Note: Your custom schedule will not be saved unless you create a new account or login to an existing account.
| Paper: | IMDSP-P10.11 |
| Session: | Image Analysis |
| Time: | Thursday, May 20, 15:30 - 17:30 |
| Presentation: |
Poster |
| Topic: |
Image and Multidimensional Signal Processing: Image and Video Analysis |
| Title: |
A NEW NON-LINEAR EXPONENTIAL 2-D ADAPTIVE FILTER AND ITS APPLICATION IN TEXTURE CHARACTERIZATION |
| Authors: |
Mounir Sayadi; ESSTT | | |
| | Samir Sakrani; ESSTT | | |
| | Farhat Fnaiech; ETS | | |
| | Mohamed Cheriet; ETS | | |
| Abstract: |
We propose in this paper a new non-linear exponential adaptive bi-dimensional (2-D) filter for image modeling. The filter coefficients are updated with the Least Mean Square (LMS) algorithm. Furthermore, the proposed non-linear model is used for texture modeling with a 2-D Auto-Regressive (AR) adaptive model. The characterization efficiency of the proposed exponential model is compared with the 2-D linear AR model updated with the LMS algorithm. The comparison criteria is based on the computation of a characterization rate using the ratio of ''between-class'' variances with respect to ''within-class'' variances of the estimated coefficients. Extensive experiments show that the exponential model coefficients give better results in texture discrimination than those of the linear model, even in a noisy context. |
| |
| Back | |