Approach For Noise Removal Research Articles

Adaptive noise removal approach for restoration of digital images corrupted by multimodal noise

Data smoothing algorithms are commonly applied to reduce the level of noise and eliminate the weak textures contained in digital images. Anisotropic diffusion algorithms form a distinct category of noise removal approaches that implement the smoothing process locally in agreement with image features such as edges that are typically determined by applying diverse partial differential equation (PDE) models. While this approach is opportune since it allows the implementation of feature-preserving data smoothing strategies, the inclusion of the PDE models in the formulation of the data smoothing process compromises the performance of the anisotropic diffusion schemes when applied to data corrupted by non-Gaussian and multimodal image noise. In this study the authors first evaluate the positive aspects related to the inclusion of a multi-scale edge detector based on the generalisation of the Di Zenzo operator into the formulation of the anisotropic diffusion process. Then, a new approach that embeds vector median filtering into discrete implementation of the anisotropic diffusion is introduced to improve the performance of the noise removal algorithm when applied to multimodal noise suppression. To evaluate the performance of the proposed data smoothing strategy, a large number of experiments on various types of digital images corrupted by multimodal noise were conducted.

Image noise removal approach based on subpixel anisotropic diffusion

Image noise removal is the first step in image processing, pattern recognition, and computer vision. A novel algorithm is proposed to remove noise on images based on anisotropic diffusion and subpixel approaches. Firstly, the subpixel difference of an image is defined and the Euler-Lagrange equation is solved. Then, the diffusion equation is solved numerically using an iterative approach. Finally, the noise is removed after the diffusion procedure is finished. The experiments show that the proposed algorithm yields better signal-to-noise ratio and has no blocky effect and less generated speckle noise in the results than the other methods do. In addition, it is easy to implement, takes less iterations, and has low computational complexity.

Signal fluctuations in fMRI data acquired with 2D-EPI and 3D-EPI at 7 Tesla

Segmented three-dimensional echo planar imaging (3D-EPI) provides higher image signal-to-noise ratio (SNR) than standard single-shot two-dimensional echo planar imaging (2D-EPI), but is more sensitive to physiological noise. The aim of this study was to compare physiological noise removal efficiency in single-shot 2D-EPI and segmented 3D-EPI acquired at 7 Tesla. Two approaches were investigated based either on physiological regressors (PR) derived from cardiac and respiratory phases, or on principal component analysis (PCA) using additional resting-state data. Results show that, prior to physiological noise removal, 2D-EPI data had higher temporal SNR (tSNR), while spatial SNR was higher in 3D-EPI. Blood oxygen level dependent (BOLD) sensitivity was similar for both methods. The PR-based approach allowed characterization of relative contributions from different noise sources, confirming significant increases in physiological noise from 2D to 3D prior to correction. Both physiological noise removal approaches produced significant increases in tSNR and BOLD sensitivity, and these increases were larger for 3D-EPI, resulting in higher BOLD sensitivity in the 3D-EPI than in the 2D-EPI data. The PCA-based approach was the most effective correction method, yielding higher tSNR values for 3D-EPI than for 2D-EPI postcorrection.

Noise Removal and Blob Detection Approach for Number Plate Identification

Noise Removal and Blob Detection Approach for Number Plate Identification

Improved polyreference time domain method for modal identification using local or global noise removal techniques

Modal identification involves estimating the modal parameters, such as modal frequencies, damping ratios, and mode shapes, of a structural system from measured data. Under the condition that noisy impulse response signals associated with multiple input and output locations have been measured, the primary objective of this study is to apply the local or global noise removal technique for improving the modal identification based on the polyreference time domain (PTD) method. While the traditional PTD method improves modal parameter estimation by over-specifying the computational model order to absorb noise, this paper proposes an approach using the actual system order as the computational model order and rejecting much noise prior to performing modal parameter estimation algorithms. Two noise removal approaches are investigated: a “local” approach which removes noise from one signal at a time, and a “global” approach which removes the noise of multiple measured signals simultaneously. The numerical investigation in this article is based on experimental measurements from two test setups: a cantilever beam with 3 inputs and 10 outputs, and a hanged plate with 4 inputs and 32 outputs. This paper demonstrates that the proposed noise-rejection method outperforms the traditional noise-absorption PTD method in several crucial aspects.

A novel cost function to estimate parameters of oscillatory biochemical systems

Oscillatory pathways are among the most important classes of biochemical systems with examples ranging from circadian rhythms and cell cycle maintenance. Mathematical modeling of these highly interconnected biochemical networks is needed to meet numerous objectives such as investigating, predicting and controlling the dynamics of these systems. Identifying the kinetic rate parameters is essential for fully modeling these and other biological processes. These kinetic parameters, however, are not usually available from measurements and most of them have to be estimated by parameter fitting techniques. One of the issues with estimating kinetic parameters in oscillatory systems is the irregularities in the least square (LS) cost function surface used to estimate these parameters, which is caused by the periodicity of the measurements. These irregularities result in numerous local minima, which limit the performance of even some of the most robust global optimization algorithms. We proposed a parameter estimation framework to address these issues that integrates temporal information with periodic information embedded in the measurements used to estimate these parameters. This periodic information is used to build a proposed cost function with better surface properties leading to fewer local minima and better performance of global optimization algorithms. We verified for three oscillatory biochemical systems that our proposed cost function results in an increased ability to estimate accurate kinetic parameters as compared to the traditional LS cost function. We combine this cost function with an improved noise removal approach that leverages periodic characteristics embedded in the measurements to effectively reduce noise. The results provide strong evidence on the efficacy of this noise removal approach over the previous commonly used wavelet hard-thresholding noise removal methods. This proposed optimization framework results in more accurate kinetic parameters that will eventually lead to biochemical models that are more precise, predictable, and controllable.

Statistical Approach for Noise Removal in Speech Signals Using LMS, NLMS, Block LMS and RLS Adaptive filters

  Abstract— In this paper the application of different adaptive filters in removing the noise present in the speech signals is presented. To analyze the performance of different adaptive filter family members, the parameters like convergence, output PSNR and CPU consumption time are considered. Results show that NLMS filter shows the better performance in CPU time consumption and output PSNR. Block LMS has the highest Convergence factor among all the members of the adaptive filter family.

A Multiplicative Noise Removal Approach Based on Partial Differential Equation Model

Multiplicative noise, also known as speckle noise, is signal dependent and difficult to remove. Based on a fourth‐order PDE model, this paper proposes a novel approach to remove the multiplicative noise on images. In practice, Fourier transform and logarithm strategy are utilized on the noisy image to convert the convolutional noise into additive noise, so that the noise can be removed by using the traditional additive noise removal algorithm in frequency domain. For noise removal, a new fourth‐order PDE model is developed, which avoids the blocky effects produced by second‐order PDE model and attains better edge‐preserve ability. The performance of the proposed method has been evaluated on the images with both additive and multiplicative noise. Compared with some traditional methods, experimental results show that the proposed method obtains superior performance on different PSNR values and visual quality.

Time‐frequency seismic data de‐noising

ABSTRACTMarine seismic data are always affected by noise. An effective method to handle a broad range of noise problems is a time‐frequency de‐noising algorithm. In this paper we explain details regarding the implementation of such a method. Special emphasis is given to the choice of threshold values, where several different strategies are investigated. In addition we present a number of processing results where time‐frequency de‐noising has been successfully applied to attenuate noise resulting from swell, cavitation, strumming and seismic interference. Our seismic interference noise removal approach applies time‐frequency de‐noising on slowness gathers (τ−p domain). This processing trick represents a novel approach, which efficiently handles certain types of seismic interference noise that otherwise are difficult to attenuate. We show that time‐frequency de‐noising is an effective, amplitude preserving and robust tool that gives superior results compared to many other conventional de‐noising algorithms (for example frequency filtering, τ−p or fx‐prediction). As a background, some of the physical mechanisms responsible for the different types of noise are also explained. Such physical understanding is important because it can provide guidelines for future survey planning and for the actual processing.

A Novel Recursive Algorithm For Detail-Preserving Impulse Noise Removal

Removing impulse noise from digital images without altering image details is a central problem in image processing. There is a huge number of filters in the literature that deal with this problem. In the state-of-the-art filters, the removal of impulse noise is usually carried out in two main steps: detection and estimation. In this paper, we show that we can take any impulse noise removal approach, where the noise detection and estimation processes are separable, and enhance its detail preservation capability. Unlike the classical recursive implementation which performs sequential row-by-row scanning, the proposed algorithm maximizes the contribution of noise-free neighbors in detecting and correcting the noisy pixels. Simulation results using standard test images demonstrate the effectiveness of the proposed algorithm in preserving image details.

Adaptive-hierarchical filtering approach for noise removal

In this paper, our contribution is to propose an adaptive-hierarchical filter that can remove the impulse noise while preserving the details in an image. The global image structure, which is estimated from a set of pyramid images, can be used as prior information in order to apply different filters adaptively. The proposed filter outperforms other methods in that it can make use of both local and global information efficiently. Experimental results show that the proposed method produces better performance than many other well-known methods do.

A HVS-Directed Neural-Network-Based Approach for Salt-Pepper Impulse Noise Removal

In this paper, a novel two-stage noise removal algorithm to deal with salt-pepper impulse noise is proposed. In the first stage, the decision-based recursive adaptive noise-exclusive median filter is applied to remove the noise cleanly and to keep the uncorrupted information as well as possible. In the second stage, the fuzzy decision rules inspired by human visual system (HVS) are proposed to classify image pixels into human perception sensitive class and non-sensitive class. A neural network is proposed to compensate the sensitive regions for image quality enhancement. According to the experimental results, the proposed method is superior to conventional methods in perceptual image quality as well as the clarity and the smoothness in edge regions of the resultant images.

Effective noise separation for magnetotelluric single site data processing using a frequency domain selection scheme

Magnetotelluric (MT) response function estimates can be severely disturbed by the effects of cultural noise. Methods to isolate and remove these disturbances are typically based on time-series editing, robust statistics, remote reference processing, or some combination of the above. Robust remote reference processing can improve the data quality at a local site, but only if synchronous recordings of at least one additional site are available and if electromagnetic noise between these sites is uncorrelated. If these prerequisites are not met, we suggest an alternative approach for noise removal, based on a combination of frequency domain editing with subsequent single site robust processing. The data pre-selection relies on a thorough visual inspection of a variety of statistical parameters such as spectral power densities, coherences, the distribution of response functions and their errors, etc. Extreme outliers and particularly noisy data segments are excluded from further data processing by setting threshold values for individual parameters. Examples from Namibia and Jordan illustrate that this scheme can improve data quality significantly. However, the examples also suggest that it is not possible to establish generally valid rules for selection as they depend strongly on the local noise conditions. High coherence, for example, can indicate a good signal-to-noise ratio or strongly correlated noise. However, we found that strong polarization of the magnetic field channels and the distribution of response function errors are two important parameters for noise detection.

Adaptive Color Image Filtering Based on Center-Weighted Vector Directional Filters

This paper presents a new filtering approach for impulsive noise removal in color images. Incorporating the nonnegative integer weight corresponding to the central sample into the structure of the ...

An Algorithm for Gradient-Free Simulation Optimization using Sampling Control

This article considers a class of continuous-variable optimization problems in which the objective function cannot be computed exactly but must instead be estimated by using, typically, a simulation. As the simulation output is stochastic, iterative optimization algorithms these problems are often augmented with noise-removal features to ensure convergence to an optimal solution. Two broad noise-removal approaches are considered: stepsize-control, which involves a decreasing stepsize, or sampling-control, which involves an increasing sample size. Of these two, stepsize-control predates sampling-control by several decades and is the most commonly used approach, although, as we show, sampling-control has some advantages over stepsize-control. Our particular focus is on optimization problems for which direct gradient estimates are not available, and that instead must be approximated using estimates of the objective function. The classic Kiefer-Wolfowitz algorithm, using stepsize-control, is one such algorithm that estimates a divided difference approximation of the gradient. This article presents a sampling-controlled version of this algorithm that also uses divided difference estimates and has the benefit of being easily parallelizable. A convergence proof and some simulation results are also included. Note that one of the advantages of the sampling-controlled method is that it is better suited to estimators that have small-sample bias but are asymptotically unbiased: because the sampling rate increases, estimator bias is gradually decreased as the number of samples increases.