Signal denoising in engineering problems through the minimum gradient method

Authors:
D. A. G. Vieira;L. Travassos;R. R. Saldanha;Vasile Palade
Affiliations:
Department of Electrical Engineering, Federal University of Minas Gerais, MG, Brazil;SENAI - Centro Integrado de Manufatura e Tecnologia, BA, Brazil;Department of Electrical Engineering, Federal University of Minas Gerais, MG, Brazil;Computing Laboratory, Oxford University, Oxford, UK
Venue:
Neurocomputing
Year:
2009

Citing 11
Cited 1

Neural networks and the bias/variance dilemma

Neural Computation
Rank-deficient and discrete ill-posed problems: numerical aspects of linear inversion

Rank-deficient and discrete ill-posed problems: numerical aspects of linear inversion
Backcalculation of flexible pavement moduli from falling weight deflectometer data using artificial neural networks

Backcalculation of flexible pavement moduli from falling weight deflectometer data using artificial neural networks
Controlling the parallel layer perceptron complexity using a multiobjective learning algorithm

Neural Computing and Applications
The use of ICA in multiplicative noise

Neurocomputing
Robustness of radial basis functions

Neurocomputing
A combined blind source separation and adaptive noise cancellation scheme with potential application in blind acoustic parameter extraction

Neurocomputing
A real-time kepstrum approach to speech enhancement and noise cancellation

Neurocomputing
Efficient estimation of multidimensional regression model using multilayer perceptrons

Neurocomputing
Letters: Noise removal using a novel non-negative sparse coding shrinkage technique

Neurocomputing
Spatially adaptive multiplicative noise image denoising technique

IEEE Transactions on Image Processing

Normalized estimate of adaptive filter for non-minimum phase transfer function component: An illustrative case study

Neurocomputing

Quantified Score

Hi-index	0.01

Visualization

Abstract

This paper applies the minimum gradient method (MGM) to denoise signals in engineering problems. The MGM is a novel technique based on the complexity control, which defines the learning as a bi-objective problem in such a way to find the best trade-off between the empirical risk and the machine complexity. A neural network trained with this method can be used to pre-process data aiming at increasing the signal-to-noise ratio (SNR). After training, the neural network behaves as an adaptive filter which minimizes the cross-validation error. By applying the general singular value decomposition (GSVD), we show the relation between the proposed approach and the Wiener filter. Some results are presented, including a toy example and two complex engineering problems, which prove the effectiveness of the proposed approach.