Multiple Noise Levels Research Articles

RETNet: Resolution enhancement Transformer network for magnetic particle imaging based on X-space

Magnetic Particle Imaging (MPI) can visualize the concentration distribution of superparamagnetic iron-oxide nanoparticles (SPIONs) in tissues with the advantages of high sensitivity and high temporal resolution. However, the low spatial resolution of MPI limits its application. Increasing the gradient strength of the selection field can improve the resolution of MPI, but also increase power consumption and noise. A feasible and cost-effective method to address this limitation is to reconstruct high gradient (HG) image from low gradient (LG) image using algorithms. Deep learning has been a powerful tool for improving the resolution of medical imaging techniques. In this study, we propose a Resolution Enhancement Transformer Network (RETNet) for reconstructing HG image with high-resolution from LG image with low-resolution as input, avoiding high power consumption and high noise in the system with HG field. RETNet leverages a shallow feature extractor to capture shallow features, a cross-scale-Transformer (CST) to focus on textural features, a residual-swin-Transformer (RST) to focus on structural features, and an image reconstruction module to aggregate these three types of features and reconstruct the HG image. Textural and structural features extracted can ensure the integrity of the details and the realization of high definition in the reconstructed image. Ablation experiments demonstrate the significant contribution of these two modules to reconstruct the HG image. Comparative experiments, including experiments at noise-free and multiple noise levels, confirm the high robustness of RETNet. Simulation, phantom, and in vivo experiments consistently demonstrate that RETNet outperforms competing methods and effectively improves the resolution of MPI.

Bearing Fault Diagnosis Based on Parameter-Optimized Variational Mode Extraction and an Improved One-Dimensional Convolutional Neural Network

When faults occur in rolling bearings, vibration signals exhibit sensitivity to periodic impact components, susceptibility to complex background noise, and non-stationary and nonlinear characteristics. Consequently, using traditional signal processing methods to effectively identify bearing faults presents significant challenges. To facilitate the accurate fault diagnosis of bearings in noisy conditions, we propose an intelligent fault diagnosis method using the Archimedes optimization algorithm (AOA), coupled with a one-dimensional multi-scale residual convolutional neural network (1D-MRCNN), to optimize the variational mode extraction (VME) parameters. First, we introduce a weighted correlated kurtosis (WCK) indicator, formulated using the correlation coefficient and correlated kurtosis as the objective function, to optimize the VME’s center frequency ω and penalty factor α, enabling targeted signal extraction. Second, deep learning techniques are employed to construct the 1D-MRCNN. The neural network then processes the extracted signal for feature extraction and automated fault-type identification. Our simulation results show that the WCK objective function effectively isolates impact components under fault conditions, and our experimental validation confirms that the proposed method accurately identifies diverse fault types across multiple noise levels.

Comparison of methods for intravoxel incoherent motion parameter estimation in the brain from flow-compensated and non-flow-compensated diffusion-encoded data.

Joint analysis of flow-compensated (FC) and non-flow-compensated (NC) diffusion MRI (dMRI) data has been suggested for increased robustness of intravoxel incoherent motion (IVIM) parameter estimation. For this purpose, a set of methods commonly used or previously found useful for IVIM analysis of dMRI data obtained with conventional diffusion encoding were evaluated in healthy human brain. Five methods for joint IVIM analysis of FC and NC dMRI data were compared: (1) direct non-linear least squares fitting, (2) a segmented fitting algorithm with estimation of the diffusion coefficient from higher b-values of NC data, (3) a Bayesian algorithm with uniform prior distributions, (4) a Bayesian algorithm with spatial prior distributions, and (5) a deep learning-based algorithm. Methods were evaluated on brain dMRI data from healthy subjects and simulated data at multiple noise levels. Bipolar diffusion encoding gradients were used with b-values 0-200 s/mm2 and corresponding flow weighting factors 0-2.35 s/mm for NC data and by design 0 for FC data. Data were acquired twice for repeatability analysis. Measurement repeatability as well as estimation bias and variability were at similar levels or better with the Bayesian algorithm with spatial prior distributions and the deep learning-based algorithm for IVIM parameters and , and for the Bayesian algorithm only for , relative to the other methods. A Bayesian algorithm with spatial prior distributions is preferable for joint IVIM analysis of FC and NC dMRI data in the healthy human brain, but deep learning-based algorithms appear promising.

Learning Energy-Based Models in High-Dimensional Spaces with Multiscale Denoising-Score Matching.

Energy-based models (EBMs) assign an unnormalized log probability to data samples. This functionality has a variety of applications, such as sample synthesis, data denoising, sample restoration, outlier detection, Bayesian reasoning and many more. But, the training of EBMs using standard maximum likelihood is extremely slow because it requires sampling from the model distribution. Score matching potentially alleviates this problem. In particular, denoising-score matching has been successfully used to train EBMs. Using noisy data samples with one fixed noise level, these models learn fast and yield good results in data denoising. However, demonstrations of such models in the high-quality sample synthesis of high-dimensional data were lacking. Recently, a paper showed that a generative model trained by denoising-score matching accomplishes excellent sample synthesis when trained with data samples corrupted with multiple levels of noise. Here we provide an analysis and empirical evidence showing that training with multiple noise levels is necessary when the data dimension is high. Leveraging this insight, we propose a novel EBM trained with multiscale denoising-score matching. Our model exhibits a data-generation performance comparable to state-of-the-art techniques such as GANs and sets a new baseline for EBMs. The proposed model also provides density information and performs well on an image-inpainting task.

Outdoor residential noise exposure and sleep in preadolescents from two European birth cohorts

ObjectiveTo examine whether outdoor residential exposure to annual average road traffic and multiple (i.e., road traffic, railway, aircraft, industry) noise levels is related with preadolescents' sleep using maternal-reported and wrist-actigraphy data in two European birth cohorts. MethodsThis cross-sectional study used data of 1245 preadolescents from the Dutch Generation R Study and 232 from the Spanish INMA-Sabadell cohort with a mean age of 12.3 years old. We used noise maps to assess average outdoor road traffic and multiple noise levels (day-evening-night noise indicator, LDEN) at each child's residential address for the year before the sleep assessment. Sleep disturbances were reported by mothers through the Sleep Disturbance Scale for Children and objectively recorded using GeneActiv wrist-actigraphy during seven subsequent days. Linear and Poisson regression models adjusted for several potential confounding variables were performed. ResultsThe mean (SD) exposure to road traffic noise was 53.2 dB (7.3) in the Generation R Study and 61.3 dB (5.9) in the INMA-Sabadell cohort. Exposure to road traffic was related with reduced total sleep time and longer wake after sleep onset (e.g. −3.62 min (95%CI -6.87; −0.37) and 6.88 min (95%CI 1.15; 12.61) per an increase of 10 dB in road traffic noise, respectively) collected by wrist-actigraphy. We observed no association between road traffic exposure and maternal-reported sleep disturbances. Results were similar for multiple noise exposure. ConclusionsThese findings indicate that sleep may be compromised for preadolescents living in areas highly exposed to outdoor residential noise. Future studies using longitudinal designs to further explore these associations during the different stages of sleep development across childhood and adolescence are warranted. Also, wrist-actigraphy measurements which provide more accurate information and may be complementary to the parental- and self-reported data should be considered.

Defects localization using the data fusion of laser Doppler and image correlation vibration measurements

This work proposes a defects localization method based on the combination of laser Doppler vibrometry (LDV) and Digital image correlation (DIC). DIC and LDV are the dominant non-contact vibration measurement methods with a matured theory and high stability. DIC gives a high spatial resolution with a measurement accuracy of several microns, whereas the Multipoint LDV has a limited spatial resolution with high-precision results for the sparse points up to nanometer level. However, the vibration-based defect localization (VBDL) necessitates certain measurement methods with both the high spatial resolution and measurement accuracy. Unfortunately, the accuracy of DIC is not sufficient to detect the data anomalies caused by the structural defects. Nevertheless, LDV cannot achieve a high-precision defect localization owing to the limited spatial resolution. In this study, a nonlinear weighted least-squares fitting has been proposed to improve the measurement accuracy and reliability. A Chebyshev polynomial has been employed as the basis function in the fitting process. The effective signal-to-noise ratio (SNR) and the correlation coefficient are employed as the weight basis of the LDV points and the DIC subsets, respectively. Based on the simulation analysis, the determination of the optimal polynomial order at multiple noise levels and the influence of the LDV points distribution are discussed. The experiment proves that the data fusion method can significantly improve spatial resolution and measurement accuracy. From the data fusion results, the reference-free defects localization in cantilever beams and plates based on residuals have been presented.

Noise-Robust Representation for Fault Identification with Limited Data via Data Augmentation

Noise will be unavoidably present in the data collected from physical environments, regardless of how sophisticated the measurement equipment is. Furthermore, collecting enough faulty data is a challenge since operating industrial machines in faulty modes not only has severe consequences to the machine health, but also may affect collateral machinery critically, from health state point of view. In this paper, we propose a method of denoising with limited data for the purpose of fault identification. In addition, our method is capable of removing multiple levels of noise simultaneously. For this purpose, inspired by unsupervised contrastive learning, we first augment the data with multiple levels of noise. Later, we construct a new feature representation using Contrastive Loss. The last step is building a classifier on top of the learned representation; this classifier can detect various faults in noisy environments. The experiments on the SOUTHEAST UNIVERSITY (SEU) dataset of bearings confirm that our method can simultaneously remove multiple noise levels.

Robustness analysis of denoising neural networks for bone scintigraphy

This paper describes and compares two neural network (NN) based noise filters developed for planar bone scintigraphy.Images taken with a gamma camera typically have a low signal-to-noise ratio and are subject to significant Poisson noise. In our work, we have designed a neural network based noise filter that can be used with planar bone scintigraphy recordings at multiple noise levels, instead of developing a separate network for each noise level.The proposed denoising solution is a convolutional neural network (CNN) inspired by U-NET architecture. A total of 1215 pairs of anterior and posterior patient images were available for training and evaluation during the analysis. The noise-filtering network was trained using bone scintigraphy recordings with real statistics according to the standard protocol, without noise-free recordings. The resulting solution proved to be robust to the noise level of the images within the examined limits.During the evaluation, we compared the performance of the networks to Gaussian and median filters and to the Block-matching and 3D filtering (BM3D) filter. Our presented evaluation method in this article does not require noiseless images and we measured the performance and robustness of our solution on specialized validation sets.We showed that particularly high signal-to-noise ratios can be achieved using noise-filtering neural networks (NNs), which are more robust than the traditional methods and can help diagnosis, especially for images with high noise content.

GRNbenchmark - a web server for benchmarking directed gene regulatory network inference methods.

Accurate inference of gene regulatory networks (GRN) is an essential component of systems biology, and there is a constant development of new inference methods. The most common approach to assess accuracy for publications is to benchmark the new method against a selection of existing algorithms. This often leads to a very limited comparison, potentially biasing the results, which may stem from tuning the benchmark's properties or incorrect application of other methods. These issues can be avoided by a web server with a broad range of data properties and inference algorithms, that makes it easy to perform comprehensive benchmarking of new methods, and provides a more objective assessment. Here we present https://GRNbenchmark.org/ - a new web server for benchmarking GRN inference methods, which provides the user with a set of benchmarks with several datasets, each spanning a range of properties including multiple noise levels. As soon as the web server has performed the benchmarking, the accuracy results are made privately available to the user via interactive summary plots and underlying curves. The user can then download these results for any purpose, and decide whether or not to make them public to share with the community.

Multilevel Noise Contrastive Network for Few-Shot Image Denoising

In recent years, most denoising methods based on deep Convolutional Neural Networks heavily rely on massive noisy-clean image pairs. Collecting massive noisy-clean image pairs is expensive and not practical in real scenes. Currently, few-shot learning has been applied to many areas to cope with the absence data. The few-shot learning, however, in image denoising severely suffers from domain gap problems, including dataset domain gap and feature domain gap, especially for the real noisy images. Therefore, this paper proposes a Multi-level Noise Contrastive Network (MNC-Net) performing few-shot image denoising. MNC-Net consists of two training stages: i) using contrastive learning to self-supervise the training of Multi-level Noise Contrastive Learner (MNCL) on the pure synthetic noisy images with multiple Gaussian noise levels to ease the acute dataset domain gap, and ii) features generated by the MNCL on limited data are fused to the second stage and alleviate the feature domain gap using our proposed denoising network. Specifically, the MNCL consists of a Contrastive Feature Extractor (CFE) and a Contrastive Feature Projector (CFP). MNCL learns the rich and complex content-invariant degradations and general multiple-level noise representations. The denoising network in the second stage is composed of Guider Feature Encoder (GFE) and Adaptive Denoising Decoder (ADD). The GFE uses contrast features from CFE to guide the produced representations on the specific input noisy images. Then, such output features are fed into the ADD to adaptively denoise the noisy images on the corresponding noise distribution. To our best knowledge, this work is the first attempt to jointly use the few-shot learning and contrastive learning in the deep denoising field. Extensive experiments on CBSD68, Kodak24, Set12, SIDD, and DND show that our method achieves promising denoising performances in the absence of data.

SRDRL: A Blind Super-Resolution Framework With Degradation Reconstruction Loss

Recent years have witnessed the remarkable success of deep learning-based single image super-resolution (SISR) methods. However, most of the existing SISR methods assume that low-resolution (LR) images are purely bicubic downsampled from high-resolution (HR) images. Once the actual degradation is not bicubic, their outstanding performance is hard to maintain. Since the real-world image degradation process can be modeled by a combination of downsampling, blurring, and noise, several SR methods have been proposed to super-resolve LR images with multiple blur kernels and noise levels. However, these SR methods require prior knowledge of the degradation process, which is difficult to obtain in practical applications. To address these issues, we propose a degradation reconstruction loss (DRL), which captures the degradation-wise differences between SR images and HR images via a degradation simulator. Empowered by the degradation simulator, the proposed loss, and an efficient SR network, a blind SR framework (SRDRL) without prior knowledge that can handle multiple degradations is formed. Extensive experimental results demonstrate that the proposed SRDRL outperforms the state-of-the-art blind SR methods and denosing+SR methods on multi-degraded datasets. The degradation reconstruction loss can be a plug-and-play loss for existing SR methods to handle multiple degradations. The source code can be found at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/FVL2020/SRDRL</uri> .

A digital image correlation-aided sampling moiré method for high-accurate in-plane displacement measurements

Sampling moiré method (SMM) is widely used for measuring high-accurate displacements and strains. However, it provides accurate measurement only when the displacement is smaller than half of grating pitch due to defects on phase difference calculation. To overcome the shortage of SMM, we improve the performance of SMM by adapting the integer-pixel digital image correlation (DIC). The improved SMM is capable of measuring much larger displacements than half of the grating pitch with high accuracy. Through computer simulation, we analysed the displacement measurement accuracy of the improved SMM under various grating pitches and sampling pitches with multiple noise levels. Then the proper range of sampling pitch was obtained. Two independent experiments with short- and long-range sight-distances, respectively, were conducted to validate the accuracy of the improved SMM successfully. In conclusion, with a mean measurement accuracy of up to 1/350 of a pitch, the improved SMM can be used for measuring high-accurate in-plane displacement whether the displacement is greater than a half of grating pitch or not.

Intelligent vibration signal denoising method based on non-local fully convolutional neural network for rolling bearings

Convolutional neural networks (CNNs) have been widely applied to machinery health management in recent years, whereas research on data-driven denoising methods is relatively limited. Therefore, this paper proposes a robust denoising method based on a non-local fully convolutional neural network (NL-FCNN). In this neural network, the Leaky-ReLU activation function is employed to maintain the information contained in the negative value of the signal. The wide kernel principle is also adopted to enlarge the receptive field. Lastly, the non-local means (NLM) is utilized to construct non-local block (NLB), which could efficiently enhance the long-range dependencies learning ability of the network. This block could enormously improve the denoising performance of the network. Moreover, the proposed method exhibits better performance compared with the three conventional denoising methods under multiple noise levels on the Case Western Reserve University (CWRU) motor bearing dataset. Ultimately, we also demonstrate its application to rolling bearing fault diagnosis.

A reconstruction error-based framework for label noise detection

Label noise is an important data quality issue that negatively impacts machine learning algorithms. For example, label noise has been shown to increase the number of instances required to train effective predictive models. It has also been shown to increase model complexity and decrease model interpretability. In addition, label noise can cause the classification results of a learner to be poor. In this paper, we detect label noise with three unsupervised learners, namely textit{principal component analysis} hbox { (PCA)}, textit{independent component analysis} hbox { (ICA)}, and autoencoders. We evaluate these three learners on a credit card fraud dataset using multiple noise levels, and then compare results to the traditional Tomek links noise filter. Our binary classification approach, which considers label noise instances as anomalies, uniquely uses reconstruction errors for noisy data in order to identify and filter label noise. For detecting noisy instances, we discovered that the autoencoder algorithm was the top performer (highest recall score of 0.90), while Tomek links performed the worst (highest recall score of 0.62).

Deep Learning-Based Denoising for High b-Value at 2000 s/mm2 Diffusion-Weighted Imaging.

Diffusion-weighted imaging (DWI) allows white matter quantification of the white matter tracts of the brain. However, at a high b-value (≥ 2000 s/mm2), DWI acquisition suffers from noise due to longer acquisition times obscuring white matter interpretation. DWI denoising techniques can be used to acquire high b-value DWI without increasing the number of signal averages. We used a residual learning-based convolutional neural network (DnCNN) to reduce noise in high b-value DWI based on the literature review. We applied the proposed denoising method on high b-value, retrospectively collected DWI data with multiple noise levels. Experimental results show an improved image quality after denoising in retrospective DWI (average PSNR before and after denoising: 27.63 ± 1.06 dB and 51.76 ± 1.95 dB, respectively). The prospective DWI included one and two signal averages for denoising. DWI with four signal averages was used as the reference. Representative images show high b-value prospective DW images denoised using the DnCNN. We demonstrated DnCNN for cases of multiple noise levels and signal averages. For the prospective study, the PSNR values for 1-NEX before and after denoising were 27.39 ± 3.75 dB and 27.68 ± 3.75 dB. For 2-NEX, the PSNR values before and after denoising were 27.51 ± 4.18 dB and 27.75 ± 4.05 dB.

Trends and knowledge gaps in field research investigating effects of anthropogenic noise.

Anthropogenic noise is a globally widespread sensory pollutant, recognized as having potentially adverse effects on function, demography, and physiology in wild animals. Human population growth and associated changes in urbanization, transportation, and resource extraction all contribute to anthropogenic noise and are predicted to increase in the coming decades. Wildlife exposure to anthropogenic noise is expected to rise correspondingly. Data collected through field research are uniquely important in advancing understanding of the real-world repercussions of human activity on wildlife. We, therefore, performed a systematic review of literature published from 2008 to 2018 that reported on field investigations of anthropogenic noise impacts. We evaluated publication metrics (e.g., publication rates and journal type), geographical distribution of studies, study subject, and methods used. Research activity increased markedly over the assessment period. However, there was a pronounced geographical bias in research, with most being conducted in North America or Europe, and a notable focus on terrestrial environments. Fewer than one-fifth of terrestrial studies were located in rural areas likely to experience urbanization by 2030, meaning data on ecosystems most likely to be affected by future changes are not being gathered. There was also bias in the taxonomic groups investigated. Most research was conducted on birds and aquatic mammals, whereas terrestrial mammals, reptiles, amphibians, fish, and invertebrates received limited attention. Almost all terrestrial studies examined diurnal species, despite evidence that nocturnality is the prevailing animal activity pattern. Nearly half the studies investigated effects of road or urban noise; the bulk of research was restricted to functional, rather than physiological or demographic consequences. Few experimental studies addressed repercussions of long-term exposure to anthropogenic noise or long-term postexposure effects, and multiple noise types or levels were rarely compared. Tackling these knowledge gaps will be vital for successful management of the effects of increasing wildlife exposure to anthropogenic noise.

Image Denoising Algorithm Based on Local Adaptive Nonlinear Response Diffusion

We describe an image denoising algorithm to improve the denoising performance and the time efficiency, while keeping the edge texture of the image from being blurred. First, the algorithm discriminates the pixel whether is an edge pixel or a noise pixel according to the similarity region values of the neighborhoods, and calculates the adaptive diffusion coefficients to construct an adaptive influence function, which can control the noise and the edges to have different diffusion speeds. Then, a 5-layer feedback convolution network with 16 small filters in each layer generates the reconstructed image through convolution, diffusion and deconvolution. Last, the algorithm uses multiple noise levels training sets to train and optimize the network parameters, and generates the finial image denoising model. The experimental results show that the proposed algorithm using adaptive diffusion coefficients can maintain the edges and the smaller filter can improve the time efficiency. The algorithm also has higher denoising accuracy for images with unknown noise level by training on multiple-noise-level training sets.

Optimal estimation of low-rank factors via feature level data fusion of multiplex signal systems

The design of fusion engines is a subject of great importance in a variety of fields. In this paper, we focus on the problem of linear fusion at the feature level for multiple signal matrices with noises, with the features being extremal eigenvectors. When given multiple similarity matrices, the objective is to find an estimate of the latent signal eigenspace. The concentration result for the inner product of features from different matrix samples is developed, utilizing the random matrix theory. Based on of the theoretical results, we proposed an efficient algorithm, EigFuse, to solve the constrained data-driven optimization problem with different level of noises. Our method is of high efficiency by comparing it with state-of-the-art baseline approaches with multiple noise levels. Comprehensive experiments on several synthetic as well as real-life networks demonstrate our method's superior performance.

Mean gradient descent: an optimization approach for single-shot interferogram analysis.

Complex object wave recovery from a single-shot interference pattern is an important practical problem in interferometry and digital holography. The most popular single-shot interferogram analysis method involves Fourier filtering of the cross term, but this method suffers from poor resolution. To obtain full pixel resolution, it is necessary to model the object wave recovery as an optimization problem. The optimization approach typically involves minimizing a cost function consisting of a data consistency term and one or more constraint terms. Despite its potential performance advantages, this method is not used widely due to several tedious and difficult tasks such as empirical tuning of free parameters. We introduce a new optimization approach, mean gradient descent (MGD), for single-shot interferogram analysis that is simple to implement. MGD does not have any free parameters whose empirical tuning is critical to the object wave recovery. The MGD iteration does not try to achieve minimization of a cost function but instead aims to reach a solution point where the data consistency and the constraint terms balance each other. This is achieved by iteratively progressing the solution in the direction that bisects the descent directions associated with the error and constraint terms. Numerical illustrations are shown for recovery of a step phase object from its corresponding off-axis as well as on-axis interferograms simulated with multiple noise levels. Our results show full pixel resolution as evident from the recovery of the phase step and excellent rms phase accuracy relative to the ground truth phase map. The concept of MGD as presented here can potentially find applications to a wider class of optimization problems.

Anomaly detection for wind turbines based on the reconstruction of condition parameters using stacked denoising autoencoders

This paper proposes an approach for detecting anomalies in a wind turbine (WT) based on multivariate analysis. Firstly, the stacked denoising autoencoders (SDAE) model with moving window and multiple noise levels is developed to reconstruct the normal operating data. The correlations among multivariable and temporal dependency inherent in each variable can be captured simultaneously with moving window processing. Both the coarse-grained and fine-grained features of input data can be learned by training with multiple noise levels. Then, the monitoring indicator is derived from the reconstruction error. The threshold value of monitoring indicator is determined by statistical analysis of the values of the monitoring indicator during normal operation. To identify the most relevant parameter related to the detected anomaly in WT, the contribution degree to which each parameter contributes to the exceedance of the threshold is calculated. Finally, the abnormal level is quantified according to the overlap between test behavior distribution and baseline condition to provide supports for operation and maintenance planning of WT. Demonstration on real SCADA data collected from a wind farm in Eastern China shows that the proposed method is effective for the anomaly detection and early warning of an actual WT.