Robust Extraction of Line Segments from Colour Images by an Iterative Weighted Polarised Hough Transform (CROSBI ID 467575)
Prilog sa skupa u zborniku | izvorni znanstveni rad | međunarodna recenzija
Podaci o odgovornosti
Šegvić, Siniša
engleski
Robust Extraction of Line Segments from Colour Images by an Iterative Weighted Polarised Hough Transform
Abstract - In this paper, a nonparametric iterative procedure for robust extraction of line segments from a colour image is introduced. On input, weighted edge map and gradient direction map are provided by applying a nonparametric modification of Canny edge detector to original colour image. Then, a sequence of following three algorithms is iteratively applied: weighted polarised Hough transform, algorithm for detecting prominent local maxima in parameter space and algorithm for extracting line segments given the list of prominent lines. An iterative approach is applied in order to overcome the "screening effect" which is identified as a cause for many problems tied with standard Hough transform. Results on three sets of colour images are reported in order to show performance and robustness of the proposed procedure.
image segmentation; edge detection; Hough transform; extraction of line segments
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
nije evidentirano
Podaci o prilogu
35-38-x.
1998.
objavljeno
Podaci o matičnoj publikaciji
Proc. 21st International Convention MIPRO '98, Vol. II
Petar Biljanović, Karolj Skala, Slobodan Ribarić, Leo Budin
Rijeka: Hrvatska udruga za informacijsku i komunikacijsku tehnologiju, elektroniku i mikroelektroniku - MIPRO
Podaci o skupu
21st International Convention MIPRO'98
predavanje
18.05.1998-22.05.1998
Opatija, Hrvatska