Services on Demand
Journal
Article
Indicators
- Cited by SciELO
- Access statistics
Related links
- Similars in SciELO
Share
Computación y Sistemas
On-line version ISSN 2007-9737Print version ISSN 1405-5546
Abstract
ORTIZ RANGEL, Estela; MEJIA-LAVALLE, Manuel and SOSSA, Humberto. Using Pulse Coupled Neural Networks to Improve Image Filtering Contaminated with Gaussian Noise. Comp. y Sist. [online]. 2017, vol.21, n.2, pp.381-395. ISSN 2007-9737. https://doi.org/10.13053/cys-21-2-2742.
An algorithm called ICM-TM to reduce the effect of Gaussian noise in grayscale images is proposed. It is based on the operation of the well- known Intersection Cortical Model (ICM), a kind of Pulse-Coupled Artificial Neural Network. A Time Matrix (TM) provides information about the iteration when the neuron fires for first time. Each neuron corresponds to a pixel. A selective filtering criteria that combines the median and average operators using the neuron´s activation time is established. The performance of the proposed algorithm is evaluated experimentally with varying degrees of Gaussian noise. Simulation results show that the effectiveness of the method is superior to the median filter, Gaussian filter, Sigma filter, Wiener filter and to the Pulse-Coupled Neural Networks with the Null Interconnections (PCNNNI). Results are mainly provided by the parameter Peak Signal to Noise Ratio (PSNR).
Keywords : Intersection Cortical Model (ICM); Gaussian noise; Wiener filter; Peak Signal to Noise Ratio (PSNR).