Characterization of Signals from Multiscale Edges
IEEE Transactions on Pattern Analysis and Machine Intelligence
On robust signal reconstruction in noisy filter banks
Signal Processing - Content-based image and video retrieval
Extensions of compressed sensing
Signal Processing - Sparse approximations in signal and image processing
Discrete-time wavelet extrema representation: design and consistentreconstruction
IEEE Transactions on Signal Processing
Wavelets and filter banks: theory and design
IEEE Transactions on Signal Processing
Properties of the multiscale maxima and zero-crossingsrepresentations
IEEE Transactions on Signal Processing
Hi-index | 0.08 |
This paper addresses the problem of characterization of the reconstructed non-destructive testing (NDT) signals from a representation in wavelet domain. A novel scheme for characterization of NDT signals from the reconstruction based on wavelet maximum curvature representation has been proposed and stability of the reconstruction from the representation is proved in the paper. It has been established that a signal in the reconstruction set based on wavelet maximum curvature point representation associated with specified non-linear operations removing insignificant information contains complete information for characterization. Data denoising and characterization based on wavelet maximum curvature representation, with a novel dual thresholding scheme, has been presented in this paper. A compression scheme in principal can be designed based on the theory. Analysis of magnetic flux leakage signal has been discussed to illustrate the efficacy of the proposed technique.