Authors:
A. A. Saraiva
1
;
M. S. de Oliveira
2
;
J. V. M. Sousa
1
;
N. M. Fonseca Ferreira
3
;
Antonio Valente
4
;
5
and
Salviano Soares
4
Affiliations:
1
State University of Piaui, Piripiri, Piaui, Brazil
;
2
University of São Paulo, Ribeirão Preto, São Paulo, Brazil
;
3
Institute of Engineering of Coimbra / Polytechnic Institute of Coimbra, Portugal. Knowledge Engineering and Decision-Support Research Center (GECAD) of the Institute of Engineering, Polytechnic Institute of Porto INESC Technology and Science (INESC TEC), Porto, Portugal
;
4
UTAD University, Vila Real, Portugal
;
5
INESC Technology and Science (INESC TEC), Porto, Portugal, School of Science and Technology, UTAD University, Vila Real, Portugal
Keyword(s):
Medical Images, Genetic Algorithm, Noise Removal.
Abstract:
The techniques of image filtering have undergone an explosive growth in the last years to make new advances and challenges. This is due to the fact, among several other reasons, the increase of the volume of images coming from several sources. Digital images have been used for a variety of purposes, from the storage of souvenirs to accurate medical exams. However, Images may be corrupted due to several factors. The challenge of suppression or noise attenuation has led to the search for improved techniques in order to preserve important characteristics of the image, but, on the other hand, there is no solution available to completely solve the problem, boosting the production of the work proposed here. In this paper proposes a method for noise attenuation in computed tomography images using a hybrid genetic algorithm, the proposed method seeks to optimize the results in the space of solutions composed by a series of techniques of noise filtering. At the end the proposed method is comp
ared statistically with two other competing methods and after the resulting filtered images are shown.
(More)