject.other	Gap metrics	en
dc.subject.other	Global threshold	en
dc.subject.other	Handwriting segmentation	en
dc.subject.other	handwritten word segmentation	en
dc.subject.other	Handwritten words	en
dc.subject.other	Linear classifiers	en
dc.subject.other	Linear SVM	en
dc.subject.other	Local feature	en
dc.subject.other	Objective functions	en
dc.subject.other	Spatial features	en
dc.subject.other	Support vector	en
dc.subject.other	Text lines	en
dc.subject.other	Word segmentation	en
dc.subject.other	Computational linguistics	en
dc.subject.other	Feature extraction	en
dc.subject.other	Image segmentation	en
dc.subject.other	Support vector machines	en
dc.subject.other	Text processing	en
dc.subject.other	Character recognition	en
dc.title	Enhancing handwritten word segmentation by employing local spatial features	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1109/ICDAR.2011.264	en
heal.identifier.secondary	http://dx.doi.org/10.1109/ICDAR.2011.264	en
heal.identifier.secondary	6065523	en
heal.publicationDate	2011	en
heal.abstract	This paper proposes an enhancement of our previously presented word segmentation method (ILSPLWseg) [1] by exploiting local spatial features. ILSP-LWseg is based on a gap metric that exploits the objective function of a soft-margin linear SVM that separates successive connected components (CCs). Then a global threshold for the gap metrics is estimated and used to classify the candidate gaps in ""within"" or ""between"" words classes. In the proposed enhancement the initial categorization is examined against the local features (i.e. margin and slope of the linear classifier for every pair of CCs in each text line) and a refined classification is applied for each text line. The method was tested on the benchmarking datasets of ICDAR07, ICDAR09 and ICFHR10 handwriting segmentation contests and performs better than the winning algorithm. © 2011 IEEE.	en
heal.journalName	Proceedings of the International Conference on Document Analysis and Recognition, ICDAR	en
dc.identifier.doi	10.1109/ICDAR.2011.264	en
dc.identifier.spage	1314	en
dc.identifier.epage	1318	en