Showing papers on "Noise reduction published in 2023"

Denoising Aggregation of Graph Neural Networks by Using Principal Component Analysis

Abstract: To avoid the overfitting phenomenon that appeared in performing graph neural networks (GNNs) on test examples, the feature encoding scheme of GNNs usually introduces the dropout procedure. However, after learning latent node representations under this scheme, Gaussian noise produced by the dropout operation is inevitably transmitted into the next neighborhood aggregation step, which necessarily hampers the unbiased aggregation ability of GNN models. To address this issue, in this article, we present a novel aggregator, denoising aggregation (DNAG), which utilizes principal component analysis (PCA) to preserve the aggregated real signals from neighboring features and simultaneously filter out the Gaussian noise. The idea is different from using PCA on traditional applications to reduce the feature dimension. We regard PCA as an aggregator to compress the neighboring node features to have better expressive denoising power. We propose new training architectures to simplify the intensive computation of PCA in DNAG. Numerical experiments show the apparent superiority of the proposed DNAG models in gaining more denoising capability and achieving the state of the art for a set of predictive tasks on several graph-structured datasets.

Research on Image Denoising in Edge Detection Based on Wavelet Transform

Abstract: Photographing images is used as a common detection tool during the process of bridge maintenance. The edges in an image can provide a lot of valuable information, but the detection and extraction of edge details are often affected by the image noise. This study proposes an algorithm for wavelet transform to denoise the image before edge detection, which can improve the signal-to-noise ratio of the image and retain as much edge information as possible. In this study, four wavelet functions and four decomposition levels are used to decompose the image, filter the coefficients and reconstruct the image. The PSNR and MSE of the denoised images were compared, and the results showed that the sym5 wavelet function with three-level decomposition has the best overall denoising performance, in which the PSNR and MSE of the denoised images were 23.48 dB and 299.49, respectively. In this study, the canny algorithm was used to detect the edges of the images, and the detection results visually demonstrate the difference between before and after denoising. In order to further evaluate the denoising performance, this study also performed edge detection on images processed by both wavelet transform and the current widely used Gaussian filter, and it calculated the Pratt quality factor of the edge detection results, which were 0.53 and 0.47, respectively. This indicates that the use of wavelet transform to remove noise is more beneficial to the improvement of the subsequent edge detection results.

A Self-Supervised Deep Learning Approach for Blind Denoising and Waveform Coherence Enhancement in Distributed Acoustic Sensing Data

Abstract: Fiber-optic distributed acoustic sensing (DAS) is an emerging technology for vibration measurements with numerous applications in seismic signal analysis, including microseismicity detection, ambient noise tomography, earthquake source characterization, and active source seismology. Using laser-pulse techniques, DAS turns (commercial) fiber-optic cables into seismic arrays with a spatial sampling density of the order of meters and a time sampling rate up to one thousand Hertz. The versatility of DAS enables dense instrumentation of traditionally inaccessible domains, such as urban, glaciated, and submarine environments. This in turn opens up novel applications such as traffic density monitoring and maritime vessel tracking. However, these new environments also introduce new challenges in handling various types of recorded noise, impeding the application of traditional data analysis workflows. In order to tackle the challenges posed by noise, new denoising techniques need to be explored that are tailored to DAS. In this work, we propose a Deep Learning approach that leverages the spatial density of DAS measurements to remove spatially incoherent noise with unknown characteristics. This approach is entirely self-supervised, so no noise-free ground truth is required, and it makes no assumptions regarding the noise characteristics other than that it is spatio-temporally incoherent. We apply our approach to both synthetic and real-world DAS data to demonstrate its excellent performance, even when the signals of interest are well below the noise level. Our proposed methods can be readily incorporated into conventional data processing workflows to facilitate subsequent seismological analyses.

RMCHN: A Residual Modular Cascaded Heterogeneous Network for Noise Suppression in DAS-VSP Records

Abstract: Distributed optical fiber acoustic sensing (DAS) is an emerging acquisition technology in seismic exploration. However, DAS records are always affected by the complex background noise, resulting in a low signal-to-noise ratio (SNR). In addition, the DAS background noise has different properties from the noise existing in conventional seismic data. Thus, conventional denoising methods may degrade the record when dealing with complex DAS data. To improve the denoising capability, a novel denoising network, called residual modular cascaded heterogeneous network (RMCHN), is proposed. In general, the network is based on the idea of heterogeneous convolution and modular convolutional neural networks. Specifically, different modules are designed to extract the discriminatory features of the DAS data through effective information integration. On this basis, heterogeneous convolution combined with long and short path feature learning strategy is employed to fuse the captured features, thereby improving the feature expression capability and avoiding the information loss. Both synthetic and field denoising results indicate that RMCHN can suppress the DAS background noise with excellent performance in signal restoration, even for the weak signals form deep strata.

DeepSeg: Deep Segmental Denoising Neural Network for Seismic Data

Abstract: Noise attenuation is a crucial phase in seismic signal processing. Enhancing the signal-to-noise ratio (SNR) of registered seismic signals improves subsequent processing and, eventually, data analysis and interpretation. In this work, a novel noise reduction framework based on an intelligent deep convolutional neural network is proposed that works on segments of the time-frequency domain and, hence named as DeepSeg. The proposed network is efficient in learning sparse representation of the data simultaneously in the time-frequency domain and adaptively capturing seismic signals corrupted with noise. DeepSeg is able to achieve impressive denoising performance even when seismic signal shares common frequency band with noise. The proposed approach properly tackles a variety of correlated (color) and uncorrelated noise, and other nonseismic signals. DeepSeg can boost the SNR considerably even in extremely noisy environments with minimal changes to the signal of interest. The effectiveness of the proposed methodology is demonstrated in enhancing passive seismic event detection/denoising. However, there are other obvious applications of the DeepSeg in active and passive seismic fields, e.g., seismic imaging, preprocessing of ambient noise data, and microseismic event monitoring. It is worth pointing out here that the deep neural network is trained exclusively using synthetic seismic data, negating the need for real data during the training phase. Furthermore, the proposed setup is general and its potential applications are not confined to passive event denoising or even seismic. The method proposed is also adaptable to other diverse signals in different settings, like medical images/signals magnetic resonance imaging (MRI), electroencephalogram (EEG) signals, electrocardiograms (ECG) signals, and retinal images, to name a few, radar signals, speech signals, fault detection in electrical/mechanical systems, daily life images, etc. Experiments on synthetic and real seismic data reveal the efficacy and supremacy of the proposed method in terms of SNR improvement and required training data when compared to the state-of-the-art deep neural network-based denoising technique.

Toward Robust Fault Identification of Complex Industrial Processes Using Stacked Sparse-Denoising Autoencoder With Softmax Classifier

Abstract: This article proposes a robust end-to-end deep learning-induced fault recognition scheme by stacking multiple sparse-denoising autoencoders with a Softmax classifier, called stacked spare-denoising autoencoder (SSDAE)-Softmax, for the fault identification of complex industrial processes (CIPs). Specifically, sparse denoising autoencoder (SDAE) is established by integrating a sparse AE (SAE) with a denoising AE (DAE) for the low-dimensional but intrinsic feature representation of the CIP monitoring data (CIPMD) with possible noise contamination. SSDAE-Softmax is established by stacking multiple SDAEs with a layerwise pretraining procedure, and a Softmax classifier with a global fine-tuning strategy. Furthermore, SSDAE-Softmax hyperparameters are optimized by a relatively new global optimization algorithm, referred to as the state transition algorithm (STA). Benefiting from the deep learning-based feature representation scheme with the STA-based hyperparameter optimization, the underlying intrinsic characteristics of CIPMD can be learned automatically and adaptively for accurate fault identification. A numeric simulation system, the benchmark Tennessee Eastman process (TEP), and a real industrial process, that is, the continuous casting process (CCP) from a top steel plant of China, are used to validate the performance of the proposed method. Experimental results show that the proposed SSDAE-Softmax model can effectively identify various process faults, and has stronger robustness and adaptability against the noise interference in CIPMD for the process monitoring of CIPs.

A denoising framework for 3D and 2D imaging techniques based on photon detection statistics

Abstract: A method to capture three-dimensional (3D) objects image data under extremely low light level conditions, also known as Photon Counting Imaging (PCI), was reported. It is demonstrated that by combining a PCI system with computational integral imaging algorithms, a 3D scene reconstruction and recognition is possible. The resulting reconstructed 3D images often look degraded (due to the limited number of photons detected in a scene) and they, therefore, require the application of superior image restoration techniques to improve object recognition. Recently, Deep Learning (DL) frameworks have been shown to perform well when used for denoising processes. In this paper, for the first time, a fully unsupervised network (i.e., U-Net) is proposed to denoise the photon counted 3D sectional images. In conjunction with classical U-Net architecture, a skip block is used to extract meaningful patterns from the photons counted 3D images. The encoder and decoder blocks in the U-Net are connected with skip blocks in a symmetric manner. It is demonstrated that the proposed DL network performs better, in terms of peak signal-to-noise ratio, in comparison with the classical TV denoising algorithm.

M₃NAS: Multi-Scale and Multi-Level Memory-Efficient Neural Architecture Search for Low-Dose CT Denoising

Abstract: Lowering the radiation dose in computed tomography (CT) can greatly reduce the potential risk to public health. However, the reconstructed images from the dose-reduced CT or low-dose CT (LDCT) suffer from severe noise, compromising the subsequent diagnosis and analysis. Recently, convolutional neural networks have achieved promising results in removing noise from LDCT images; the network architectures used are either handcrafted or built on top of conventional networks such as ResNet and U-Net. Recent advance on neural network architecture search (NAS) has proved that the network architecture has a dramatic effect on the model performance, which indicates that current network architectures for LDCT may be sub-optimal. Therefore, in this paper, we make the first attempt to apply NAS to LDCT and propose a multi-scale and multi-level NAS for LDCT denoising, termed MANAS. On the one hand, the proposed MANAS fuses features extracted by different scale cells to capture multi-scale image structural details. On the other hand, the proposed MANAS can search a hybrid cell- and network-level structure for better performance. Extensively experimental results on three different dose levels demonstrate that the proposed MANAS can achieve better performance in terms of preserving image structural details than several state-of-the-art methods. In addition, we also validate the effectiveness of the multi-scale and multi-level architecture for LDCT denoising.

Noise Suppression With Similarity-Based Self-Supervised Deep Learning

Abstract: Image denoising is a prerequisite for downstream tasks in many fields. Low-dose and photon-counting computed tomography (CT) denoising can optimize diagnostic performance at minimized radiation dose. Supervised deep denoising methods are popular but require paired clean or noisy samples that are often unavailable in practice. Limited by the independent noise assumption, current self-supervised denoising methods cannot process correlated noises as in CT images. Here we propose the first-of-its-kind similarity-based self-supervised deep denoising approach, referred to as Noise2Sim, that works in a nonlocal and nonlinear fashion to suppress not only independent but also correlated noises. Theoretically, Noise2Sim is asymptotically equivalent to supervised learning methods under mild conditions. Experimentally, Nosie2Sim recovers intrinsic features from noisy low-dose CT and photon-counting CT images as effectively as or even better than supervised learning methods on practical datasets visually, quantitatively and statistically. Noise2Sim is a general self-supervised denoising approach and has great potential in diverse applications.

Fractional-order diffusion coupled with integer-order diffusion for multiplicative noise removal

Abstract: The problem of multiplicative noise removal has attracted much attention in recent years. Unlike additive noise, multiplicative noise destroys almost all information of an image; therefore, it is more difficult to remove them from corrupted images. In this paper, we propose a denoising model of fractional-order diffusion coupled with integer-order diffusion for multiplicative gamma noise, which takes advantage of both texture-preserving property of fractional-order diffusion and edge-preserving property of integer-order diffusion. We explore the mutual mechanism of two diffusion equations in the diffusion process, i.e., mutual transfer of texture and edge information respectively from two filtering images, and mutual regularization on the diffusion coefficients in both equations. We design an alternating numerical scheme based on semi-implicit finite difference and discrete Fourier transform for solving the coupled diffusion system. The proposed model is tested on some commonly-used images and is compared with six state-of-the-art models, qualitatively and quantitatively. Experimental results show the superiority of the proposed model in reducing multiplicative gamma noise while preserving textures and edges.

Ocean bottom seismometer (OBS) noise reduction from horizontal and vertical components using harmonic–percussive separation algorithms

Abstract: Abstract. Records from ocean bottom seismometers (OBSs) are highly contaminated by noise, which is much stronger compared to data from most land stations, especially on the horizontal components. As a consequence, the high energy of the oceanic noise at frequencies below 1 Hz considerably complicates the analysis of the teleseismic earthquake signals recorded by OBSs. Previous studies suggested different approaches to remove low-frequency noises from OBS recordings but mainly focused on the vertical component. The records of horizontal components, which are crucial for the application of many methods in passive seismological analysis of body and surface waves, could not be much improved in the teleseismic frequency band. Here we introduce a noise reduction method, which is derived from the harmonic–percussive separation algorithms used in Zali et al. (2021), in order to separate long-lasting narrowband signals from broadband transients in the OBS signal. This leads to significant noise reduction of OBS records on both the vertical and horizontal components and increases the earthquake signal-to-noise ratio (SNR) without distortion of the broadband earthquake waveforms. This is demonstrated through tests with synthetic data. Both SNR and cross-correlation coefficients showed significant improvements for different realistic noise realizations. The application of denoised signals in surface wave analysis and receiver functions is discussed through tests with synthetic and real data.

Iterative α -(de)Blending: a Minimalist Deterministic Diffusion Model

Abstract: We derive a minimalist but powerful deterministic denoising-diffusion model. While denoising diffusion has shown great success in many domains, its underlying theory remains largely inaccessible to non-expert users. Indeed, an understanding of graduate-level concepts such as Langevin dynamics or score matching appears to be required to grasp how it works. We propose an alternative approach that requires no more than undergrad calculus and probability. We consider two densities and observe what happens when random samples from these densities are blended (linearly interpolated). We show that iteratively blending and deblending samples produces random paths between the two densities that converge toward a deterministic mapping. This mapping can be evaluated with a neural network trained to deblend samples. We obtain a model that behaves like deterministic denoising diffusion: it iteratively maps samples from one density (e.g., Gaussian noise) to another (e.g., cat images). However, compared to the state-of-the-art alternative, our model is simpler to derive, simpler to implement, more numerically stable, achieves higher quality results in our experiments, and has interesting connections to computer graphics.

Image Denoising Based on Quantum Calculus of Local Fractional Entropy

Abstract: Images are frequently disrupted by noise of all kinds, making image restoration very challenging. There have been many different image denoising models proposed over the last few decades. Some models preserve the image’s smooth region, while others preserve the texture margin. One of these methods is by using quantum calculus. Quantum calculus is a branch of mathematics that deals with the manipulation of functions and operators in a quantum mechanical setting. It has been used in image processing to improve the speed and accuracy of image-processing algorithms. In quantum computing, entropy can be defined as a measure of the disorder or randomness of a quantum state. The concept of local fractional entropy has been used to study a wide range of quantum systems. In this study, an image denoising model is proposed based on the quantum calculus of local fractional entropy (QC-LFE) to remove a Gaussian noise. The local fractional entropy is used to estimate the image pixel probability, while the quantum calculus is used to estimate the convolution window mask for image denoising. A processing fractional mask with n x n elements was used in the suggested denoising algorithm. The proposed image denoising algorithm uses mask convolution to process each corrupted pixel one at a time. The proposed denoising algorithm’s effectiveness is assessed using peak signal-to-noise ratio and visual perception (PSNR). The experimental findings show that, compared to other similar fractional operators, the proposed method can better preserve texture details when denoising.

Low Cost and Latency Event Camera Background Activity Denoising

Abstract: Dynamic Vision Sensor (DVS) event camera output includes uninformative background activity (BA) noise events that increase dramatically under dim lighting. Existing denoising algorithms are not effective under these high noise conditions. Furthermore, it is difficult to quantitatively compare algorithm accuracy. This paper proposes a novel framework to better quantify BA denoising algorithms by measuring receiver operating characteristics with known mixtures of signal and noise DVS events. New datasets for stationary and moving camera applications of DVS in surveillance and driving are used to compare 3 new low-cost algorithms: Algorithm 1 checks distance to past events using a tiny fixed size window and removes most of the BA while preserving most of the signal for stationary camera scenarios. Algorithm 2 uses a memory proportional to the number of pixels for improved correlation checking. Compared with existing methods, it removes more noise while preserving more signal. Algorithm 3 uses a lightweight multilayer perceptron classifier driven by local event time surfaces to achieve the best accuracy over all datasets. The code and data are shared with the paper as DND21.

Denoising Fault-Aware Wavelet Network: A Signal Processing Informed Neural Network for Fault Diagnosis

Abstract: Abstract Deep learning (DL) is progressively popular as a viable alternative to traditional signal processing (SP) based methods for fault diagnosis. However, the lack of explainability makes DL-based fault diagnosis methods difficult to be trusted and understood by industrial users. In addition, the extraction of weak fault features from signals with heavy noise is imperative in industrial applications. To address these limitations, inspired by the Filterbank-Feature-Decision methodology, we propose a new Signal Processing Informed Neural Network (SPINN) framework by embedding SP knowledge into the DL model. As one of the practical implementations for SPINN, a denoising fault-aware wavelet network (DFAWNet) is developed, which consists of fused wavelet convolution (FWConv), dynamic hard thresholding (DHT), index-based soft filtering (ISF), and a classifier. Taking advantage of wavelet transform, FWConv extracts multiscale features while learning wavelet scales and selecting important wavelet bases automatically; DHT dynamically eliminates noise-related components via point-wise hard thresholding; inspired by index-based filtering, ISF optimizes and selects optimal filters for diagnostic feature extraction. It's worth noting that SPINN may be readily applied to different deep learning networks by simply adding filterbank and feature modules in front. Experiments results demonstrate a significant diagnostic performance improvement over other explainable or denoising deep learning networks. The corresponding code is available at https://github.com/albertszg/DFAWnet .

De-Noising of Photoacoustic Microscopy Images by Attentive Generative Adversarial Network

Abstract: As a hybrid imaging technology, photoacoustic microscopy (PAM) imaging suffers from noise due to the maximum permissible exposure of laser intensity, attenuation of ultrasound in the tissue, and the inherent noise of the transducer. De-noising is an image processing method to reduce noise, and PAM image quality can be recovered. However, previous de-noising techniques usually heavily rely on manually selected parameters, resulting in unsatisfactory and slow de-noising performance for different noisy images, which greatly hinders practical and clinical applications. In this work, we propose a deep learning-based method to remove noise from PAM images without manual selection of settings for different noisy images. An attention enhanced generative adversarial network is used to extract image features and adaptively remove various levels of Gaussian, Poisson, and Rayleigh noise. The proposed method is demonstrated on both synthetic and real datasets, including phantom (leaf veins) and in vivo (mouse ear blood vessels and zebrafish pigment) experiments. In the in vivo experiments using synthetic datasets, our method achieves the improvement of 6.53 dB and 0.26 in peak signal-to-noise ratio and structural similarity metrics, respectively. The results show that compared with previous PAM de-noising methods, our method exhibits good performance in recovering images qualitatively and quantitatively. In addition, the de-noising processing speed of 0.016 s is achieved for an image with 256×256 pixels, which has the potential for real-time applications. Our approach is effective and practical for the de-noising of PAM images.

MR Image Denoising and Super-Resolution Using Regularized Reverse Diffusion

Abstract: Patient scans from MRI often suffer from noise, which hampers the diagnostic capability of such images. As a method to mitigate such artifacts, denoising is largely studied both within the medical imaging community and beyond the community as a general subject. However, recent deep neural network-based approaches mostly rely on the minimum mean squared error (MMSE) estimates, which tend to produce a blurred output. Moreover, such models suffer when deployed in real-world situations: out-of-distribution data, and complex noise distributions that deviate from the usual parametric noise models. In this work, we propose a new denoising method based on score-based reverse diffusion sampling, which overcomes all the aforementioned drawbacks. Our network, trained only with coronal knee scans, excels even on out-of-distribution in vivo liver MRI data, contaminated with a complex mixture of noise. Even more, we propose a method to enhance the resolution of the denoised image with the same network. With extensive experiments, we show that our method establishes state-of-the-art performance while having desirable properties which prior MMSE denoisers did not have: flexibly choosing the extent of denoising, and quantifying uncertainty.

Deep Learning Models for Single-Channel Speech Enhancement on Drones

Abstract: Speech enhancement for drone audition is made challenging by the strong ego-noise from the rotating motors and propellers, which leads to extremely low signal-to-noise ratios (e.g. SNR < -15 dB) at onboard microphones. In this paper, we extensively assess the ability of single-channel deep learning approaches to ego-noise reduction on drones. We train twelve representative deep neural network (DNN) models, covering three operation domains (time-frequency magnitude domain, time-frequency complex domain and end-to-end time domain) and three distinct architectures (sequential, encoder-decoder and generative). We critically discuss and compare the performance of these models in extremely low-SNR scenarios, ranging from -30 to 0 dB. We show that time-frequency complex domain and UNet encoder-decoder architectures outperform other approaches on speech enhancement measures while providing a good trade-off with other criteria, such as model size, computation complexity and context length. Specifically, the best-performing model is DCUNet, a UNet model operating in the time-frequency complex domain, which, at input SNR -15 dB, improves ESTOI from 0.1 to 0.4, PESQ from 1.0 to 1.9 and SI-SDR from -15 dB to 3.7 dB. Based on the insights drawn from these findings, we discuss future research in drone ego-noise reduction.