|
A NONLINEAR ANISOTROPIC DIFFUSION-BASED EDGE DETECTION SCHEME
|
|
|
T. Barbu
|
|
|
||
|
|
|
|
1314-2704
|
|
|
||
|
English
|
|
|
17
|
|
|
21
|
|
|
|
|
|
||
|
A novel image edge detection technique is proposed in this paper. The considered boundary detection approach uses a nonlinear parabolic partial differential equation (PDE) - based model. We develop a nonlinear anisotropic diffusion scheme consisting in a second-order parabolic equation and some boundary conditions. An effective edge-stopping function is constructed for this PDE model. Then, we construct an explicit iterative numerical approximation scheme for this model, by using the finite-difference method. This numerical discretization scheme is applied on the analyzed image and two results, achieved after different iterations, are considered. The image representing the absolute difference between those two states is then determined and a thresholding procedure is then applied to it. Some morphology-based operations are applied on that image, its edges being determined from these morphologically processes. Numerous experiments and method comparison are performed using the proposed PDE-based edge detection framework.
|
|
|
conference
|
|
|
||
|
||
|
17th International Multidisciplinary Scientific GeoConference SGEM 2017
|
|
|
17th International Multidisciplinary Scientific GeoConference SGEM 2017, 29 June - 5 July, 2017
|
|
|
Proceedings Paper
|
|
|
STEF92 Technology
|
|
|
International Multidisciplinary Scientific GeoConference-SGEM
|
|
|
Bulgarian Acad Sci; Acad Sci Czech Republ; Latvian Acad Sci; Polish Acad Sci; Russian Acad Sci; Serbian Acad Sci & Arts; Slovak Acad Sci; Natl Acad Sci Ukraine; Natl Acad Sci Armenia; Sci Council Japan; World Acad Sci; European Acad Sci, Arts & Letters; Ac
|
|
|
21-28
|
|
|
29 June - 5 July, 2017
|
|
|
website
|
|
|
cdrom
|
|
|
2932
|
|
|
edge detection; PDE model; anisotropic diffusion; numerical approximation scheme; morphology-based operation
|
|