Adaptive Denoising of Images using Single Layer Neural Network and Normalized Signed Regressor LMS Algorithm

Denoising of images is very important aspect during image enhancement process. In this paper, a novel image denoising model based on adaptive threshold estimation using functional link artificial neural network (FLANN) is proposed. The proposed FLANN structure is trained by Normalised Signed Regress...

Full description

Saved in:
Bibliographic Details
Published in2019 IEEE 5th International Conference for Convergence in Technology (I2CT) pp. 1 - 4
Main Authors Sarangi, Sunita, Rath, Nrusingha Prasad, Kumar Sahoo, Harish
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.03.2019
Subjects
Adaptive threshold selection
Median filter
Normalized Signed Regressor LMS Algorithm
Random-valued impulsive noise
Online AccessGet full text
DOI10.1109/I2CT45611.2019.9033742

Cover

More Information
Summary:Denoising of images is very important aspect during image enhancement process. In this paper, a novel image denoising model based on adaptive threshold estimation using functional link artificial neural network (FLANN) is proposed. The proposed FLANN structure is trained by Normalised Signed Regressor LMS Algorithm (NSRA). The denoising model uses a second-order difference operator and median filtering that work on only degraded pixels. The method is tested with different test images under various noise ratios that the performance of proposed technique provides a significant improvement over other existing methods.
DOI:10.1109/I2CT45611.2019.9033742