Image denoising and fuzziness measures

Authors:
Vincenzo Niola;Giuseppe Quaremba
Affiliations:
Department of Mechanical Engineering for Energetics, University of Naples "Federico II", Napoli, Italy;University of Naples "Federico II", Napoli, Italy
Venue:
NN'06 Proceedings of the 7th WSEAS International Conference on Neural Networks
Year:
2006

Citing 6
Cited 1

Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations

Parallel distributed processing: explorations in the microstructure of cognition, vol. 1: foundations
Neural network, self-organization and object extraction

Pattern Recognition Letters - Special issue on artificial neural networks
Image processing and data analysis: the multiscale approach

Image processing and data analysis: the multiscale approach
Nonlinear wavelet image processing: variational problems, compression, and noise removal through wavelet shrinkage

IEEE Transactions on Image Processing
Efficient and reliable schemes for nonlinear diffusion filtering

IEEE Transactions on Image Processing
Robust anisotropic diffusion

IEEE Transactions on Image Processing

Benchmark database and GUI environment for printed Arabic text recognition research

WSEAS Transactions on Information Science and Applications

Quantified Score

Hi-index	0.00

Visualization

Abstract

A self-organizing multilayer neural network suitable for image processing applications is proposed. The output of the neurons in the output layer has been viewed as a fuzzy set and measures of fuzziness have been used to model the error (instability of the network) of the system. Various mathematical models for calculation of fuzziness of this fuzzy set have been described. The weight updating rules under each model have been developed. This error is then back-propagated to correct weights so that the system error is reduced in the next stage. A comparative study (both analytical and experimental) on the rate of learning for different error measures is also done. Results also show that the rate of learning affects the output, especially when the noise level is very high.