Intensity difference based neuro-fuzzy system for impulse noisy image restoration: ID-NFS

Author

Kumari, Ratnesh ; Gambhir, Deepak ; Kumar, Vipin

Author_Institution

G.D. Goenka World Inst., Gurgaon, India

fYear

2014

fDate

20-21 Feb. 2014

Firstpage

178

Lastpage

183

Abstract

Restoration of the image corrupted by impulse noise is proposed in this paper. Adaptive neuro fuzzy inference system (ANFIS) has been used to detect the impulse noisy pixels to keep preserve the fine details of the image. Feed-forward neural network with resilient backpropagation method is used to estimate the value of the pixel by which the corrupted pixel is replaced by the estimated value. Proposed method is experimented on some popular test images and the results are shown for visual analysis as well as for quantitative measures.

Keywords

adaptive systems; feedforward neural nets; fuzzy neural nets; image denoising; image restoration; ANFIS; ID-NFS; adaptive neuro fuzzy inference system; corrupted pixel; feedforward neural network; impulse noise; impulse noisy image restoration; impulse noisy pixels; intensity difference based neuro-fuzzy system; resilient backpropagation method; visual analysis; Adaptive systems; Filtering algorithms; Finite impulse response filters; Neural networks; Noise; Noise measurement; Signal processing algorithms; Adaptive Neuro-Fuzzy Inference System; Artificial Neural Network; Image Processing; Impulse Noise; Resilient Backpropagation Algorithm;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing and Integrated Networks (SPIN), 2014 International Conference on

Conference_Location

Noida

Print_ISBN

978-1-4799-2865-1

Type

conf

DOI

10.1109/SPIN.2014.6776944

Filename

6776944