122 Int. J. Computer Applications in Technology, Vol. 49, No. 2, 2014
Anisotropic and nonlinear diffusion applied to image
enhancement and edge detection
M. Ait Oussous*
Faculty of Science and Technology,
UFR Metrology Automatic and System Analysis,
Laboratory LAMAI, BP 549, Marrakech, Morocco
E-mail: aitoussous.med@gmail.com
*Corresponding author
N. Alaa
Faculty of Science and Technology, Laboratory LAMAI,
University Cadi Ayyad,
P.O. Box 549, Marrakech, Morocco
E-mail: n.alaa@uca.ma
Y. Ait Khouya
Faculty of Science and Technology,
UFR Metrology Automatic and System Analysis,
Laboratory LAMAI, BP 549, Marrakech, Morocco
E-mail: youssef2005133@yahoo.fr
Abstract: In this paper, we suggest a new processing algorithm based on anisotropic diffusion
and nonlinear process for removing noise and image enhancement. The principle of the proposed
algorithm is to apply a Gaussian filter to the image gradient when computing the diffusion coefficient
and to choose the gradient threshold parameter depending on the gradient of the image at each
iteration in the model adopted by Morfu (2009). We compare this algorithm with that of Morfu. A
number of experimental results are described to illustrate its performance and indicate that it is very
efficient in removing noise, image enhancement and edge preserving.
Keywords: image processing; anisotropic; nonlinear reaction-diffusion equations; PDEs; partial
differential equations.
Reference to this paper should be made as follows: Ait Oussous, M., Alaa, N. and Ait Khouya, Y.
(2014) ‘Anisotropic and nonlinear diffusion applied to image enhancement and edge detection’,
Int. J. Computer Applications in Technology, Vol. 49, No. 2, pp.122–133.
Biographical notes: Mohammed Ait Oussous received his Master in Telecommunications and
Computer Networks from the University of Cadi Ayyad, Morocco (2008). He is actually working as
a qualifying high school professor and prepares his PhD in Image Processing. His research interests
include image segmentation, restoration and pattern recognition.
NourEddine Alaa received his Master of Science and PhD from the University of NancyI
France respectively (1986) and (1989). He received the HDR in Applied Mathematics from
the University of Cadi Ayyad, Morocco (1996). He is currently Professor of Modeling and
Scientific Computing at the Faculty of Sciences and Technology of Marrakech. His research
is geared towards non-linear mathematical models and their analysis and digital processing
applications.
Youssef Ait Khouya received his Master in Telecommunications and Computer Networks from the
University of Cadi Ayyad, Morocco (2008). He is currently a PhD student at the Faculty of Sciences
and Technology Marrakech. His research interests include Image Segmentation, Restoration and
Pattern Recognition.
Copyright © 2014 Inderscience Enterprises Ltd.