Additive Noise Removal Research Articles

Adaptive Bayesian speckle reduction in the 2D DOST domain using 2D MC-GARCH-M model: Preserving image edges and textural information

Image denoising and image quality enhancement are major issues in image processing. Additive and multiplicative noise removal is one of the main image enhancement approaches. In this paper, we propose a novel 2D Merged Complex Generalized Autoregressive Conditional Heteroscedasticity Mixture (MC-GARCH-M) model for 2D complex stochastic processes. This model effectively captures non-Gaussian statistics and all three types of dependencies within the processes (RRD, IID, and RID) using the proposed linear mapping transformation T. Our statistical analysis shows that the 2D MC-GARCH-M model provides the best fit to the non-Gaussian statistics of the mapped matrix of 2D DOST coefficients compared to Gaussian, Generalized Gaussian, and Stable distributions. We confirm the presence of heteroscedasticity in the mapped matrix using the Engle hypothesis test, verifying ARCH/GARCH effects. The 2D MC-GARCH-M model, with its location-dependent conditional variances, provides a flexible approach for noise removal and modeling complex stochastic processes. Experimental results on artificially speckled aerial and actual SAR images demonstrate that our method outperforms state-of-the-art approaches in terms of speckle removal and preserving the edges and textural information of the image. Key innovations include: a new statistical analysis using linear mapping T showing non-Gaussian heavy-tailed distributions; an adaptive Bayesian estimator, MCMAP, based on the heteroscedastic model for speckle noise removal; and the method's applicability to images with varying mutual information (MI) between real and imaginary parts of 2D DOST coefficients. Our method is adaptive and robust to initial parameter settings, effectively utilizing magnitude and phase information, providing an optimal closed-form solution, reducing memory and computational requirements, and demonstrating robustness to speckle power levels. While requiring more computational resources compared to thresholding methods, our approach is less demanding than other Bayesian algorithms, making it suitable for offline processing.

SDAUNet: A simple dual attention mechanism UNet for mixed noise removal

AbstractConvolutional neural networks (CNNs) have demonstrated impressive results in additive white Gaussian noise removal due to their strong fitting ability. However, their performance in mixed noise removal remains unsatisfactory, primarily due to their limited receptive field that focuses only on the local features of images and disregards global information. To ameliorate this issue, recent state‐of‐the‐art approaches employ attention mechanism (AM) to capture the global information. However, most AM based methods still suffer from low computational efficiency. In this paper, a novel model named simple dual attention mechanism UNet (SDAUNet) for mixed noise removal is proposed. In SDAUNet, the UNet architecture is used to gradually acquire multi‐scale image features and provide a more comprehensive and accurate representation of the image features than other CNNs. A simple dual attention convolutional block is presented to acquire the global image features that can successfully capture image details with a low burden. The experimental results demonstrate that the SDAUNet model can achieve better measurement metrics and visual performance than other state‐of‐the‐art methods.

Comparative Analysis on De-Noising of MRI Uterus Image for Identification of Endometrial Carcinoma

Purpose: The anatomical and physiological processes of the human body are pictured in radiology using different modalities. Magnetic Resonance Imaging (MRI) supports capturing the images of organs using magnetic field gradients. The quality of MR images is generally affected by various noises such as Gaussian, speckle, salt and pepper, Rayleigh, Rican etc. Removal of these noises from the MR images is essential for further diagnostic procedures.
 Materials and Methods: In this article, Gaussian noise, speckle noise, and salt and pepper noise are added to the MR uterus image for which different filters are applied to remove the noise for precise identification of endometrial carcinoma.
 Results: The different filters incorporated for the additive noise removal process are the bilateral filter, Non-Local Means (NLM) filter, anisotropic diffusion filter, and Convolution Neural Network (CNN). The efficiency of the filter is calculated by evaluating the response of the filter by gradually increasing the noise intensity of the MR images.
 Conclusion: Further, peak Signal-to-Noise Ratio (SNR), structural similarity index measure, image quality index and computational cost parameters are computed and analyzed.

A serial attention module‐based deep convolutional neural network for mixed Gaussian‐impulse removal

AbstractThe removal of mixed noise is a challenging task because the attenuation of the noise distribution cannot be described precisely. The coupling of additive white Gaussian noise and impulse noise (IN) is a typical case. At present, most methods use a two‐phase strategy, that is, IN detection coupled with additive white Gaussian noise removal, often leading to poor denoising results with an increase in the ratio of IN. In this paper, an effective convolutional neural network (CNN) model is proposed, namely a serial attention module‐based CNN (SACNN), for mixed noise removal. In contrast to the existing two‐phase methods, SACNN unifies the denoising process into a single CNN framework. In SACNN, residual learning and batch normalization are used to train the model, which speeds up the convergence and improves the mixed noise removal performance. Meanwhile, the serial attention module is applied to better preserve the texture details. The experimental results reveal that SACNN achieves superior quality metrics and visual appearance when compared to several leading approaches.

Speech Enhancement and Dereverberation With Diffusion-Based Generative Models

In this work, we build upon our previous publication and use diffusion-based generative models for speech enhancement. We present a detailed overview of the diffusion process that is based on a stochastic differential equation and delve into an extensive theoretical examination of its implications. Opposed to usual conditional generation tasks, we do not start the reverse process from pure Gaussian noise but from a mixture of noisy speech and Gaussian noise. This matches our forward process which moves from clean speech to noisy speech by including a drift term. We show that this procedure enables using only 30 diffusion steps to generate high-quality clean speech estimates. By adapting the network architecture, we are able to significantly improve the speech enhancement performance, indicating that the network, rather than the formalism, was the main limitation of our original approach. In an extensive cross-dataset evaluation, we show that the improved method can compete with recent discriminative models and achieves better generalization when evaluating on a different corpus than used for training. We complement the results with an instrumental evaluation using real-world noisy recordings and a listening experiment, in which our proposed method is rated best. Examining different sampler configurations for solving the reverse process allows us to balance the performance and computational speed of the proposed method. Moreover, we show that the proposed method is also suitable for dereverberation and thus not limited to additive background noise removal. Code and audio examples are available online <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup> <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://github.com/sp-uhh/sgmse</uri> .

Additive Gaussian noise removal based on generative adversarial network model and semi-soft thresholding approach

Additive Gaussian noise removal based on generative adversarial network model and semi-soft thresholding approach

Polarimetric SAR Despeckling With Convolutional Neural Networks

Coherent imaging systems such as synthetic aperture radar (SAR) are subject to speckle, the reduction of which is an active area of study. Methods based on deep convolutional neural networks (CNNs) have recently demonstrated state-of-the-art performance in the removal of additive noise from natural images and speckle from single-channel SAR images. The application of deep learning to multichannel SAR modalities such as polarimetric SAR (PolSAR) is complicated in part by the nature of the data as images of complex-valued covariance matrices. In this article, we propose a CNN-based PolSAR despeckling approach that uses an invertible transformation involving a matrix logarithm to facilitate CNN processing of the PolSAR data. A residual learning strategy is adopted, in which the CNN is trained to identify the speckle component which is then removed from the corrupted image. The experimental results on simulated and measured PolSAR data show the proposed approach to markedly reduce speckle and preserve scene features.

Shock filter coupled with a high-order PDE for additive noise removal and image quality enhancement

We are interested in repairing blurry images degraded by additive noise. We present a new filter for image enhancement based on the combination of a shock filter with a fourth order diffusion equation. The algorithm that we propose eliminates not only the blurry but also the noise. The control of the diffusion weight and that of the enhancement of the model is done by a parameter. After experimentation on some test images and on some real images of degraded old documents, the results obtained are satisfactory compared to certain models in the literature.

Speckle noise removal based on structural convolutional neural networks with feature fusion for medical image

Application of convolutional neural networks (CNNs) for image additive white Gaussian noise (AWGN) removal has attracted considerable attentions with the rapid development of deep learning in recent years. However, the work of image multiplicative speckle noise removal is rarely done. Moreover, most of the existing speckle noise removal algorithms are based on traditional methods with human priori knowledge, which means that the parameters of the algorithms need to be set manually. Nowadays, deep learning methods show clear advantages on image feature extraction. Multiplicative speckle noise is very common in real life images, especially in medical images. In this paper, a novel neural network structure is proposed to recover noisy images with speckle noise. Our proposed method mainly consists of three subnetworks. One network is rough clean image estimate subnetwork. Another is subnetwork of noise estimation. The last one is an information fusion network based on U-Net and several convolutional layers. Different from the existing speckle denoising model based on the statistics of images, the proposed network model can handle speckle denoising of different noise levels with an end-to-end trainable model. Extensive experimental results on several test datasets clearly demonstrate the superior performance of our proposed network over state-of-the-arts in terms of quantitative metrics and visual quality.

A Novel approach in Hybrid Median Filtering for Denoising Medical images

Image denoising is a key pre-processing step in medical image analysis. At current, deep learning-based models have shown a great promise, which outperformed many conventional methods over the past three decades. Speckle noise removal is a major issue in preserving all the delicate details and the edges in ultrasound image processing as it degrades the visual evaluation of ultrasound images. The multiplicative behavior of speckle-noise is converted into additive by using log transform as the additive noise removal is easy as compared to multiplicative noise. An innovative approach for denoising highly distorted images affected by speckle noise is proposed. This paper presents a result of significant work in image denoising and exploring several thresholding methods of denoising images such as SureShrink, VisuShrink and BayesShrink. The results of different approaches of wavelet-based image denoising methods are tabulated to find the best method. The main aim is to show the result of wavelet coefficients on a new basis, so that the noise can be minimized or removed from the data.

An Enhanced Optimal Technique For Accurate Detection Of Color Face Images With Different Illuminations

Secured authentication system is one of the most challenging tasks focused now-a-days by many researchers and greatly achieved by means of face detection technique. Face image recognition is currently realized by Adaptive singular value decomposition in two-dimensional discrete Fourier domain (ASVDF). The recognition systems ability enhancement is attained for face images recognition by side light influence reduction on a color face image for inadequate light. The prevailing researches does not focuses the following points: No correct output during face recognition process, Face spoofing is not concentrated thereby face recognition may effect in imprecise result, Optimal feature extraction. Optimized Face Recognition System with Illumination and Rotation Consideration (OFRS-IRC) is one of the promising solutions for mitigating all those issues. Various methods are presented for ensuring accurate face recognition. Additive White Gaussian Noise removal technique is utilized for eliminating noise when the image is captured through sensor devices. Illuminate invariant features and locality preserving projection approach is exploited for segmented image recognition. As a final step, Fuzzy neural network is deployed for precise prediction on the basis of locality preserving projection approach results. MATLAB simulation tool is exploited for evaluating this research, where improved performance are attained by proposed method than prevailing methods. The proposed method shows 7.42% better detection rate than the existing work.

Review of noise removal techniques in ECG signals

An electrocardiogram (ECG) records the electrical signal from the heart to check for different heart conditions, but it is susceptible to noises. ECG signal denoising is a major pre-processing step which attenuates the noises and accentuates the typical waves in ECG signals. Researchers over time have proposed numerous methods to correctly detect morphological anomalies. This study discusses the workflow, and design principles followed by these methods, and classify the state-of-the-art methods into different categories for mutual comparison, and development of modern methods to denoise ECG. The performance of these methods is analysed on some benchmark metrics, viz., root-mean-square error, percentage-root-mean-square difference, and signal-to-noise ratio improvement, thus comparing various ECG denoising techniques on MIT-BIH databases, PTB, QT, and other databases. It is observed that Wavelet-VBE, EMD-MAF, GAN2, GSSSA, new MP-EKF, DLSR, and AKF are most suitable for additive white Gaussian noise removal. For muscle artefacts removal, GAN1, new MP-EKF, DLSR, and AKF perform comparatively well. For base-line wander, and electrode motion artefacts removal, GAN1 is the best denoising option. For power-line interference removal, DLSR and EWT perform well. Finally, FCN-based DAE, DWT (Sym6) soft, MABWT (soft), CPSD sparsity, and UWT are promising ECG denoising methods for composite noise removal.

A globally convergent algorithm for a class of gradient compounded non-Lipschitz models applied to non-additive noise removal

Non-Lipschitz regularization has got much attention in image restoration with additive noise removal recently, which can preserve neat edges in the restored image. In this paper, we consider a class of minimization problems with gradient compounded non-Lipschitz regularization applied to non-additive noise removal, with Poisson and multiplicative one as examples. The existence of a solution of the general model is discussed. We also extend the recent iterative support shrinkage strategy to give an algorithm to minimize it, where the subproblem at each iteration is allowed to be solved inexactly. Moreover, this paper is the first one to give the subdifferential of the gradient compounded non-Lipschitz regularization term, based on which we are able to establish the global convergence of the iterative sequence to a stationary point of the original objective function. This is, to our best knowledge, stronger than all the convergence results for gradient compounded non-Lipschitz minimization problems in the current published literature. Numerical experiments show that our proposed method performs well.

Data-driven estimation of noise variance stabilization parameters for low-dose x-ray images

Purpose. Denoising x-ray images corrupted by signal-dependent mixed noise is usually approached either by considering noise statistics directly or by using noise variance stabilization (NVS) techniques. An advantage of the latter is that the noise variance can be stabilized to a known constant throughout the image, facilitating the application of denoising algorithms designed for the removal of additive Gaussian noise. A well-performing NVS is the generalized Anscombe transform (GAT). To calculate the GAT, the system gain as well as the variance of electronic noise are required. Unfortunately, these parameters are difficult to predict from the x-ray tube settings in clinical practice, because the system gain observed at the detector depends on the beam hardening caused by the patient. Materials and Methods. We propose a data-driven method for estimating the parameters required to carry out an NVS using the GAT. It utilizes the energy compaction property of the discrete cosine transform to obtain the NVS parameters using a robust regression approach relying on a linear Poisson-Gaussian model. The method has been experimentally validated with respect to beam hardening as well as denoising performance for different dose and scatter levels. Results. Across a range of low-dose x-ray settings, the proposed robust regression approach has estimated both system gain and electronic noise level with an average error of only 4.2%. When used to perform a GAT followed by the denoising of low-dose x-ray images, performance gains of 5% for peak-signal-to-noise ratio and 4% for structural similarity index can be obtained. Conclusion. The parameters needed to calculate the GAT can be estimated efficiently and robustly using a data-driven approach. The improved parameter estimation method facilitates a more accurate GAT-based NVS and, hence, better denoising of low-dose x-ray images when algorithms designed for additive Gaussian noise are applied.

Detection of Defects in Additively Manufactured Stainless Steel 316L with Compact Infrared Camera and Machine Learning Algorithms

Additive manufacturing (AM) is an emerging method for cost-efficient fabrication of nuclear reactor parts. AM of metallic structures for nuclear energy applications is currently based on the laser powder bed fusion process, which can introduce internal material flaws, such as pores and anisotropy. Integrity of AM structures needs to be evaluated nondestructively because material flaws could lead to premature failures due to exposure to high temperature, radiation and corrosive environments in a nuclear reactor. Thermal tomography (TT) provides a capability for non-destructive evaluation of sub-surface defects in arbitrary size structures. We investigate TT of AM stainless steel 316L specimens with imprinted internal porosity defects using a relatively low-cost, small form factor infrared camera based on an uncooled micro-bolometer detector. Sparse coding-related K-means singular value decomposition machine learning, image processing algorithms are developed to improve the quality of TT images through removal of additive white Gaussian noise without blurring the images.

Sentinel-1 additive noise removal from cross-polarization extra-wide TOPSAR with dynamic least-squares

Sentinel-1 is a synthetic aperture radar (SAR) platform with an operational mode called extra wide (EW) that offers large regions of ocean areas to be observed. A major issue with EW images is that the cross-polarized HV and VH channels have prominent additive noise patterns relative to low backscatter intensity, which disrupts tasks that require manual or automated interpretation. The European Space Agency (ESA) provides a method for removing the additive noise pattern by means of lookup tables, but applying them directly produces unsatisfactory results because characteristics of the noise still remain. Furthermore, evidence suggests that the magnitude of the additive noise dynamically depends on factors that are not considered by the ESA estimated noise field.To address these issues we propose a quadratic objective function to model the mis-scale of the provided noise field on an image. We consider a linear denoising model that re-scales the noise field for each subswath, whose parameters are found from a least-squares solution over the objective function. This method greatly reduces the presence of additive noise while not requiring a set of training images, is robust to heterogeneity in images, dynamically estimates parameters for each image, and finds parameters using a closed-form solution.Two experiments were performed to validate the proposed method. The first experiment simulated noise removal on a set of RADARSAT-2 images with noise fields artificially imposed on them. The second experiment conducted noise removal on a set of Sentinel-1 images taken over the five oceans. Afterwards, quality of the noise removal was evaluated based on the appearance of open-water. The two experiments indicate that the proposed method marks an improvement both visually and through numerical measures.

Multi-band weighted [formula omitted] norm minimization for image denoising

Low rank matrix approximation (LRMA) has a wide range of applications in computer vision and has drawn much attention in recent years. A typical nuclear norm minimization (NNM) is often used to solve a LRMA, but this is likely to overshrink the rank components due to having the same threshold. To address this problem, we propose a flexible and precise model named multi-band weighted lp norm minimization (MBWPNM). We have reformulated it into nonconvex lp norm subproblems under certain weight conditions, and we solve these subproblems via a generalized soft-thresholding algorithm. We then adopt MBWPNM for image (grayscale, color and multispectral) denoising. The proposed MBWPNM not only guarantees a more accurate approximation with a Schatten p-norm in case of a change of noise levels, but it also considers prior knowledge, for which different rank components have varying importance. Extensive experiments on additive white Gaussian noise removal and realistic noise removal demonstrate that the proposed MBWPNM achieves better performance than several state-of-the-art algorithms.

Cauchy Noise Removal by Weighted Nuclear Norm Minimization

Recently, weighted nuclear norm minimization (WNNM), which regularizes singular values of an input matrix with different strengths according to given weights, has demonstrated impressive results in low-level vision tasks such as additive Gaussian noise removal, deblurring and image inpainting [14, 15, 33]. In this study, we apply WNNM to remove additive Cauchy noise in images. A variational model is adopted based on maximum a posteriori estimate, which contains a data fidelity term that is appropriate for noise following the Cauchy distribution. Weighted nuclear norm is used as a regularizer in the proposed algorithm, and we utilized similar patches in the image by nonlocal similarity. We adopted the nonconvex alternating direction method of multiplier to solve the problem iteratively. Numerical experiments are presented to demonstrate the superior denoising performance of our algorithm compared with other existing methods in terms of quantitative measure and visual quality.

Blind Universal Bayesian Image Denoising with Gaussian Noise Level Learning.

Blind and universal image denoising consists of using a unique model that denoises images with any level of noise. It is especially practical as noise levels do not need to be known when the model is developed or at test time. We propose a theoretically-grounded blind and universal deep learning image denoiser for additive Gaussian noise removal. Our network is based on an optimal denoising solution, which we call fusion denoising. It is derived theoretically with a Gaussian image prior assumption. Synthetic experiments show our network's generalization strength to unseen additive noise levels. We also adapt the fusion denoising network architecture for image denoising on real images. Our approach improves real-world grayscale additive image denoising PSNR results for training noise levels and further on noise levels not seen during training. It also improves state-of-the-art color image denoising performance on every single noise level, by an average of 0.1dB, whether trained on or not.

Anisotropic Diffusion Based Multiplicative Speckle Noise Removal.

Multiplicative speckle noise removal is a challenging task in image processing. Motivated by the performance of anisotropic diffusion in additive noise removal and the structure of the standard deviation of a compressed speckle noisy image, we address this problem with anisotropic diffusion theories. Firstly, an anisotropic diffusion model based on image statistics, including information on the gradient of the image, gray levels, and noise standard deviation of the image, is proposed. Although the proposed model can effectively remove multiplicative speckle noise, it does not consider the noise at the edge during the denoising process. Hence, we decompose the divergence term in order to make the diffusion at the edge occur along the boundaries rather than perpendicular to the boundaries, and improve the model to meet our requirements. Secondly, the iteration stopping criteria based on kurtosis and correlation in view of the lack of ground truth in real image experiments, is proposed. The optimal values of the parameters in the model are obtained by learning. To improve the denoising effect, post-processing is performed. Finally, the simulation results show that the proposed model can effectively remove the speckle noise and retain minute details of the images for the real ultrasound and RGB color images.