Pre-processing of Electroencephalography Signals Using Stationary Wavelet Transform-Enhanced Fixed-Point Fast-ICA

2021 ◽  
pp. 387-396
Author(s):  
Aashi Rastogi ◽  
Vikrant Bhateja
Keyword(s):  
Fixed Point ◽  
Wavelet Transform ◽  
Stationary Wavelet Transform

scholarly journals Implementation of Spatial-Scale Domain Based De-Noising Techniques using Different Thresholding

2020 ◽  
Vol 9 (3) ◽  
pp. 3594-3603
Keyword(s):  
Wavelet Transform ◽  
Root Mean Square Error ◽  
Signal To Noise Ratio ◽  
Objective Evaluation ◽  
Contourlet Transform ◽  
Discrete Wavelet ◽  
Mean Square ◽  
Stationary Wavelet Transform ◽  
Signal To Noise ◽  
Different Levels

This paper aims in presenting a thorough comparison of performance and usefulness of multi-resolution based de-noising technique. Multi-resolution based image denoising techniques overcome the limitation of Fourier, spatial, as well as, purely frequency based techniques, as it provides the information of 2-Dimensional (2-D) signal at different levels and scales, which is desirable for image de-noising. The multiresolution based de-noising techniques, namely, Contourlet Transform (CT), Non Sub-sampled Contourlet Transform (NSCT), Stationary Wavelet Transform (SWT) and Discrete Wavelet Transform (DWT), have been selected for the de-noising of camera images. Further, the performance of different denosing techniques have been compared in terms of different noise variances, thresholding techniques and by using well defined metrics, such as Peak Signal-to-Noise Ratio (PSNR) and Root Mean Square Error (RMSE). Analysis of result shows that shift-invariant NSCT technique outperforms the CT, SWT and DWT based de-noising techniques in terms of qualititaive and quantitative objective evaluation

Sign in / Sign up

Export Citation Format

Share Document