Denoising Model Research Articles

An Architecture-Oriented Analysis of Stacked Denoising Autoencoders

One of the major threats in computer vison and image processing areas is image tarnishing due to noisy pixels. Image tarnishing occurs when clean images are subjected to multiple feature changes, length transmission, or due to improper image capturing. The growth of Autoencoders in Artificial Neural Networks has paved way to efficient noise removal. In this work, an architecture oriented analysis of stacked denoising autoencoders is implemented by considering two perspectives of noise, namely noise as an inverse problem and noise as a residual problem. Autoencoders have been modelled by considering the approaches: Shallow Model, Deep-Series and Parallel Series models. Autoencoders with Skip Connections and Denoising CNNs have also been implemented and analyzed in order to propose an efficient, optimized and best working model for denoising. The dataset used is Common Object in Context (COCO) and the performance of the model is evaluated based on Structural Similarity Index (SSIM) and Peak Signal to Noise Ratio (PSNR). The analysis has presented the impact of the model structures of the autoencoders, which implicitly tend to treat noise as an inverse problem.

An Improved VMD and Wavelet Hybrid Denoising Model for Wearable SSVEP-BCI

An Improved VMD and Wavelet Hybrid Denoising Model for Wearable SSVEP-BCI

Local estimates for vectorial Rudin–Osher–Fatemi type problems in one dimension

We consider the Rudin–Osher–Fatemi variational denoising model with general regularizing term and L2 fidelity in one-dimensional, vector-valued setting. We obtain local estimates on the singular part of the variation measure of the minimizer in terms of the singular part of the variation measure of the datum. In the case of homogeneous regularizer, we prove local estimates on the whole variation measure of the minimizer and deduce an analogous result for the gradient flow of the regularizer. We also discuss the question of extending our results to other fidelities.

Weighted and Well-Balanced Nonlinear TV-Based Time-Dependent Model for Image Denoising

The partial differential equation (PDE)-based models are widely used to remove additive Gaussian white noise and preserve edges, and one of the most widely used methods is the total variation denoising algorithm. Total variation (TV) denoising algorithm-based time-dependent models have seen considerable success in the field of image-denoising and edge detection. TV denoising algorithm is based on that signals with spurious detail have a high total variation and reduction of unwanted signals to achieve noise-free images. It is a constrained optimization-type algorithm. The Lagrange multiplier and gradient descent method are used to solve the TV algorithm to reach the PDE-based time dependent model. To eliminate additive noise and preserve edges, we investigate a class of weighted time-dependent model in this study. The proposed method is investigated in a well-balanced flow form that extends the time-dependent model with an adaptive fidelity element. Adaptive function is fusing into the regularization term of the classical time-dependent model which successfully enhances the intensity of the regularizer function. We maintain the ability of the time-dependent model without any oscillation effects. Furthermore, we want to prove the viscosity solution of our weighted and well balanced time-dependent model, demonstrating its existence and uniqueness. The finite difference method is applied to discretize the nonlinear time-dependent models. The numerical results are expressed as a statistic known as the peak signal-to-noise ratio (PSNR) and structural similarity index metric (SSIM). Numerical experiments demonstrate that the proposed model yields good performance compared with the previous time-dependent model.

CicadaNet: Deep learning based automatic cicada chorus filtering for improved long-term bird monitoring

Passive acoustic monitoring has been an effective tool for bird sound analysis. However, bird sounds often include cicada noise, which is an obstacle for investigating bird sounds. For example, cicada noise can result in large deviations of acoustic index, which will lead to the mismonitoring of species richness trends. Therefore, there is a critical need to filter cicada noise for helping bird sound analysis. We develop a novel end-to-end deep learning model, named CicadaNet for filtering cicada chorus from recordings containing bird sound. CicadaNet utilizes a convolutional encoder-decoder network to encode and decode acoustic features and a conformer module for global and local sequence modeling. We build a clean bird sound dataset and collect a large amount of real cicada noise data for model evaluation. We compare CicadaNet with current state-of-the-art deep denoising models and traditional denoising algorithms. Experimental results show that CicadaNet achieves the best denoising performance (SegSNR is improved by 9.59 dB and SI-SNR is improved by 20.08 dB when the noisy SNR = 0 dB). Meanwhile, CicadaNet achieves good performance for the real-time denoising of cicada noise. Furthermore, CicadaNet achieves bird species-independent noise reduction. We evaluate the effectiveness of CicadaNet for bird diversity survey. CicadaNet achieves the best performance, which can effectively eliminate the deviation caused by cicada noise to the acoustic index. CicadaNet can be easily extended to the cancellation of other environmental noise, and we propose it for the acoustic denoising of other vocalizing animals.

Swin Transformer for simultaneous denoising and interpolation of seismic data

Seismic data are often characterized by low quality due to noise contamination or missing traces. Convolutional neural networks are popular in dealing with denoising and interpolation. However, fixed-size convolution kernels have limited feature extraction range, and popular networks aim at either denoising or interpolating. In this paper, we propose a Swin Transformer convolutional residual network (SCRN) for simultaneous denoising and interpolation. We step-by-step show on synthetic datasets that SCRN can effectively denoise, interpolate, or denoise and interpolate simultaneously. Following this, the trained simultaneous denoising and interpolation model is directly used to handle multi-tasks on the field datasets. We assess the reconstruction effect of SCRN based on synthetic and field datasets. Comparison methods include both supervised ones (DnCNN, Unet and RIDNet) and unsupervised ones (DDUL, MultiResUNet and DL-POCS). Experimental results show that SCRN outperforms the counterparts in terms of (1) quantitative evaluation indices, (2) event continuity and weak signal retention, (3) generalization seismic data ability to suppress various noise levels and interpolate different sampling rates, and (4) preservation of first-arriving edge signals. These results indicate that SCRN can effectively reconstruct seismic data.

Flexible and accurate total variation and cascaded denoisers-based image reconstruction algorithm for hyperspectrally compressed ultrafast photography

Hyperspectrally compressed ultrafast photography (HCUP) based on compressed sensing and time- and spectrum-to-space mappings can simultaneously realize the temporal and spectral imaging of non-repeatable or difficult-to-repeat transient events with a passive manner in single exposure. HCUP possesses an incredibly high frame rate of tens of trillions of frames per second and a sequence depth of several hundred, and therefore plays a revolutionary role in single-shot ultrafast optical imaging. However, due to ultra-high data compression ratios induced by the extremely large sequence depth, as well as limited fidelities of traditional algorithms over the image reconstruction process, HCUP suffers from a poor image reconstruction quality and fails to capture fine structures in complex transient scenes. To overcome these restrictions, we report a flexible image reconstruction algorithm based on a total variation (TV) and cascaded denoisers (CD) for HCUP, named the TV-CD algorithm. The TV-CD algorithm applies the TV denoising model cascaded with several advanced deep learning-based denoising models in the iterative plug-and-play alternating direction method of multipliers framework, which not only preserves the image smoothness with TV, but also obtains more priori with CD. Therefore, it solves the common sparsity representation problem in local similarity and motion compensation. Both the simulation and experimental results show that the proposed TV-CD algorithm can effectively improve the image reconstruction accuracy and quality of HCUP, and may further promote the practical applications of HCUP in capturing high-dimensional complex physical, chemical and biological ultrafast dynamic scenes.

Addressing Information Inequality for Text-Based Person Search via Pedestrian-Centric Visual Denoising and Bias-Aware Alignments

Text-based person search is an important task in video surveillance, which aims to retrieve the corresponding pedestrian images with a given description. In this fine-grained retrieval task, accurate cross-modal information matching is an essential yet challenging problem. However, existing methods usually ignore the information inequality between modalities, which could introduce great difficulties to cross-modal matching. Specifically, in this task, the images inevitably contain some pedestrian-irrelevant noise like background and occlusion, and the descriptions could be biased to partial pedestrian content in images. With that in mind, in this paper, we propose a Text-Guided Denoising and Alignment (TGDA) model to alleviate the information inequality and realize effective cross-modal matching. In TGDA, we first design a prototype-based denoising module, which integrates pedestrian knowledge from textual features into a prototype vector and uses it as guidance to filter out pedestrian-irrelevant noise from visual features. Thereafter, a bias-aware alignment module is introduced, which guides our model to focus on the description-biased pedestrian content in cross-modal features consistently. Through extensive experiments, the effectiveness of both modules has been validated. Besides, our TGDA achieves state-of-the-art performance on various related benchmarks.

AdaRes: A deep learning-based model for ultrasound image denoising: Results of image quality metrics, radiomics, artificial intelligence, and clinical studies.

The quality of ultrasound images is degraded by speckle and Gaussian noises. This study aims to develop a deep-learning (DL)-based filter for ultrasound image denoising. A novel DL-based filter using adaptive residual (AdaRes) learning was proposed. Five image quality metrics (IQMs) and 27 radiomics features were used to evaluate denoising results. The effect of our proposed filter, AdaRes, on four pre-trained convolutional neural network (CNN) classification models and three radiologists was assessed. AdaRes filter was tested on both natural and ultrasound image databases. IQMs results indicate that AdaRes could remove noises in three different noise levels with the highest performances. In addition, a radiomics study proved that AdaRes did not distort tissue textures and it could preserve most radiomics features. AdaRes could also improve the performance classification using CNNs in different settings. Finally, AdaRes also improved the mean overall performance (AUC) of three radiologists from 0.494 to 0.702 in the classification of benign and malignant lesions. AdaRes filtered out noises on ultrasound images more effectively and can be used as an auxiliary preprocessing step in computer-aided diagnosis systems. Radiologists may use it to remove unwanted noises and improve the ultrasound image quality before the interpretation.

Deep Learning-Based Multifunctional End-to-End Model for Optical Character Classification and Denoising

Optical Character Recognition (OCR) has revolutionized document processing by converting scanned documents, PDFs, and images captured by cameras into editable and searchable text. This technology is crucial for digitizing historical documents, streamlining data entry processes, and improving accessibility for the visually impaired through text-to-speech technologies. Despite its widespread application, OCR faces significant challenges, especially in accurately recognizing text in noisy or degraded images. Traditionally, OCR systems have treated noise reduction and character classification as separate stages, which can compromise the overall effectiveness of text recognition. Our research introduces a groundbreaking Multifunctional End-to-End Model for Optical Character Classification and Denoising, which integrates these functions within a unified framework. By employing a dual-output autoencoder, our model concurrently denoises images and recognizes characters, thereby enhancing both the efficiency and accuracy of OCR. This paper outlines the model's development and implementation, explores the interplay between denoising and classification, and presents compelling experimental results that demonstrate marked improvements over conventional OCR methods.

Image denoising and segmentation model construction based on IWOA-PCNN

The research suggests a method to improve the present pulse coupled neural network (PCNN), which has a complex structure and unsatisfactory performance in image denoising and image segmentation. Then, a multi strategy collaborative improvement whale optimization algorithm (WOA) is proposed, and an improved whale optimization algorithm (IWOA) is constructed. IWOA is used to find the optimal parameter values of PCNN to optimize PCNN. By combining the aforementioned components, the IWOA-PCNN model had the best image denoising performance, and the produced images were crisper and preserve more information. IWOA-PCNN processed pictures have an average PSNR of 35.87 and an average MSE of 0.24. The average processing time for photos with noise is typically 24.80 s, which is 7.30 s and 7.76 s faster than the WTGAN and IGA-NLM models, respectively. Additionally, the average NU value measures 0.947, and the average D value exceeds 1000. The aforementioned findings demonstrate that the suggested method can successfully enhance the PCNN, improving its capability for image denoising and image segmentation. This can, in part, encourage the use and advancement of the PCNN.

Quantitative analysis of deep learning-based denoising model efficacy on optical coherence tomography images with different noise levels

BackgroundTo quantitatively evaluate the effectiveness of the Noise2Noise (N2N) model, a deep learning (DL)-based noise reduction algorithm, on enhanced depth imaging-optical coherence tomography (EDI-OCT) images with different noise levels. MethodsThe study included 30 subfoveal EDI-OCT images averaged with 100 frames from 30 healthy participants. Artificial Gaussian noise at 25.00, 50.00, and 75.00 standard deviations were added to the averaged (original) images, and the images were grouped as 25N, 50N, and 75N. Afterward, noise-added images were denoised with the N2N model and grouped as 25dN, 50dN, and 75dN, according to previous noise levels. The choroidal vascularity index (CVI) and deep choroidal contrast-to-noise ratio (CNR) were calculated for all images, and noise-added and denoised images were compared with the original images. The structural similarity of the noise-added and denoised images to the original images was assessed by the Multi-Scale Structural Similarity Index (MS-SSI). ResultsThe CVI and CNR parameters of the original images (68.08 ± 2.47 %, and 9.71 ± 2.80) did not differ from the only 25dN images (67.97 ± 2.34 % and 8.50 ± 2.43) (p:1.000, and p:0.062, respectively). Noise reduction improved the MS-SSI at each noise level (p < 0.001). However, the highest MS-SSI was achieved in 25dN images. ConclusionsThe DL-based N2N denoising model can be used effectively for images with low noise levels, but at increasing noise levels, this model may be insufficient to provide both the original structural features of the choroid and structural similarity to the original image.

Noise adaptive filtering model integrating spatio-temporal feature for soft sensor

Due to uncertain production conditions and external disturbances, there are a lot of noise in the industrial process dataset. Traditional denoising models can only solve a single noise problem. However, the noise in industrial process data has diversity and inhomogeneity. Therefore, the noise adaptive filtering (NAF) model integrating spatio-temporal feature (ST) (ST-NAF) is proposed for the soft sensor, which achieves a balance between noise and data features to reduce the influence of noise on soft sensor results. The gated recurrent unit extracts temporal features of the industrial process data. Then, the dynamically gated graph attention extracts the interaction relationship between monitoring points to obtain spatio-temporal features. Moreover, the NAF converts time-domain features into frequency-domain features, and adaptively selects frequency-domain components with important information through the noise filtering matrix. The selected frequency-domain components are converted to the temporal domain to obtain the denoised spatio-temporal features. Finally, the self-attention learns the dynamic relationships between denoised spatio-temporal features and output variables to predict the key indicator. Compared with other denoising soft sensor models in a public dataset and an actual industrial process dataset, the ST-NAF achieves state-of-the-art results. And the prediction accuracy of the proposed model improved by 4.5% and 10.0% in terms of R2, respectively. Furthermore, the experimental results show that the ST-NAF can adaptively select useful information from industrial noise data to achieve accurate prediction of key indicators.

Transformer-based deep learning denoising of single and multi-delay 3D arterial spin labeling.

To present a Swin Transformer-based deep learning (DL) model (SwinIR) for denoising single-delay and multi-delay 3D arterial spin labeling (ASL) and compare its performance with convolutional neural network (CNN) and other Transformer-based methods. SwinIR and CNN-based spatial denoising models were developed for single-delay ASL. The models were trained on 66 subjects (119 scans) and tested on 39 subjects (44 scans) from three different vendors. Spatiotemporal denoising models were developed using another dataset (6 subjects, 10 scans) of multi-delay ASL. A range of input conditions was tested for denoising single and multi-delay ASL, respectively. The performance was evaluated using similarity metrics, spatial SNR and quantification accuracy of cerebral blood flow (CBF), and arterial transit time (ATT). SwinIR outperformed CNN and other Transformer-based networks, whereas pseudo-3D models performed better than 2D models for denoising single-delay ASL. The similarity metrics and image quality (SNR) improved with more slices in pseudo-3D models and further improved when using M0 as input, but introduced greater biases for CBF quantification. Pseudo-3D models with three slices achieved optimal balance between SNR and accuracy, which can be generalized to different vendors. For multi-delay ASL, spatiotemporal denoising models had better performance than spatial-only models with reduced biases in fitted CBF and ATT maps. SwinIR provided better performance than CNN and other Transformer-based methods for denoising both single and multi-delay 3D ASL data. The proposed model offers flexibility to improve image quality and/or reduce scan time for 3D ASL to facilitate its clinical use.

Pest Detection Using Image Denoising and Cascaded Unet Segmentation for Pest Images

This study proposes a novel approach for pest detection in pest images using image denoising and cascaded UNET segmentation. The proposed approach involves the use of a hybrid neural network RESNET50 with CNN for image dataset training, optimized using Believed Adam Optimization. The images are then preprocessed using a superior MLP model for image denoising, which enhances the image quality and reduces the noise present in the image. The images are then passed through an adaptive UNET architecture for image segmentation, which is based on domain adaptation and semantic segmentation. The cascaded UNET segmentation improves the segmentation accuracy, and the domain adaptation ensures that the model can be applied to new datasets without requiring additional training. The proposed approach achieves a high accuracy rate of 98.5% in detecting pests images. This approach can be used in various applications related to pest detection and management, including agriculture and pest control.

Reducing the risk of hallucinations with interpretable deep learning models for low-dose CT denoising: comparative performance analysis

Objective. Reducing CT radiation dose is an often proposed measure to enhance patient safety, which, however results in increased image noise, translating into degradation of clinical image quality. Several deep learning methods have been proposed for low-dose CT (LDCT) denoising. The high risks posed by possible hallucinations in clinical images necessitate methods which aid the interpretation of deep learning networks. In this study, we aim to use qualitative reader studies and quantitative radiomics studies to assess the perceived quality, signal preservation and statistical feature preservation of LDCT volumes denoised by deep learning. We aim to compare interpretable deep learning methods with classical deep neural networks in clinical denoising performance. Approach. We conducted an image quality analysis study to assess the image quality of the denoised volumes based on four criteria to assess the perceived image quality. We subsequently conduct a lesion detection/segmentation study to assess the impact of denoising on signal detectability. Finally, a radiomic analysis study was performed to observe the quantitative and statistical similarity of the denoised images to standard dose CT (SDCT) images. Main results. The use of specific deep learning based algorithms generate denoised volumes which are qualitatively inferior to SDCT volumes(p < 0.05). Contrary to previous literature, denoising the volumes did not reduce the accuracy of the segmentation (p > 0.05). The denoised volumes, in most cases, generated radiomics features which were statistically similar to those generated from SDCT volumes (p > 0.05). Significance. Our results show that the denoised volumes have a lower perceived quality than SDCT volumes. Noise and denoising do not significantly affect detectability of the abdominal lesions. Denoised volumes also contain statistically identical features to SDCT volumes.

Unsupervised domain alignment of fingerprint denoising models using pseudo annotations

Unsupervised domain alignment of fingerprint denoising models using pseudo annotations

Multiplicative noise removal using complementary total generalized variation and nonlocal low rank regularization

The problem of multiplicative image denoising is a typical ill-posed inverse problem which has been extensively studied in the past few decades and is still an open problem. Exploiting appropriate image priors is critical to solve ill-posed inverse image restoration problem. The existing total generalized variation (TGV) based approaches have achieved significant performance in removing the multiplicative noise. However, these TGV regularized approaches only consider the local smoothness prior of the potential clean image, which limits their ability for image restoration. Other than the local smoothness property, the nonlocal self-similarity prior of images can also be exploited to offer helpful remedy for better image reconstruction. In this paper, to simultaneously make use of both the local smoothness prior and nonlocal self-similarity prior, we integrate the TGV regularizer with the nonlocal low rank regularizer to propose a novel variational model for multiplicative image denoising. Specifically, in the objective function, the nonlocal low rank regularization is expressed by the celebrated weighted nuclear norm minimization (WNNM) which has recently been applied to image denoising and presents state-of-art result. Then, we apply the split Bregman algorithm to efficiently solve the resulting model. Numerical experiments demonstrate that the proposed method achieve superior performance over the competing methods, including the state-of-the-art SAR-BM3D method.

Joint Low-Light Image Enhancement and Denoising via a New Retinex-Based Decomposition Model

It is well known that images taken in low-light conditions frequently suffer from unknown noise and low visibility, which can pose challenges for image enhancement. The majority of Retinex-based decomposition algorithms usually attempt to directly design prior regularization for illumination or reflectance. Nevertheless, noise can be involved in such schemes. To address these issues, a new Retinex-based decomposition model for simultaneous enhancement and denoising has been developed. In this paper, an extended decomposition scheme is introduced to extract the illumination and reflectance components, which helps to better describe the prior information on illumination and reflectance. Subsequently, spatially adaptive weights are designed for two regularization terms. The main motivation is to provide a small amount of smoothing in near edges or bright areas and stronger smoothing in dark areas, which could preserve useful information and remove noise effectively during image-enhancement processing. Finally, the proposed algorithm is validated on several common datasets: LIME, LOL, and NPE. Extensive experiments show that the presented method is superior to state-of-the-art methods both in objective index comparisons and visual quality.

Time Image De-Noising Method Based on Sparse Regularization

The blurring of texture edges often occurs during image data transmission and acquisition. To ensure the detailed clarity of the drag-time images, we propose a time image de-noising method based on sparse regularization. First, the image pixel sparsity index is set, and then an image de-noising model is established based on sparse regularization processing to obtain the neighborhood weights of similar image blocks. Second, a time image de-noising algorithm is designed to determine whether the coding coefficient reaches the standard value, and a new image de-noising method is obtained. Finally, the images of electronic clocks and mechanical clocks are used as two kinds of time images to compare different image de-noising methods, respectively. The results show that the sparsity regularization method has the highest peak signal-to-noise ratio among the six compared methods for different noise standard deviations and two time images. The image structure similarity is always above which shows that the proposed method is better than the other five image de-noising methods.