No Detail Found!

An enhanced active contour for image segmentation with strong noise

Author(s)
OMAR GOUASNOUANE Laboratory of Mathematics, Computer Science and Applications (LMCSA), University Hassan II of Casablanca, FST Mohammedia, P.O Box 146, 20650, Morocco; Laboratory of Fundamental and Applied Mathematics (LMFA), University Hassan II of Casablanca, Ain Chock Faculty of Science, Km 8 Route d'El Jadida, B.P 5366 Maarif, Casablanca, Morocco.
SOUMAYA BOUJENA Laboratory of Fundamental and Applied Mathematics (LMFA), University Hassan II of Casablanca, Ain Chock Faculty of Science, Km 8 Route d'El Jadida, B.P 5366 Maarif, Casablanca, Morocco.
KARIMA KABLI Laboratory of Modelling, Analysis, Control and Statistics (MACS), University Hassan II of Casablanca, Ain Chock Faculty of Science, Km 8 Route d'El Jadida, B.P 5366 Maarif, Casablanca, Morocco.
Abstract
Image segmentation is a useful and important technique in image processing, its purpose is to simplify and/or change the representation of an image into components that are more meaningful and easier to analyze. Nevertheless, even after the evolution of segmentation methods, they unfortunately lead to over-segmentation or to under-segmentation when applied to noisy images. Among different techniques for the segmentation of pixels of interest from an image we focus on segmentation by active contour models. Those models have been widely successfully used, in recent years, for different applications. However, for noisy images it is necessary to perform a denoising preprocessing step for a satisfactory segmentation. In this work we propose a segmentation model wich is a combination between a nonlinear diffusion model for image denoising and classical active contour models which use mean curvature motion techniques. For this we take, in the edge indicator function, the gradient of the image to be segmented which has been denoised by a nonlinear diffusion instead a Gaussian filter. Finally, we will present various experimental results and in particular some examples for which the classical snakes methods based on the gradient are not applicable.
Publication Details
Page(s) 3205-3211
Published Journal: Journal of Theoretical and Applied Information Technology, Volume: 100, Issue: 10, Year: 2022

Request the full text from PASTIC

Keywords
Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Nonlinear Diffusion, Active Contour Models
0

Citations

0

Downloads

86

Views