| Paper: | IMDSP-P15.9 | ||
| Session: | Image Restoration and Denoising | ||
| Time: | Friday, May 19, 10:00 - 12:00 | ||
| Presentation: | Poster | ||
| Topic: | Image and Multidimensional Signal Processing: Morphological Processing | ||
| Title: | Modular Morphological Neural Network Training via Adaptive Genetic Algorithm for Designing Translation Invariant Operators | ||
| Authors: | Ricardo de A. Araújo, Francisco Madeiro, Robson P. de Sousa, Catholic University of Pernambuco, Brazil; Lúcio F. C. Pessoa, Freescale Semiconductor, Inc., United States | ||
| Abstract: | In the present paper, adaptive genetic algorithm (AGA) is applied to training a Modular Morphological Neural Network (MMNN) to design translation invariant operators via Matheron decomposition as well as to Banon and Barrera decomposition. The operators are applied to restoration of images corrupted by salt and pepper noise. The AGA is used to determine the set of structuring elements (neural network weights), architecture of the neural network and number of MMNN modules. Results in terms of noise to signal ratio show that the method proposed in the present work lead to a better operator performance when compared to other methods previously proposed in the literature. | ||
©2018 Conference Management Services, Inc. -||- email: webmaster@icassp2006.org -||- Last updated Friday, August 17, 2012