Author/Authors :
SAĞIROĞLU, Şeref Gazi Üniversitesi - Engineering Faculty - Computer Engineering Department, Turkey , BEŞDOK, Erkan Erciyes University - Engineering Faculty - Geodesy Photogrammetry, Engineering Department, Turkey
Title Of Article :
A Novel Approach for Image Denoising Based on Artificial Neural Networks
Abstract :
This study presents a novel approach based on artificial neural networks (ANNs) to remove noises from defected images. ANNs were trained with two different learning algorithms, Levenberg-Marquardt and Extended-Delta-Bar-Delta, for speeding up the training and feedforward calculation processes. The restored results were also compared to the classical techniques, FFT, Wiener+Median filtering and wavelet denoising. The results were shown that the proposed novel neural model provides simplicity and accuracy to remove noises from defected images without estimating any mathematical model than the others.
NaturalLanguageKeyword :
Image Restoration , Wavelet , Wiener , Median , FFT , Artificial neural networks
JournalTitle :
Journal Of Polytechnic
