|
Title:
|
3D IMAGE SEGMENTATION BY CHAOTIC NEURALNETWORKS: DYNAMIC REDUCTION ON OBJECTS CONTOURS USING EXCITATORY-INHIBITORY PAIRS |
|
Author(s):
|
Abdelouahid Bouhouche , Abdenacer Nafir , Laidi Foughali |
|
ISBN:
|
978-972-8924-39-3 |
|
Editors:
|
António Palma dos Reis, Katherine Blashki and Yingcai Xiao (series editors:Piet Kommers, Pedro Isaías and Nian-Shing Chen) |
|
Year:
|
2007 |
|
Edition:
|
Single |
|
Keywords:
|
Neural networks, Chaos control, 3D image segmentation, Contour detection. |
|
Type:
|
Reflection Paper |
|
First Page:
|
159 |
|
Last Page:
|
163 |
|
Language:
|
English |
|
Cover:
|
|
|
Full Contents:
|
click to dowload 
|
|
Paper Abstract:
|
This paper presents a particular model of chaotic neural network where each element of the network consists of a pair of
two neurons: one excitatory and the other inhibitory. The neural network will be used to detect edge Pixels separating
patches of object surfaces in a 3D scene represented by a range image. The pair of neurons exhibits a chaotic behavior,
observed by transition between different dynamics. The latter can be controlled by an external stimulus.
Coupling pairs of excitatory-inhibitory neurons in a planer structure enables performing several tasks in low and medium
levels of image processing. This paper presents a particular implementation of an excitatory-inhibitory model used to
perform a contour detection task in 3D dense image. |
|
|
|
|
|
|
