Noise Corruption Research Articles

Solving Multi‐Term Time‐Fractional Diffusion Equations Using a High‐Order Fractional Finite Difference Scheme Accompanied by Neural Network Techniques

ABSTRACTThis article is devoted to solving multi‐term time‐fractional diffusion equations with a high‐order finite difference scheme. We approximate the temporal fractional derivatives by the L1‐2‐3 formula, an expansion of the L1 formula with higher accuracy. We employ neural network strategies and automatic differentiation techniques for approximating integer‐order derivatives. In the present work, we propose very important lemmas and theorems to prove the unconditional stability and convergence of the proposed scheme. Extensive numerical experiments for one‐ and two‐dimensional problems, especially for curved boundaries, confirm the theoretical convergence orders. Comparisons with fractional physics‐informed neural networks and finite difference methods in solving the multi‐term time‐fractional diffusion equations demonstrate the superiority of the proposed method. It is also illustrated that the proposed scheme can efficiently and accurately extract the pattern of the solutions even when noise corruption up to ten percent is imposed on the forcing term.

DeScoD-ECG: Deep Score-Based Diffusion Model for ECG Baseline Wander and Noise Removal.

Electrocardiogram (ECG) signals commonly suffer noise interference, such as baseline wander. High-quality and high-fidelity reconstruction of the ECG signals is of great significance to diagnosing cardiovascular diseases. Therefore, this paper proposes a novel ECG baseline wander and noise removal technology. We extended the diffusion model in a conditional manner that was specific to the ECG signals, namely the Deep Score-Based Diffusion model for Electrocardiogram baseline wander and noise removal (DeScoD-ECG). Moreover, we deployed a multi-shots averaging strategy that improved signal reconstructions. We conducted the experiments on the QT Database and the MIT-BIH Noise Stress Test Database to verify the feasibility of the proposed method. Baseline methods are adopted for comparison, including traditional digital filter-based and deep learning-based methods. The quantities evaluation results show that the proposed method obtained outstanding performance on four distance-based similarity metrics with at least 20% overall improvement compared with the best baseline method. This paper demonstrates the state-of-the-art performance of the DeScoD-ECG for ECG baseline wander and noise removal, which has better approximations of the true data distribution and higher stability under extreme noise corruptions. This study is one of the first to extend the conditional diffusion-based generative model for ECG noise removal, and the DeScoD-ECG has the potential to be widely used in biomedical applications.

An accurate paradigm for denoising degraded ultrasound images based on artificial intelligence systems.

Ultrasound images are susceptible to various forms of quality degradation that negatively impact diagnosis. Common degradations include speckle noise, Gaussian noise, salt and pepper noise, and blurring. This research proposes an accurate ultrasound image denoising strategy based on firstly detecting the noise type, then, suitable denoising methods can be applied for each corruption. The technique depends on convolutional neural networks to categorize the type of noise affecting an input ultrasound image. Pre-trained convolutional neural network models including GoogleNet, VGG-19, AlexNet and AlexNet-support vector machine (SVM) are developed and trained to perform this classification. A dataset of 782 numerically generated ultrasound images across different diseases and noise types is utilized for model training and evaluation. Results show AlexNet-SVM achieves the highest accuracy of 99.2% in classifying noise types. The results indicate that, the present technique is considered one of the top-performing models is then applied to real ultrasound images with different noise corruptions to demonstrate efficacy of the proposed detect-then-denoise system. RESEARCH HIGHLIGHTS: Proposes an accurate ultrasound image denoising strategy based on detecting noise type first. Uses pre-trained convolutional neural networks to categorize noise type in input images. Evaluates GoogleNet, VGG-19, AlexNet, and AlexNet-support vector machine (SVM) models on a dataset of 782 synthetic ultrasound images. AlexNet-SVM achieves highest accuracy of 99.2% in classifying noise types. Demonstrates efficacy of the proposed detect-then-denoise system on real ultrasound images.

Multiscale dynamically parallel shrinkage network for fault diagnosis of aviation hydraulic pump and its generalizable applications

Aiming to address the multiscale characteristics and noise corruption problems in the vibration signals of aviation hydraulic pumps, this article develops a novel Multiscale Dynamically Parallel Shrinkage Network (MDPSN) to learn complementary and rich fault-related multiscale features, with the ultimate goal of yielding higher diagnostic accuracy. One significant property is the development of a novel dynamically parallel shrinking module (DPSM) that adaptively generates independent soft thresholds for different scales, effectively shrinking noise-related features to zeros. On one hand, DPSM aggregates and interacts with features at all scales to construct a global feature representation containing richer fault-related information, which is served as the foundation for soft thresholding generation, significantly improving the accuracy and rationality of the generated thresholds. On the other hand, DPSM can adaptively generate individual soft threshold for each scale, allowing each scale to use an independent threshold tailored to its own characteristic to eliminate noise-related information. This avoids the issues of over-denoising or under-denoising caused by the uniform application of thresholds across all scales. Finally, the effectiveness of MDPSN is validated by a series of experiment comparisons on an aviation hydraulic pump dataset and two bearing datasets with various types of noise. The experimental results demonstrate that MDPSN achieves superior diagnostic accuracy compared to five other comparison methods.

Low‐light visibility enhancement for improving visual surveillance in intelligent waterborne transportation systems

AbstractUnder low‐light imaging conditions, visual scenes captured by intelligent waterborne transportation systems often suffer from low‐intensity illumination and noise corruption. The visual quality degradation would lead to negative effects in maritime surveillance, e.g., vessel detection, positioning and tracking, etc. To restore the low‐light images, we develop an effective visibility enhancement method, which contains a coarse‐to‐fine framework of spatially‐smooth illumination estimation. In particular, the refined illumination is effectively generated by optimizing a novel structure‐preserving variational model on the coarse version, estimated through the Max‐RGB method. The proposed variational model has the capacity of suppressing the textural details while preserving the main structures in the refined illumination map. To further boost imaging performance, the refined illumination is adjusted through the Gamma correction to increase brightness in dark regions. We then estimate the refined reflection map by implementing the joint denoising and detail boosting strategies on the original reflection. In this work, the original reflection is yielded by dividing the input image using the refined illumination. We finally produce the enhanced image by multiplying the adjusted illumination and the refined reflection. Experiments on synthetic and realistic datasets illustrate that our method can achieve comparable results to the state‐of‐the‐art techniques under different imaging conditions.

Re-evaluating the scoring criteria of the Interacoustics Sera Neonatal Hearing Screener

Abstract Introduction: Newborn hearing screeners via otoacoustic emissions (OAEs) have become the standard in audiological clinical practice. Despite the application of OAE protocols for more than 3 decades, there has been no consensus on the standardization of the recorded responses. Usually, manufacturers based the scoring criteria of the OAE responses on small scale samples and in many cases on data published in the literature. Since there are no standards in the OAE probe construction, the resulted OAE responses are quite different between OAE screeners. The project evaluated the scoring criteria used in the Interacoustics Sera OAE screener. Methods: Eight hundred and eleven infants were initially screened by the Accuscreen OAE screener for a Pass or a Fail outcome. Seven hundred and ninety-two infants resulted as Pass were re-evaluated with the Sera transient-evoked OAE (TEOAE) and distortion-product OAE (DPOAE) protocols. Results: In terms of screening efficiency, the DPOAE protocol performed better, showing results very similar to the Accuscreen performance (Fail %: 2.4 vs. 2.3). The data showed that the recordings of the TEOAE responses were more prone to noise corruption. Using 10th percentile normative distribution values, DPOAE-optimized scoring criteria were estimated as 3.1, 7.9, 9.9, and 8.9 dB for the tested frequencies of 2.0, 3.0, 4.0, and 5.0 kHz. Unfortunately, the TEOAE signal-to-noise (S/N) data resulted as too noisy (despite relatively large TEOAE responses); therefore, the TEOAE responses were considered as scoring criteria candidates. The latter were estimated as 4.61, 6.35, 7, 1 and 3, 35 dB at 1.4, 2.0, 2.8, 4.0 kHz. Conclusion: It is possible to improve the screening efficiency of the DPOAE Sera protocol using optimized scoring criteria, aiming the correct identification of at least 90% of all normal hearing subjects. The Sera TEOAE protocol needs further investigation to identify the source of the noise, which lowers significantly the reported S/N ratios. The collected OAE data confirm that there are differences among the responses obtained by various screeners; therefore, an OAE probe standardization is necessary.

Multimodal identification of the mouse brain using simultaneous Ca 2+ imaging and fMRI.

Individual differences in neuroimaging are of interest to clinical and cognitive neuroscientists based on their potential for guiding the personalized treatment of various heterogeneous neurological conditions and diseases. Despite many advantages, the workhorse in this arena, BOLD (blood-oxygen-level-dependent) functional magnetic resonance imaging (fMRI) suffers from low spatiotemporal resolution and specificity as well as a propensity for noise and spurious signal corruption. To better understand individual differences in BOLD-fMRI data, we can use animal models where fMRI, alongside complementary but more invasive contrasts, can be accessed. Here, we apply simultaneous wide-field fluorescence calcium imaging and BOLD-fMRI in mice to interrogate individual differences using a connectome-based identification framework adopted from the human fMRI literature. This approach yields high spatiotemporal resolution cell-type specific signals (here, from glia, excitatory, as well as inhibitory interneurons) from the whole cortex. We found mouse multimodal connectome- based identification to be successful and explored various features of these data.

SAR Target Recognition Method based on Adaptive Weighted Decision Fusion of Deep Features

Background: This paper proposes a synthetic aperture radar (SAR) target recognition method based on adaptive weighted decision fusion of multi-level deep features. Methods: The trained ResNet-18 is employed to extract multi-level deep features from SAR images. Afterwards, based on the joint sparse representation (JSR) model, the multi-level deep features are represented to obtain the corresponding reconstruction error vectors. Considering the differences in the abilities of different levels of features to distinguish the target, the reconstruction error vectors are analyzed based on entropy theory, and their corresponding weights are adaptively obtained. Finally, the fused reconstruction error result is obtained through adaptively weighted fusion, and the target label is determined accordingly. Results: Experiments are conducted on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset under different conditions, and the proposed method is compared with published methods, including multi-feature decision fusion, JSR-based decision fusion and other types of ResNets. Conclusion: The experimental results under standard operating condition (SOC) and extended operating conditions (EOCs) including depression angle variance and noise corruption validate the advantages of the proposed method.

Uncertainty-guided label correction with wavelet-transformed discriminative representation enhancement

Label noises, categorized into closed-set noise and open-set noise, are prevalent in real-world scenarios and can seriously hinder the generalization ability of models. Identifying noise is challenging because noisy samples closely resemble true positives. Existing approaches often assume a single noise source, oversimplify closed-set noise, or treat open-set noise as toxic and eliminate it, resulting in limited practical effects. To address these issues, we present a novel approach named uncertainty-guided label correction with wavelet-transformed discriminative representation enhancement (Ultra), designed to mitigate the effects of mixed noise. Specifically, our approach considers a more practical noise setting. To achieve robust mixed-noise identification, we initially look into a learnable wavelet filter for obtaining discriminative features and filtering spurious cues automatically at the representation level. Subsequently, we introduce a two-fold uncertainty estimation to stably locate noise within the corrupted supervised signal level. These insights pave the way for a simple yet potent label correction technique, enabling comprehensive utilization of open-set noise, which can be rendered non-toxic in a specific manner, in contrast to harmful closed-set noise. Experimental validation on datasets with synthetic mixed noise, web noise corruption, and a real-world dataset confirms the effectiveness and generality of Ultra. Furthermore, our approach enhances the application of efficient techniques (e.g., supervised contrastive learning) within label noise scenarios.

The ringed residual u-net with non-natural regions feature for image splicing forgery detection and localization

Recently, with the emergence of many image editing tools (photoshop, Topaz studio, etc.), the authenticity of images has been severely challenged. However, the performance of some existing traditional feature extraction methods and detection methods based on convolutional neural network (CNN) is poor, and the information provided by the features extracted from the network is limited and single. In this paper, an end-to-end ringed residual U-Net is proposed to detect image splicing forgery by blending features of non-natural regions. Some regions with significant differences from the image background are defined as non-natural regions(such as the irregular border at the splicing of images). In this paper, a feature enhancement module for non-natural regions is constructed, which the image through the pooling of four different scales, and these features are then combined with the original image and input to the backbone network for processing, aiming to highlight regions of the image that differ significantly from the background. Therefore, after adding the feature enhancement module for non-natural regions to the end-to-end ring residual U-Net, more attention will be paid to the tampering regions in the feature extraction stage, image manipulation detection and localization will also become more accurate. Compared with some mainstream methods, this method achieves better performance on the three standard datasets(CASIA2.0, NIST2016, COLUMBIA). In addition, it has excellent robustness under JPEG compression attack and noise corruption attack.

Relaxation linear regression with spectral–spatial constrained locality adaptive regularization for hyperspectral image classification

AbstractRecently, relaxation linear regression has received increasing attention in image analysis. However, the current relaxation linear regression methods fail to consider the local geometrical structure and spatial information while they are applied for hyperspectral image (HSI) classification. To address the above problems, this letter proposes a novel relaxation linear regression with spectral–spatial constrained locality adaptive regularization (SSLA‐RLR) method for HSI classification. The SSLA‐RLR method not only integrates the locality adaptive graph with relaxation linear regression to adaptively exploit the local geometrical structure for relieving the side effects of noise corruptions, but also takes the spatial correlations between each data point and its neighbours into consideration via spectral–spatial constraint. Finally, extensive experiments are conducted on three benchmark HSI datasets to demonstrate the effectiveness of the proposed method.

A novel robust multivariate Laplace distribution-based distributed consensus information fusion

The multivariate Laplace distribution (MLD) possesses a lesser set of prior parameters compared to the Student’s t-distribution, thus simplifying the modeling of measurement data subject to heavy-tailed noise corruption. This short communication introduces a novel distributed hybrid consensus fusion algorithm utilizing the MLD modeling and variational Bayesian approach. Firstly, we formulate the measurement noise associated with individual sensor nodes and introduce the hierarchical Gaussian form of MLD to enhance the feasibility of variational inference. Secondly, we opt for the inverse-Wishart (IW) distribution to serve as the conjugate prior to the positive definite matrix, enabling adaptive parameter learning within the MLD context. As a result, a novel hierarchical Gaussian state–space model (SSM) is established for multi-sensors network. Subsequently, we derive a novel local information filter to acquire higher precision information and innovation vector–matrix pairs, outperforming the Student’s t distribution-based filter. These derivations are grounded in the new hierarchical SSM, and the vector–matrix pairs aid in information fusion across the network. Lastly, we present a distributed robust consensus filter for fusing local information using a hybrid consensus approach that devoid of a central fusion node. Simulation results demonstrate that the proposed consensus filter can obtain better estimation accuracy under the interference of heavy-tailed measurement outliers.

Scale and pattern adaptive local binary pattern for texture classification

Local binary pattern (LBP) with a fixed sampling template is sensitive to scale changes. Furthermore, under rotation changes or noise corruptions, one uniform LBP pattern can be corrupted to fall into a non-uniform pattern which loses its discrimination power to describe the corresponding texture feature. To overcome these two main drawbacks, we propose a scale and pattern adaptive local binary pattern (SPALBP). Firstly, in the gradient-based sampling radius adaptive scheme, eight directional adaptive sampling radius of each center pixel can be obtained by using its eight Kirsch gradient values. Secondly, in the noise and rotation robust neighborhood sampling scheme, three neighborhood sampling templates are used to extract three kinds of averaging neighborhood pixels. Thirdly, for each center pixel, three kinds of LBPriu2 patterns can be extracted by sampling these three kinds of averaging neighborhood pixels along eight directional adaptive sampling radius. Finally, an optimal SPALBP uniform pattern can be adaptively selected from these three LBPriu2 patterns. Hence, all SPALBP patterns show more robustness against scale changes, rotation changes and noise corruptions. Extensive experiments are conducted on four standard texture databases: Outex, UIUC, CUReT and XU_HR. Comparing with state-of-the-art LBP-based variants, the proposed SPALBP method consistently shows superior performance both in dramatic environment changes and high-levels of noise conditions, meanwhile it maintains a lower texture feature dimension.

Nonparametric estimation for SDE with sparsely sampled paths: An FDA perspective

We consider the problem of nonparametric estimation of the drift and diffusion coefficients of a Stochastic Differential Equation (SDE), based on n independent replicates Xi(t):t∈[0,1]1≤i≤n, observed sparsely and irregularly on the unit interval, and subject to additive noise corruption. By sparse we intend to mean that the number of measurements per path can be arbitrary (as small as two), and can remain constant with respect to n. We focus on time-inhomogeneous SDE of the form dX(t)=μ(t)X(t)αdt+σ(t)X(t)βdB(t), where α∈{0,1} and β∈{0,1/2,1}, which includes prominent examples such as Brownian motion, Ornstein–Uhlenbeck process, geometric Brownian motion and Brownian bridge. Our estimators are constructed by relating the local (drift/diffusion) parameters of the SDE to its global parameters (mean/covariance, and their derivatives) by means of an apparently novel Partial Differential Equation (PDE). This allows us to use methods inspired by functional data analysis, and pool information across the sparsely measured paths. The methodology we develop is fully non-parametric and avoids any functional form specification on the time-dependency of either the drift function or the diffusion function. We establish almost sure uniform asymptotic convergence rates of the proposed estimators as the number of observed paths n grows to infinity. Our rates are non-asymptotic in the number of measurements per path, explicitly reflecting how different sampling frequency might affect the speed of convergence. Our framework suggests possible further fruitful interactions between FDA and SDE methods in problems with replication.

Identification of Uncertain Parameter in Flight Vehicle Using Physics-Informed Deep Learning

This paper presents the estimation method for uncertain parameters in flight vehicles, especially missile systems, based on physics-informed neural networks (PINNs) augmented with a novel integration-based loss. The proposed method identifies four types of structured uncertainty: burnout time, rocket motor tilt angle, location of the center of pressure, and control fin bias, which significantly affect the missile performance. In the estimation framework, as neural networks (NNs) are updated, these uncertainties are also identified simultaneously because they are also included in the structure of NNs. After testing 100 simulation data, the average estimation errors are within 1% of the mean value for each type of uncertainty. The methodology is able to identify the parameters despite noise corruption in the time-series data. Compared with the conventional PINNs, adding the new loss based on the integration of differential equations yields a more reliable estimation performance for all types of uncertainty. This approach can be effective for complex systems and ill-posed inverse problems, which makes it applicable to other aerospace systems.

Robust Multi-View Clustering Through Partition Integration on Stiefel Manifold

Multi-view clustering aims at integrating information from different views to improve clustering performance. Recent methods integrate multiple view-specific partition matrices to seek a consensus one and have demonstrated promising clustering performance in various applications. However, the clustering performance of such methods heavily relies on the consensus partition matrix estimated by the arithmetic mean in Euclidean space and thus is highly susceptible to noise corruption. To this end, this paper proposes to learn a consensus partition matrix through the geometric mean on the manifold to achieve robust clustering. Specifically, the multiple view-specific partition matrices can be regarded as points residing in the Stiefel manifold and enable a manifold-based integration. Consequently, the view-specific partition matrices are integrated by estimating a consensus partition matrix as the center point on the Stiefel manifold. Such a partition integration boils down to the Fréchet mean problem on a manifold, which is solved by the intrinsic manifold-based optimization and proves effective in providing a more robust estimation against noise. Experimental results on seven benchmark datasets demonstrate the effectiveness and noise-robustness of our proposed method in comparison to eight competitive methods.

State-of-health estimation for lithium-ion battery using model-based feature optimization and deep extreme learning machine

Accurate and reliable estimation of state of health (SOH) is essential for the safe and efficient operation of battery energy storages. However, random charging/discharging behaviors and ambient conditions complicate the online SOH estimation. Aiming at an accurate and robust online SOH estimation for lithium-ion batteries, this paper proposes a SOH estimation method using model-based feature optimization and improved machine learning. First, an empirical model-based voltage reconstruction method is proposed to reconstruct the voltage curve for solving disturbance-free different time (DT) curves under measurement noises and high C-rates and extract model-associated health features (HFs). Then, an optimal charging voltage window (CVW) determination method is proposed, which extracts the CVW-associated HFs by determining the optimal CVW from partial charging voltage ranges. Finally, a set of informative and multi-attribute HFs are extracted to train an improved deep extreme learning machine (DELM) mode for online SOH estimation. The proposed method is validated based on three datasets of batteries with different materials. The results demonstrate high accuracy and strong robustness to partial voltage range, noise corruption, and ambient temperature fluctuation.

Fuzzy Salt and Pepper Noise Removing for Color Images

Within the field of digital image processing applications, observed images frequently exposed to noise corruption stemming from image acquisition or transmission processes. This noise degradation reduces image quality and yields unfavorable outcomes in subsequent processing stages (e.g., segmentation, pattern recognition, and enhancement). Consequently, the mitigation of noise in images assumes paramount importance in the domain of image processing. This study introduces an algorithm centered around fuzzy logic for removing impulse noise from color images. The efficiency of the proposed algorithms is assessed by comparing their performance against various noise reduction methods. Objective metrics, namely peak signal-to-noise ratio and mean square error, substantiate that the proposed algorithms yield commendable outcomes in noise reduction and the preservation of intricate image details across a wide spectrum of noise densities

Image Enhancement using Adaptive Median Filter

Impulse noise frequently taints digital images as a result of errors produced by noisy sensors, faults during the analog-to-digital conversion process, and errors produced in communication channels. This persistent bug modifies the brightness of some pixels while keeping the brightness of other pixels unaltered. We looked at the Median Filter, Weighted Median Filter, and Multi-Stage Median Filter to lessen impulsive noise and enhance the quality of the impacted image. These filters enhance the image quality by removing or successfully suppressing impulse noise in the image while maintaining the information about the image's edges. In MATLAB simulations, the effectiveness of these filters was assessed on a picture that has undergone different types of impulse noise corruption (Salt-and-Pepper and Gaussian noise).

Hybrid Pseudorandom Sequence for Broadband Impedance Measurements of Lithium-Ion Batteries

Broadband battery impedance spectroscopy brings valuable information that facilitates states estimation and fault diagnosis, which are of extensive research interests at present. pseudo random sequence (PRS) provides a promising tool for realizing online battery impedance measurements, yet it suffers from an insufficient signal-to-noise ratio (SNR). Motivated by this, this paper proposes a novel hybrid pseudo random sequence (HPRS) to boost the signal power in the low-frequency region, where the measurement data is exceptionally robust against noise corruption. The disturbances are further suppressed by applying a specially designed adaptive bilateral impedance filter (ABIF), which considers signal power content and impedance measurements&#x0027; distribution enabling a nearly unbiased result. The feasibility of the proposed HPRS signal is verified experimentally by comparing it with the typical PRS signal and step signal. The results suggest superior broadband impedance measurements&#x0027; accuracy and robustness under different ambient temperatures and state of charges (SOCs) and the maximum normalized root mean square error (NRMSE) is limited to 0.77<inline-formula><tex-math notation="LaTeX">$\%$</tex-math></inline-formula>.