A Simple Fuzzy Method to Remove Mixed Gaussian-Impulsive Noise From Color Images
| Autor: | Almanzor Sapena, Valentín Gregori, Joan-Gerard Camarena, Samuel Morillas |
|---|---|
| Rok vydání: | 2013 |
| Předmět: |
Vector Median Filter
symbols.namesake Artificial Intelligence Image noise Median filter Fuzzy Rule Value noise Mathematics Noise measurement business.industry Applied Mathematics Salt-and-pepper noise Pattern recognition Fuzzy Metrics Gradient noise Color Image Filter Noise Computational Theory and Mathematics Control and Systems Engineering Gaussian noise symbols Artificial intelligence MATEMATICA APLICADA business LENGUAJES Y SISTEMAS INFORMATICOS |
| Zdroj: | RiuNet. Repositorio Institucional de la Universitat Politécnica de Valéncia instname |
| ISSN: | 1941-0034 1063-6706 |
| DOI: | 10.1109/tfuzz.2012.2234754 |
| Popis: | Mixed impulsive and Gaussian noise reduction from digital color images is a challenging task because it is necessary to appropriately process both types of noise that in turn need to be distinguished from the original image structures such as edges and details. Fuzzy theory is useful to build simple, efficient, and effective solutions for this problem. In this paper, we propose a fuzzy method to reduce Gaussian and impulsive noise from color images. Our method uses one only filtering operation: a weighted averaging. A fuzzy rule system is used to assign the weights in the averaging so that both noise types are reduced and image structures are reserved. We provide experimental results to show that the performance of the method is competitive with respect to state-of-the-art filters. This work was supported by the Spanish Ministry of Science and Innovation under Grant MTM2009-12872-C02-01. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|