Találati lista | Tudóstér

001-es BibID:	BIBFORM053615
Első szerző:	Lázár István (programtervező matematikus)
Cím:	Retinal Microaneurysm Detection Based on Intensity Profile Analysis / István Lázár, András Hajdu
Dátum:	2010
Megjegyzések:	The reliable detection of microaneurysms (MAs) on digital fundus images is essential for any diabetic retinopathy (DR) screening system, since the presence of MAs is usually the first sign of this disease. Manual identification of MAs is time-consuming, and unacceptable in an automatic screening system. Due to the improtance of this task, the literature on automatic MA detection is extensive, the proposed methods have varying accuracy and computational resource requirements.In this paper we present a novel approach to the extraction of MA candidates. The proposed algorithm is based on the construction of a two dimensional map, where MAs appear as isolated bright spots. To create this map, the entire image is scanned along lines with different directions to obtain one dimensional intensity profiles. Each intensity profile is thresholded through an adaptive thresholding step, and the resulting indicies corresponding to foreground pixels are transformed to two dimensional coordinates. Thresholding of the map results in a set of connected components, whereby MA candidates are selected based on diameter.
ISBN:	978 963 9894 72 3
Tárgyszavak:	Természettudományok Matematika- és számítástudományok előadáskivonat diabetic retinopathy retinal image automatic screening microaneurysm candidate intensity profile image processing
Megjelenés:	Proceedings of the 8th International Conference on Applied Informatics (ICAI2010) Vol. 1. / edited by Attila Egri-Nagy, Emőd Kovács, Gergely Kovásznai, Gábor Kusper, Tibor Tómács. - p. 157-165. -
További szerzők:	Hajdu András (1973-) (matematikus, informatikus)
Internet cím:	Szerző által megadott URL Intézményi repozitóriumban (DEA) tárolt változat
Borító:
Saját polcon: