• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 1
  • Tagged with
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Detecção de vasos sanguineos em imagens de fundo de olhos / Detection of blood vessel in retinal images

PAIVA JÚNIOR, Sérgio de Sá Leitão 17 February 2006 (has links)
Submitted by (ana.araujo@ufrpe.br) on 2016-08-12T15:12:18Z No. of bitstreams: 1 Sergio de Sa Leitao Paiva Junior.pdf: 1027714 bytes, checksum: 382b44b0daa25e3bc9eed83acf4e7b18 (MD5) / Made available in DSpace on 2016-08-12T15:12:18Z (GMT). No. of bitstreams: 1 Sergio de Sa Leitao Paiva Junior.pdf: 1027714 bytes, checksum: 382b44b0daa25e3bc9eed83acf4e7b18 (MD5) Previous issue date: 2006-02-17 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior - CAPES / We describe an automated method for blood vessel segmentation in two-dimensional color retinal images. This method may be used by specialized software for automated screening of diabetic rethinopaty and other eye diseases, for evaluation of treatment evolution, and clinical study. The proposed method is based on a combination of “2d matched filter” response procedure (for vessel location), and the Hoshen-Kopelman algorithm (for vessel segmentation). We compare the results obtained through this method with the results of manual segmentation of 20 imagens of the DRIVE project, where ROC(Receiver Operating Characteristic curve) analysis was used for quantitative comparison with others works published on this subject. / Descrevemos neste trabalho um método automático para localizar e segmentar vasos sanguíneos em imagens de fundo de olhos. Este método pode ser usado por um software especialista em imagens de retina para diagnosticar automaticamente doenças como retinopatia diabética entre outra s , para avaliação da evolução do tratamento, e para estudos clínicos. O método proposto é baseado na combinação do procedimento “2d matched fi l t e r” ( usado na localização dos vasos) com o algoritmo de “Hoshen-Kopelman” (usado na segmentação dos vasos). Resultados obtidos com este método foram comparados, com a segmentação manual em 20 imagens do projeto DRIVE,usando ROC(Receiver Operating Characterist i c curve) para comparação quant i t a t i v a com os resultados de outros trabalhos publicados sobre este assunto.

Page generated in 0.0695 seconds