Search

Image denoising and super-resolution play vital roles in imaging systems, greatly reducing the preprocessing cost of many AI techniques for object detection, segmentation, and tracking. Various advancements have been accomplished in this field, but progress is still needed. In this paper, we have proposed a novel technique named the Enhanced Learning Enriched Features (ELEF) mechanism using a deep convolutional neural network, which makes significant improvements to existing techniques. ELEF consists of two major processes: (1) Denoising, which removes the noise from images; and (2) Super-resolution, which improves the clarity and details of images. Features are learned through deep CNN and not through traditional algorithms so that we can better refine and enhance images. To effectively capture features, the network architecture adopted Dual Attention Units (DUs), which align with the Multi-Scale Residual Block (MSRB) for robust feature extraction, working sidewise with the feature-matching Selective Kernel Extraction (SKF). In addition, resolution mismatching cases are processed in detail to produce high-quality images. The effectiveness of the ELEF model is highlighted by the performance metrics, achieving a Peak Signal-to-Noise Ratio (PSNR) of 42.99 and a Structural Similarity Index (SSIM) of 0.9889, which indicates the ability to carry out the desired high-quality image restoration and enhancement. [ABSTRACT FROM AUTHOR]

Infrared images are susceptible to interference from a variety of factors during acquisition and transmission, resulting in the inclusion of mixed noise, which seriously affects the accuracy of subsequent vision tasks. To solve this problem, we designed a mixed noise removal algorithm for infrared images based on improved CycleGAN. First, we proposed a ResNet-E Block that incorporates the EMA (Efficient Multi-Scale Attention Module) and build a generator based on it using the skip-connection structure to improve the network's ability to remove mixed noise of different strengths. Second, we added the PSNR (Peak Signal-to-Noise Ratio) as an extra calculation item of cycle consistency loss, so that the network can effectively retain the detailed information of infrared images while denoising. Finally, we conducted experimental validation on both synthetic noisy images and real noisy images, which proved that our algorithm can effectively remove the mixed noise in infrared images and the denoising effect is better than other similar methods. [ABSTRACT FROM AUTHOR]

Demosaicking is a popular scientific area that is being explored by a vast number of scientists. Current digital imaging technologies capture colour images with a single monochrome sensor. In addition, the colour images were captured using a sensor coupled with a Colour Filter Array (CFA). Furthermore, the demosaicking procedure is required to obtain a full-colour image. Image denoising and image demosaicking are the two important image restoration techniques, which have increased popularity in recent years. Finding a suitable strategy for multiple image restoration is critical for researchers. Hence, a deep learning (DL) based image denoising and image demosaicking is developed in this research. Moreover, the Autoregressive Circle Wave Optimization (ACWO) based Demosaicking Convolutional Neural Network (DMCNN) is designed for image demosaicking. The Quantum Wavelet Transform (QWT) is used in the image denoising process. Similarly, Quantum Wavelet Transform (QWT) is used to analyse the abrupt changes in the input image with noise. The transformed image is then subjected to a thresholding technique, which determines an appropriate threshold range. Once the threshold range has been determined, soft thresholding is applied to the resulting wavelet coefficients. After that, the extraction and reconstruction of the original image is carried out using the Inverse Quantum Wavelet Transform (IQWT). Finally, the fused image is created by combining the results of both processes using a weighted average. The denoised and demosaicked images are combined using the weighted average technique. Furthermore, the proposed QWT+DMCNN-ACWO model provided the ideal values of Peak signal-to-noise ratio (PSNR), Second derivative like measure of enhancement (SDME), Structural Similarity Index (SSIM), Figure of Merit (FOM) of 0.890, and computational time of 49.549 dB, 59.53 dB, 0.963, 0.890, and 0.571, respectively. [ABSTRACT FROM AUTHOR]

Background: Residual image noise is substantial in positron emission tomography (PET) and one of the factors limiting lesion detection, quantification, and overall image quality. Thus, improving noise reduction remains of considerable interest. This is especially true for respiratory-gated PET investigations. The only broadly used approach for noise reduction in PET imaging has been the application of low-pass filters, usually Gaussians, which however leads to loss of spatial resolution and increased partial volume effects affecting detectability of small lesions and quantitative data evaluation. The bilateral filter (BF) — a locally adaptive image filter — allows to reduce image noise while preserving well defined object edges but manual optimization of the filter parameters for a given PET scan can be tedious and time-consuming, hampering its clinical use. In this work we have investigated to what extent a suitable deep learning based approach can resolve this issue by training a suitable network with the target of reproducing the results of manually adjusted case-specific bilateral filtering. Methods: Altogether, 69 respiratory-gated clinical PET/CT scans with three different tracers ( [ 18 F ] FDG, [ 18 F ] L-DOPA, [ 68 Ga ] DOTATATE) were used for the present investigation. Prior to data processing, the gated data sets were split, resulting in a total of 552 single-gate image volumes. For each of these image volumes, four 3D ROIs were delineated: one ROI for image noise assessment and three ROIs for focal uptake (e.g. tumor lesions) measurements at different target/background contrast levels. An automated procedure was used to perform a brute force search of the two-dimensional BF parameter space for each data set to identify the "optimal" filter parameters to generate user-approved ground truth input data consisting of pairs of original and optimally BF filtered images. For reproducing the optimal BF filtering, we employed a modified 3D U-Net CNN incorporating residual learning principle. The network training and evaluation was performed using a 5-fold cross-validation scheme. The influence of filtering on lesion SUV quantification and image noise level was assessed by calculating absolute and fractional differences between the CNN, manual BF, or original (STD) data sets in the previously defined ROIs. Results: The automated procedure used for filter parameter determination chose adequate filter parameters for the majority of the data sets with only 19 patient data sets requiring manual tuning. Evaluation of the focal uptake ROIs revealed that CNN as well as BF based filtering essentially maintain the focal SUV max values of the unfiltered images with a low mean ± SD difference of δ SUV max CNN , STD = (−3.9 ± 5.2)% and δ SUV max BF , STD = (−4.4 ± 5.3)%. Regarding relative performance of CNN versus BF, both methods lead to very similar SUV max values in the vast majority of cases with an overall average difference of δ SUV max CNN , BF = (0.5 ± 4.8)%. Evaluation of the noise properties showed that CNN filtering mostly satisfactorily reproduces the noise level and characteristics of BF with δ Noise CNN , BF = (5.6 ± 10.5)%. No significant tracer dependent differences between CNN and BF were observed. Conclusions: Our results show that a neural network based denoising can reproduce the results of a case by case optimized BF in a fully automated way. Apart from rare cases it led to images of practically identical quality regarding noise level, edge preservation, and signal recovery. We believe such a network might proof especially useful in the context of improved motion correction of respiratory-gated PET studies but could also help to establish BF-equivalent edge-preserving CNN filtering in clinical PET since it obviates time consuming manual BF parameter tuning. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) can effectively extract structural information in image denoising. However, they tend to ignore texture information. To tackle this problem, we present a texture-guided CNN for image denoising (TDCNN), which depends on blocks for texture extraction, refinement, and transformation to realize excellent denoising performance on both quantitative and visual metrics. A texture-extraction block combines non-local similarity and two sub-networks to extract texture and structural information. A refinement block with a stacked architecture mines accurate information from complementary features. A transformation block is used to obtain clean output images. A joint loss function, including perceptual loss and mean square error, enhances the robustness of the proposed denoiser. Experiments show that the proposed TDCNN is superior to some popular methods for denoising synthetic and real images. [ABSTRACT FROM AUTHOR]

Digital images are corrupted with noise, and image denoising is an important step in image processing modules. In this review, the latest developments in filtering methods for color image restoration are analyzed. These algorithms are compared in terms of objective image quality measures and divided into major classes, such as spatial domain, switching and wavelet filtering methods. These classes are based on the particular methodology used in image denoising algorithms and further subdivided to show their classification in terms of noise models utilized, application style, and stages the filters applied in images. In particular, we present a review of filtering methods in color image denoising, published over the past two decades. Our classification and succinct descriptions of color image restoration by these mathematical filtering techniques and their characterizations can help choose the appropriate ones for various downstream image processing tasks. [ABSTRACT FROM AUTHOR]

The traditional methods for 3D reconstruction mainly involve using image processing techniques or deep learning segmentation models for rib extraction. After post-processing, voxel-based rib reconstruction is achieved. However, these methods suffer from limited reconstruction accuracy and low computational efficiency. To overcome these limitations, this paper proposes a 3D rib reconstruction method based on point cloud adaptive smoothing and denoising. We converted voxel data from CT images to multi-attribute point cloud data. Then, we applied point cloud adaptive smoothing and denoising methods to eliminate noise and non-rib points in the point cloud. Additionally, efficient 3D reconstruction and post-processing techniques were employed to achieve high-accuracy and comprehensive 3D rib reconstruction results. Experimental calculations demonstrated that compared to voxel-based 3D rib reconstruction methods, the 3D rib models generated by the proposed method achieved a 40% improvement in reconstruction accuracy and were twice as efficient as the former. [ABSTRACT FROM AUTHOR]

Image denoising is a significant procedure in image processing, and a good denoising method can improve the image quality. This study focuses on enhancing the estimation model of wavelet coefficients by improving the threshold function. Thus, a general construction method for a class of threshold functions is introduced, and sufficient conditions are given to ensure that the threshold functions are continuous, first-order differentiable and higher-order differentiable. It makes an existing class of improved threshold functions a particular case and enables the construction of other needed threshold functions based on this general approach. Experiments have shown that using the new threshold functions yields better denoising results for images containing salt and pepper noise, Gaussian noise and speckle noise. Moreover, as the smoothness of the threshold function is enhanced, the denoising effect is also improved. The improved wavelet coefficients estimation model can effectively boost the quality of image denoising, further enhancing the accuracy and effectiveness of computer vision algorithms. [ABSTRACT FROM AUTHOR]

In this paper, we propose a new image denoising technique which consists in applying a stationary wavelet transform (SWT)-based image denoising technique, in the domain of 2D dual-tree discrete wavelet transform (DWT). In fact, this proposed technique consists first of applying the 2D dual-tree DWT to the noisy image. Then, the noisy wavelet coefficients obtained from this application are denoised by applying to each of them, a SWT-based image denoising technique. Finally, the denoised image is reconstructed by applying the inverse of the 2D dual-tree DWT to the obtained denoised wavelet coefficients. For applying this SWT-based image denoising technique, we use the soft thresholding, the Daubechies 4 as the mother wavelet, and the decomposition level is equal to 5. The performance of this image denoising technique proposed in this work is proven by its comparison to three other denoising techniques existing in the literature. These three techniques are the denoising technique based on the soft thresholding in the SWT domain, the image denoising technique based on soft thresholding in the domain of 2D dual-tree DWT and the image denoising approach using deep neural network. All the previously mentioned techniques, including our proposed denoising approach, are applied to a number of noisy images, and the obtained results are in terms of peak signal-to-noise ratio (PSNR) and structural similarity (SSIM). Those results show that this proposed denoising technique outperforms the other three denoising techniques used in this evaluation. [ABSTRACT FROM AUTHOR]

In this paper, a novel non-convex fractional-order image denoising model is proposed to suppress the staircase effect produced by the TV model while maintaining a neat contour. The model combines ℓ q (0 < q < 1) quasi-norm and fractional-order regularization, and employs a diffusion coefficient with a faster convergence rate to preserve more image edges and details. Additionally, an adaptive regularization parameter is designed to adjust the denoising performance of the algorithm. To obtain the optimal approximate solution of the model, an enhanced primal-dual algorithm is adopted and the complexity and convergence of the algorithm are theoretically analyzed. Finally, the effectiveness of the proposed method is demonstrated through numerical experiments. • A new non-convex FOTV model is proposed. • A new diffusion coefficient is introduced to retain more edge details of the image. • The existing primal-dual algorithm is improved, and the convergence of this algorithm is analyzed. • An adaptive regularization parameter is designed. [ABSTRACT FROM AUTHOR]

Scanning electron microscope (SEM) enables imaging of micro-nano scale objects. It is an analytical tool widely used in the material, earth and life sciences. However, SEM images often suffer from high noise levels, influenced by factors such as dwell time, the time during which the electron beam remains per pixel during acquisition. Slower dwell times reduce noise but risk damaging the sample, while faster ones introduce uncertainty. To this end, the latest state-of-the-art denoising techniques must be explored. Experimentation is crucial to identify the most effective methods that balance noise reduction and sample preservation, ensuring high-quality SEM images with enhanced clarity and accuracy. A thorough analysis tracing the evolution of image denoising techniques was conducted, ranging from classical methods to deep learning approaches. A comprehensive taxonomy of this reverse problem solutions was established, detailing the developmental flow of these methods. Subsequently, the latest state-of-the-art techniques were identified and reviewed based on their reproducibility and the public availability of their source code. The selected techniques were then tested and investigated using scanning electron microscope images. After in-depth analysis and benchmarking, it is clear that the existing deep learning-based denoising techniques fall short in maintaining a balance between noise reduction and preserving crucial information for SEM images. Issues like information removal and over-smoothing have been identified. To address these constraints, there is a critical need for the development of SEM image denoising techniques that prioritize both noise reduction and information preservation. Additionally, one can see that the combination of several networks, such as the generative adversarial network and the convolutional neural network (CNN), known as BoostNet, or the vision transformer and the CNN, known as SCUNet, improves denoising performance. It is recommended to use blind techniques to denoise real noise while taking into account detail preservation and tackling excessive smoothing, particularly in the context of SEM. In the future the use of explainable AI will facilitate the debugging and the identification of these problems. [ABSTRACT FROM AUTHOR]

Image denoising is crucial for enhancing image quality, improving visual effects, and boosting the accuracy of image analysis and recognition. Most of the current image denoising methods perform superior on synthetic noise images, but their performance is limited on real-world noisy images since the types and distributions of real noise are often uncertain. To address this challenge, a multi-scale information fusion generative adversarial network method is proposed in this paper. Specifically, In this method, the generator is an end-to-end denoising network that consists of a novel encoder–decoder network branch and an improved residual network branch. The encoder–decoder branch extracts rich detailed and contextual information from images at different scales and utilizes a feature fusion method to aggregate multi-scale information, enhancing the feature representation performance of the network. The residual network further compensates for the compressed and lost information in the encoder stage. Additionally, to effectively aid the generator in accomplishing the denoising task, convolution kernels of various sizes are added to the discriminator to improve its image evaluation ability. Furthermore, the dual denoising loss function is presented to enhance the model’s capability in performing noise removal and image restoration. Experimental results show that the proposed method exhibits superior objective performance and visual quality than some state-of-the-art methods on three real-world datasets. [ABSTRACT FROM AUTHOR]

Image-space denoising of rendered images has become a commonly adopted approach since this post-rendering process often drastically reduces required sample counts (thus rendering times) for producing a visually pleasing image without noticeable noise. It is a common practice to conduct such denoising while preserving image details by exploiting auxiliary information (e.g., G-buffers) as well as input colors. However, it is still challenging to devise an ideal denoising framework that fully considers the two inputs with different characteristics, e.g., noisy but complete shading information in the input colors and less noisy but partial shading information in the auxiliary buffers. This paper proposes a transformer-based denoising framework with a new self-attention mechanism that infers a joint self-attention map, i.e., self-similarity in input features, through dual attention scores: one from noisy colors and another from auxiliary buffers. We demonstrate that this separate consideration of the two inputs allows our framework to produce more accurate denoising results than state-of-the-art denoisers for various test scenes. [ABSTRACT FROM AUTHOR]

In the process of data generation or transmission, the quality of data may degrade and not meet the required level for subsequent processing steps. Improving data quality is one of the crucial steps that needs to be taken to obtain accurate information hidden within the data in any field. Researchers have proposed various methods to perform this process, which differ based on the type of data. However, it is important to note that often these methods do not consider the existing similarities in different dimensions of the data simultaneously. This can have an undesirable or detrimental impact on certain parts of the data and may not improve the damaged segments. As a result, the obtained output will not contain all the desired information. In this paper, a new method is introduced in which data quality improvement is carried out using a set of collaborative nodes in an interactive network structure. This method enhances resistance against various types of degradation by employing a set of nodes. The performance of the proposed method is compared with six other state-of-the-art data quality improvement methods on real degraded datasets. The results obtained from the simulation show that the proposed method outperforms the other compared methods. [ABSTRACT FROM AUTHOR]

Artificial intelligence (AI) in positron emission tomography/computed tomography (PET/CT) can be used to improve image quality when it is useful to reduce the injected activity or the acquisition time. Particular attention must be paid to ensure that users adopt this technological innovation when outcomes can be improved by its use. The aim of this study was to identify the aspects that need to be analysed and discussed to implement an AI denoising PET/CT algorithm in clinical practice, based on the representations of Nuclear Medicine Technologists (NMT) from Western-Switzerland, highlighting the barriers and facilitators associated. Two focus groups were organised in June and September 2023, involving ten voluntary participants recruited from all types of medical imaging departments, forming a diverse sample of NMT. The interview guide followed the first stage of the revised model of Ottawa of Research Use. A content analysis was performed following the three-stage approach described by Wanlin. Ethics cleared the study. Clinical practice, workload, knowledge and resources were de 4 themes identified as necessary to be thought before implementing an AI denoising PET/CT algorithm by ten NMT participants (aged 31–60), not familiar with this AI tool. The main barriers to implement this algorithm included workflow challenges, resistance from professionals and lack of education; while the main facilitators were explanations and the availability of support to ask questions such as a "local champion". To implement a denoising algorithm in PET/CT, several aspects of clinical practice need to be thought to reduce the barriers to its implementation such as the procedures, the workload and the available resources. Participants emphasised also the importance of clear explanations, education, and support for successful implementation. To facilitate the implementation of AI tools in clinical practice, it is important to identify the barriers and propose strategies that can mitigate it. [ABSTRACT FROM AUTHOR]

The generation of photogrammetric point clouds from satellite images is often based on image correlation techniques. Correlation errors can arise for a wide variety of reasons: transient objects, homogeneous areas, shadows, and surface discontinuities. Therefore, a simple 3D Gaussian distribution at the point cloud level is not an appropriate model. In this paper, we propose a new point cloud denoising method integrated into the Multiview Stereo Pipeline CARS, dedicated to satellite imagery. Building upon bilateral filtering principles, our approach introduces a novel utilization of color information, confidence estimation and geometric constraints alongside point positions and normals. While the use of point color increases the level of detail, the addition of geometric constraints and confidence awareness guides processing towards a realistic solution. We propose an ablation study and compare our solution against a previously established bilateral filter with LiDAR data as ground truth. [ABSTRACT FROM AUTHOR]

Medical image denoising plays a critical role in enhancing the quality of diagnostic imaging, where noise reduction without compromising image details is paramount. In this paper, we propose a novel hybrid approach aimed at improving the denoising efficacy for medical images. Initially, we employ an adaptive Kalman filter to attenuate noise, leveraging its proficiency in state estimation from noisy measurements. Unlike conventional Kalman filters with fixed parameters, our adaptive Kalman filter dynamically adjusts its parameters based on the noise characteristics of the input image, thus offering enhanced accuracy in estimating the underlying true state of the system represented by the medical image. Subsequently, both a non-local means (NLM) method and a median filter are introduced as post-processing steps to further refine the denoised image. The NLM method leverages the similarities between image patches to effectively reduce noise, while the median filter further enhances the denoised image by suppressing residual noise and preserving image details. However, the effectiveness of NLM and the median filter is highly dependent on carefully chosen parameters, which traditionally necessitates extensive computational resources for optimization. To address this challenge, we introduce the innovative use of Latin square optimization, a structured experimental design technique, to efficiently determine optimal parameters for NLM. By systematically exploring parameter combinations using Latin square optimization, we mitigate the complexity of experiments while enhancing denoising performance. The experimental results on medical images demonstrate the effectiveness of our proposed approach, showcasing significant improvements in noise reduction and the preservation of image features compared to conventional methods. Our hybrid approach not only advances the state-of-the-art in medical image denoising but also presents a practical solution for optimizing parameter selection in NLM, thereby facilitating their broader adoption in medical imaging applications. [ABSTRACT FROM AUTHOR]

Patch level-based noise level estimation (NLE) is often inaccurate and inefficient because of the harsh criteria required to select a small number of homogeneous patches. In this paper, a fast image NLE method based on a global search for similar pixels is proposed to solve the above problem. Specifically, the mean square distance (MSD) is first expressed in the form of the standard deviation (std) and mean value of image patches. Afterward, the two values, std and mean, are calculated and stored in advance. Then, a 2D statistical histogram and summed area table are adopted to speed up the search for similar patches. Further, the most similar pixels are selected from similar patches to obtain an initial estimation. Finally, we correct the deviation of the initial estimation by re-injecting noise for secondary estimation. Experimental results show that the proposed method outperforms the state-of-the-art techniques in fast NLE and guided denoising. [ABSTRACT FROM AUTHOR]

Depth maps produced by consumer-grade sensors suffer from inaccurate measurements and missing data from either system or scene-specific sources. Data-driven denoising algorithms can mitigate such problems; however, they require vast amounts of ground truth depth data. Recent research has tackled this limitation using self-supervised learning techniques, but it requires multiple RGB-D sensors. Moreover, most existing approaches focus on denoising single isolated depth maps or specific subjects of interest highlighting a need for methods that can effectively denoise depth maps in real-time dynamic environments. This paper extends state-of-the-art approaches for depth-denoising commodity depth devices, proposing SelfReDepth, a self-supervised deep learning technique for depth restoration, via denoising and hole-filling by inpainting of full-depth maps captured with RGB-D sensors. The algorithm targets depth data in video streams, utilizing multiple sequential depth frames coupled with color data to achieve high-quality depth videos with temporal coherence. Finally, SelfReDepth is designed to be compatible with various RGB-D sensors and usable in real-time scenarios as a pre-processing step before applying other depth-dependent algorithms. Our results demonstrate our approach’s real-time performance on real-world datasets shows that it outperforms state-of-the-art methods in denoising and restoration performance at over 30 fps on Commercial Depth Cameras, with potential benefits for augmented and mixed-reality applications. [ABSTRACT FROM AUTHOR]

Image denoising presents a significant challenge in computer vision, aiming to eliminate unwanted noise from images and restore their original quality. Cauchy noise, characterized by its heavy-tailed distribution, poses a unique hurdle among various types of noise. While several denoising models have been proposed, Total Variation (TV)-based methods, such as the Remove Cauchy Noise model, are widely used for their effectiveness. However, these approaches often suffer from issues such as oversmoothing, staircase artifacts, and the introduction of false details. To address these limitations, recent research has explored the combination of high-order TV with Overlapping Group Sparsity (OGS), showing promising results in noise removal. Inspired by this, our article introduces a novel Cauchy denoising model. Our approach leverages OGS and directional higher-order TV to effectively remove Cauchy noise while preserving image details and minimizing aliasing and smoothing artifacts. The Chambolle-Pock algorithm efficiently solves the underlying optimization problem. Through qualitative and quantitative evaluations, including visualization and parameter measurements, we demonstrate the competitiveness of our model compared to existing methods. [ABSTRACT FROM AUTHOR]

Recovering high-quality inscription images from unknown and complex inscription noisy images is a challenging research issue. Different from natural images, character images pay more attention to stroke information. However, existing models mainly consider pixel-level information while ignoring structural information of the character, such as its edge and glyph, resulting in reconstructed images with mottled local structure and character damage. To solve these problems, we propose a novel generative adversarial network (GAN) framework based on an edge-guided generator and a discriminator constructed by a dual-domain U-Net framework, i.e. EDU-GAN. Unlike existing frameworks, the generator introduces the edge extraction module, guiding it into the denoising process through the attention mechanism, which maintains the edge detail of the restored inscription image. Moreover, a dual-domain U-Net-based discriminator is proposed to learn the global and local discrepancy between the denoised and the label images in both image and morphological domains, which is helpful to blind denoising tasks. The proposed dual-domain discriminator and generator for adversarial training can reduce local artifacts and keep the denoised character structure intact. Due to the lack of a real-inscription image, we built the real-inscription dataset to provide an effective benchmark for studying inscription image denoising. The experimental results show the superiority of our method both in the synthetic and real-inscription datasets. [ABSTRACT FROM AUTHOR]

Nano-computed tomography (Nano-CT) is an emerging, high-resolution imaging technique. However, due to their low-light properties, tabletop Nano-CT has to be scanned under long exposure conditions, which the scanning process is time-consuming. For 3D reconstruction data, this paper proposed a lightweight 3D noise reduction method for desktop-level Nano-CT called AAD-ResNet (Axial Attention DeNoise ResNet). The network is framed by the U-net structure. The encoder and decoder are incorporated with the proposed 3D axial attention mechanism and residual dense block. Each layer of the residual dense block can directly access the features of the previous layer, which reduces the redundancy of parameters and improves the efficiency of network training. The 3D axial attention mechanism enhances the correlation between 3D information in the training process and captures the long-distance dependence. It can improve the noise reduction effect and avoid the loss of image structure details. Experimental results show that the network can effectively improve the image quality of a 0.1-s exposure scan to a level close to a 3-s exposure, significantly shortening the sample scanning time. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Frontiers of Computer Science & Technology is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The existence of noise components will affect the quality of image super-resolution reconstruction, so an image super-resolution reconstruction method based on improved cyclic generative countermeasure network is proposed. Using the image denoising regularization method, the internal noise of the original image is removed. By introducing the twin attention mechanism into the cyclic generative countermeasure network, an improved cyclic generative countermeasure network is obtained. In the improved cyclic generative countermeasure network, the twin attention model is used to extract the denoised image features, and the super-resolution reconstruction image is generated with the generator. The network discriminator is used to identify whether the reconstructed image is a real image, and the output identification result is a real image to obtain the relevant image super-resolution reconstruction results. Experiments show that this method can effectively denoise the original image and extract image features, and can also reconstruct the image with high quality to improve image resolution and clarity. At different image magnifications, the structural similarity of image reconstruction using this method is high. The subjective opinion score of the image super-resolution reconstruction result of this method is high, with a maximum score of 4.8. The perception index and Frachet inception distance are both small, with values of 21.65 and 14.84, respectively. The image super-resolution reconstruction effect is good. [ABSTRACT FROM AUTHOR]

Convolutional neural networks (CNNs) have gained significant popularity in image denoising. In recent years, many CNN-based image denoising methods have been developed. However, some of these methods extract the noise from noisy images by only stacking numerous relatively common convolutional layers, which can easily lead to some issues, including more loss of image details and insufficient processing of complex real-world noisy images. To address these challenges more effectively, a new progressive multi-scale denoising network (PMSDNet) is proposed. Its denoising effectiveness is attributed to two key modules, namely progressive multi-scale fusion block (PMSFB) and pixel attention block (PAB), in which PMSFB enhances the noise extraction capability of PMSDNet by enlarging its receptive field and reducing important image feature information loss, and PAB further facilitates PMSDNet to effectively capture the noise and preserve more image details by guiding it to focus more on multi-noised pixels and high-frequency image regions. Experimental results demonstrate the inspiring denoising performance of PMSDNet in facing three types of image noises, i.e., non-blind synthetic noise, blind synthetic noise and real-world noise. Additionally, the PMSDNet is extended to single-image deraining, showing the promising deraining performance. [ABSTRACT FROM AUTHOR]

The total variation method has been widely used because it can preserve important sharp edges and target boundaries in the image, but one of its shortcomings is that the L 1 norm would cause excessive punishment. To overcome excessive penalization, suppress step effects, and maintain a clean contour, this article introduces the L q non-convex function to design a new regularization term and proposes a new non-convex variational model. For non-convex variational optimization problems, the algorithm employs the alternating direction method of multipliers to obtain optimal approximate solutions, and rigorous convergence proofs are provided based on rich mathematical theory. By comparing with advanced fractional-order and other total variation recovery algorithms, it can be observed that this algorithm achieves better visual effects and quantitative analysis results on images with complex texture structures. [ABSTRACT FROM AUTHOR]

Non-local means (NL-Means) algorithm which removes the noise from the image have been used in the field widely due to its good performance especially for magnetic resonance images which consists of three dimensional data. Its main idea is using all the pixels which are local and non-local in an image and taking weighted averaging of all values. One negative side of this method is that it considers all pixels in the image without looking at their similarity. This paper proposes an NL-Means algorithm with pixel selection by applying linear regression analysis using root mean squared error (RMSE) value. After regression analysis, RMSE of the neighborhoods is used to exclude non-similar pixels during the noise removal. Lastly, obtained results were compared by four different methods which are NL-Means algorithm and, Gaussian, anisotropic diffusion and median filterings. All of the methods were outperformed by our method on structured similarity index and peak signal-to-noise ratio quantitative metrics. Moreover, the level of increase on visual qualities are also represented as a qualitative analysis. [ABSTRACT FROM AUTHOR]

A gear box fault diagnosis method based on improved time series gray image and deep learning is put forward in order to solve the problems such as complex actual working environment of gear box, inadequate performance of extracting features by traditional methods and gray image extraction features. After the vibration signal is decomposed into several intrinsic mode components (IMFs) by EEMD, the IMFs are divided into high-frequency and low-frequency components by cumulative mean criterion, in which the high-frequency components are denoised by wavelet threshold. The high frequency IMFs and low frequency IMFs after noise reduction are reconstructed, and the reconstructed signal is coded by gray image method. Two-dimensional improved time series gray image is sent to convolution neural network for training, so as to exert the advantages of convolution network in feature extraction of pictures and display the results by confusion matrix. Finally, the model results and different gray scale images are compared with traditional diagnosis methods. The results show that compared with common gray image and global denoising gray image, the method proposed in this paper improves the accuracy of gear box fault diagnosis by 4 and 1.8 percentage points respectively, and the convergence speed is significantly faster. Compared with BP neural network and ELM diagnosis method, the method proposed in this paper significantly improves the accuracy of gear box fault diagnosis. [ABSTRACT FROM AUTHOR]

Image denoising faces significant challenges, arising from the sources of noise. Specifically, Gaussian, impulse, speckle noises, salt and pepper noise are complicated sources of noise in imaging. Convolutional neural network (CNN) has increasingly received attention in image denoising task. Due to the fast inference and good performance, discriminative learning methods have been widely studied in image denoising. However, these methods mostly learn a specific model for each noise level and require multiple models for denoising images with different noise levels. They also lack flexibility to deal with spatially variant noise, limiting their applications in practical denoising. To address these issues, we propose a deep iterative down-up CNN for image denoising, which alternates between lowering and raising the determination of the feature maps. The proposed network performed on down-sampled sub-images, achieving a good trade-off between inference speed and denoising performance. In contrast to the existing denoisers, the proposed network has several desirable properties, including (i) the ability to handle a wide range of noise levels (i.e., [0, 75]) effectively with a single network and (ii) the ability to remove spatially variant noise by specifying a non-uniform noise level map. In addition, train a network model for five different types of noises, and achieve better performance. Extensive experiments on synthetic and real noisy images are conducted to evaluate the proposed model in comparison with state-of-the-art denoisers. The results show that the proposed model is effective and efficient, making it highly attractive for practical denoising applications. [ABSTRACT FROM AUTHOR]

Uncontrollable growth of cells in the lung tissue is the condition that results in lung cancer. Early screening is ideal for lung cancer detection and treatment to prevent these kinds of disorders. The proposed model performed lung cancer detection through three phases. The Chest CT images are initially preprocessed by applying some preprocessing strategies like image demising and contrast enhancement. Here, General Adversarial Network (GAN) is considered for performing the image denoising process, and adaptive histogram equalization is adapted for the enhancement of image contrast. Then, an active contour algorithm is applied to segment the tumor affected region of the chest CT image. Finally, the detection is done through the 3DCNN model, which categorizes whether a segmented region is affected by lung cancer or not. The proposed lung cancer detection model's effectiveness is estimated with several performance evaluation metrics. Moreover, some recent prior standard works are also considered to prove the effectiveness of the lung cancer detection model. The proposed 3DCNN model attained an accuracy of 98%, a precision of 97.2%, a recall of 98.5%, an error of 2%, a specificity of 97.7% and an NPV of 97.1%. The proposed lung cancer detection approach makes it possible to identify the disease at an early stage and take preventative measures. [ABSTRACT FROM AUTHOR]

Hyperspectral images are easily contaminated by multi-modal noise in the process of data acquisition. Generally, hyperspectral signatures and noise appear unexpectedly scattered in the spatial-spectral domain. Conversely, in the transform domain, the principle components of an image are gathered in the low-frequency bands, while the details and noise often occur in high-frequency parts. The existing denoising methods usually divide the whole frequency domain into two parts, that is, high frequency and low frequency, which can not make the most of the knowledge contained at different frequencies or scales, especially in the multiscale transform domain. Furthermore, aiming to restore a clean image from the noise-corrupted data, hyperspectral image denoising often amounts to an ill-posed problem, in which reshaping cubic data into matrix form results in the loss of structure information. To address the aforementioned issues, this paper incorporates the reweighted smoothing regularization in the transform domain into the noise-perturbed degradation model to reformulate a novel optimization problem. First, the multiscale curvelet is leveraged to transform the spatial matrix into frequency domain, where the coefficients are carefully designed to weight the components at different scales and orientations. Second, ${l_1}$ l 1 -norm is imposed on the abundance term to promote sparsity in the transform domain for performance enhancement. Finally, using proximal alternating optimization, an efficient algorithm is well developed to obtain the closed-form solutions to the proposed denoising problem. The experimental results on several synthetic and real datasets, illustrate that the proposed method improves PSNR by at least 1.5 dB and reduces ERGAS error by more than 35% than the state-of-the-art approaches under the multi-modal noise mixed environments, which verifies the validity of regularization terms in hyperspectral images denoising. [ABSTRACT FROM AUTHOR]

Universal image restoration (UIR) aims to accurately restore images with a variety of unknown degradation types and levels. Existing methods, including both learning-based and prior-based approaches, heavily rely on low-quality image features. However, it is challenging to extract degradation information from diverse low-quality images, which limits model performance. Furthermore, UIR necessitates the recovery of images with diverse and complex types of degradation. Inaccurate estimations further decrease restoration performance, resulting in suboptimal recovery outcomes. To enhance UIR performance, a viable approach is to introduce additional priors. The current UIR methods have problems such as poor enhancement effect and low universality. To address this issue, we propose an effective framework based on a diffusion model (DM) for universal image restoration, dubbed ETDiffIR. Inspired by the remarkable performance of text prompts in the field of image generation, we employ text prompts to improve the restoration of degraded images. This framework utilizes a text prompt corresponding to the low-quality image to assist the diffusion model in restoring the image. Specifically, a novel text–image fusion block is proposed by combining the CLIP text encoder and the DA-CLIP image controller, which integrates text prompt encoding and degradation type encoding into time step encoding. Moreover, to reduce the computational cost of the denoising UNet in the diffusion model, we develop an efficient restoration U-shaped network (ERUNet) to achieve favorable noise prediction performance via depthwise convolution and pointwise convolution. We evaluate the proposed method on image dehazing, deraining, and denoising tasks. The experimental results indicate the superiority of our proposed algorithm. [ABSTRACT FROM AUTHOR]

To address the problems of noise interference and image blurring in hyperspectral imaging (HSI), this paper proposes a denoising method for HSI based on deep learning and a total variation (TV) prior. The method minimizes the first-order moment distance between the deep prior of a Fast and Flexible Denoising Convolutional Neural Network (FFDNet) and the Enhanced 3D TV (E3DTV) prior, obtaining dual priors that complement and reinforce each other's advantages. Specifically, the original HSI is initially processed with a random binary sparse observation matrix to achieve a sparse representation. Subsequently, the plug-and-play (PnP) algorithm is employed within the framework of generalized alternating projection (GAP) to denoise the sparsely represented HSI. Experimental results demonstrate that, compared to existing methods, this method shows significant advantages in both quantitative and qualitative assessments, effectively enhancing the quality of HSIs. [ABSTRACT FROM AUTHOR]

This paper focuses on converting complex 3D maps created by LiDAR and SLAM technology into simple 2D maps to make them easier to understand. While 3D maps provide a lot of useful details for robots and computer programs, they can be difficult to read for humans who are used to flat maps. We developed a new system to clean up these 3D maps and convert them into intuitive and accurate 2D maps. The system uses three steps designed to correct different kinds of errors found in 3D LiDAR scan data: clustering-based denoising, height-based denoising, and Statistical Outlier Removal. In particular, height-based denoising is the method we propose in this paper, an algorithm that leaves only indoor structures such as walls. The paper proposes an algorithm that considers the entire range of the point cloud, rather than just the points near the ceiling, as is the case with existing methods, to make denoising more effective. This makes the final 2D map easy to understand and useful for building planning or emergency preparedness. Our main goal is to map the interior of buildings faster and more effectively, creating 2D drawings that reflect accurate and current information. We want to make it easier to use LiDAR and SLAM data in our daily work and increase productivity. [ABSTRACT FROM AUTHOR]

Compared to supervised denoising models based on deep learning, the unsupervised Deep Image Prior (DIP) denoising approach offers greater flexibility and practicality by operating solely with the given noisy image. However, the random initialization of network input and network parameters in the DIP leads to a slow convergence during iterative training, affecting the execution efficiency heavily. To address this issue, we propose the Masked-Pre-training-Based Fast DIP (MPFDIP) Denoising Model in this paper. We enhance the classical Restormer framework by improving its Transformer core module and incorporating sampling, residual learning, and refinement techniques. This results in a fast network called FRformer (Fast Restormer). The FRformer model undergoes offline supervised training using the masked processing technique for pre-training. For a specific noisy image, the pre-trained FRformer network, with its learned parameters, replaces the UNet network used in the original DIP model. The online iterative training of the replaced model follows the DIP unsupervised training approach, utilizing multi-target images and an adaptive loss function. This strategy further improves the denoising effectiveness of the pre-trained model. Extensive experiments demonstrate that the MPFDIP model outperforms existing mainstream deep-learning-based denoising models in reducing Gaussian noise, mixed Gaussian–Poisson noise, and low-dose CT noise. It also significantly enhances the execution efficiency compared to the original DIP model. This improvement is mainly attributed to the FRformer network's initialization parameters obtained through masked pre-training, which exhibit strong generalization capabilities for various types and intensities of noise and already provide some denoising effect. Using them as initialization parameters greatly improves the convergence speed of unsupervised iterative training in the DIP. Additionally, the techniques of multi-target images and the adaptive loss function further enhance the denoising process. [ABSTRACT FROM AUTHOR]

AbstractAn investigation is performed to study the damage mechanisms of dual phase steels using experimental analysis, local finite element simulation, and nonlocal peridynamics (PDs) method. The uniaxial tensile tests and preparation of metallography and SEM images have been used to peruse the pattern of deformation and voids in different phases. Understanding of the relationship between real microstructure and damage mechanisms can often not be obtained from microstructural investigation using experiments only. Here, a novel numerical approach is presented to investigate damage mechanisms in multiphase materials at micro scale level using both local and nonlocal numerical analyses and microstructural images from a scanning electron microscope (SEM). The nonlocal PDs method is capable to overcome many challenges associated with convergence problems in other numerical methods and to detail-oriented study of damage initiation and propagation autonomously. In this study, images of real microstructure are used as RVEs and imported to the FEM software and PD-developed code. Moreover, the effects of different constitutive models for the interface of ferrite and martensite phases on the bond breakage and damage patterns have been also investigated. [ABSTRACT FROM AUTHOR]

Conventional and electron microscopy visualize structures in the micrometer to nanometer range, and such visualizations contribute decisively to our understanding of biological processes. Due to different factors in recording processes, microscopy images are subject to noise. Especially at their respective resolution limits, a high degree of noise can negatively effect both image interpretation by experts and further automated processing. However, the deteriorating effects of strong noise can be alleviated to a large extend by image enhancement algorithms. Because of the inherent high noise, a requirement for such algorithms is their applicability directly to noisy images or, in the extreme case, to just a single noisy image without a priori noise level information (referred to as blind zero-shot setting). This work investigates blind zero-shot algorithms for microscopy image denoising. The strategies for denoising applied by the investigated approaches include: filtering methods, recent feed-forward neural networks which were amended to be trainable on noisy images, and recent probabilistic generative models. As datasets we consider transmission electron microscopy images including images of SARS-CoV-2 viruses and fluorescence microscopy images. A natural goal of denoising algorithms is to simultaneously reduce noise while preserving the original image features, e.g., the sharpness of structures. However, in practice, a tradeoff between both aspects often has to be found. Our performance evaluations, therefore, focus not only on noise removal but set noise removal in relation to a metric which is instructive about sharpness. For all considered approaches, we numerically investigate their performance, report their denoising/sharpness tradeoff on different images, and discuss future developments. We observe that, depending on the data, the different algorithms can provide significant advantages or disadvantages in terms of their noise removal vs. sharpness preservation capabilities, which may be very relevant for different virological applications, e.g., virological analysis or image segmentation. Author summary: For any image (e.g., recorded using digital photography or electron microscopy), there is a limiting resolution for which the image still looks sharp. Beyond this resolution (by further zooming in), an image becomes increasingly noisy. Resolution limits may be improved technically (bigger lenses, more light, longer exposure times etc.). Alternatively, or additionally, modern denoising algorithms can be applied after the image is recorded. Here we investigate state-of-the-art such algorithms with many of them being developed in field of Machine Learning. The algorithms work by first learning representations of images (e.g., edges, textures etc.), which they then use to remove the noise from an image. To be applicable at high resolutions, algorithms need to be able to learn from noisy images. If just one noisy image is available for learning without a priori noise level information, the task is also referred to as blind zero-shot denoising. Importantly, we argue that the implicit goal of denoising is noise removal without majorly diminishing image sharpness. All the here investigated blind zero-shot algorithms are consequently evaluated based on both: noise removal and sharpness preservation. We observed salient strengths and weaknesses of different approaches for different image types, which can significantly impact different tasks. [ABSTRACT FROM AUTHOR]

Low-dose computed tomography (LDCT) has gained increasing attention owing to its crucial role in reducing radiation exposure in patients. However, LDCT-reconstructed images often suffer from significant noise and artifacts, negatively impacting the radiologists' ability to accurately diagnose. To address this issue, many studies have focused on denoising LDCT images using deep learning (DL) methods. However, these DL-based denoising methods have been hindered by the highly variable feature distribution of LDCT data from different imaging sources, which adversely affects the performance of current denoising models. In this study, we propose a parallel processing model, the multi-encoder deep feature transformation network (MDFTN), which is designed to enhance the performance of LDCT imaging for multisource data. Unlike traditional network structures, which rely on continual learning to process multitask data, the approach can simultaneously handle LDCT images within a unified framework from various imaging sources. The proposed MDFTN consists of multiple encoders and decoders along with a deep feature transformation module (DFTM). During forward propagation in network training, each encoder extracts diverse features from its respective data source in parallel and the DFTM compresses these features into a shared feature space. Subsequently, each decoder performs an inverse operation for multisource loss estimation. Through collaborative training, the proposed MDFTN leverages the complementary advantages of multisource data distribution to enhance its adaptability and generalization. Numerous experiments were conducted on two public datasets and one local dataset, which demonstrated that the proposed network model can simultaneously process multisource data while effectively suppressing noise and preserving fine structures. The source code is available at https://github.com/123456789ey/MDFTN. [ABSTRACT FROM AUTHOR]

In vivo revealing how brain subregions are structurally connected during neonatal period via diffusion magnetic resonance imaging (dMRI) is critical for understanding brain development and pediatric mental disorders. However, even if preprocess was performed on dMRI images including denoising, eddy and motion correction, and unring, residual artifacts still affect the construction of brain structural networks. In this study, nonlocal estimation of multispectral magnitudes (NESMA) was extended to further smooth the preprocessed brain dMRI images of 46 healthy infants from the developing Human Connectome Project (dHCP). The proposed method smoothed dMRI images by exploiting similar multispectral diffusion-weighted signal pattern and the signal redundance among 3D patches. After structural connectivity networks were constructed from the smoothed dMRI images, network-level and nodal topological measures were estimated. While characteristic path length remained unchanged, significantly higher global efficiency, average clustering coefficient, and transitivity were observed in the infant structural networks built from NESMA-smoothed dMRI images. Additionally, more brain subregions with clustering coefficient > = 0.035 and local efficiency > = 0.05 were identified. In summary, higher efficiency was observed in the structural connectivity networks of healthy infants. Nonlocal estimation of multispectral diffusion-weighted volumes has nonnegligible effect on topological analysis of infant brain structural networks. The code for this algorithm is publicly available at https://github.com/freedom1979/NESMA-dMRI. [ABSTRACT FROM AUTHOR]

Deep learning has gained significant interest in image denoising, but there are notable distinctions in the types of deep learning methods used. Discriminative learning is suitable for handling Gaussian noise, while optimization models are effective in estimating real noise. However, there is limited research that summarizes the different deep learning techniques for image denoising. This paper conducts a comprehensive review of techniques and methods used for image denoising and identifying challenges associated with existing approaches. In this paper, a comparative study of deep techniques is offered in image denoising. The study conducted a comprehensive review of 68 papers on image denoising published between 2018 and 2023, providing a detailed analysis of the field's progress and methodologies over a period of 5 years. Through its literature review, the paper provides a comprehensive summary of image denoising in deep learning, including machine learning methods for image denoising, CNNs for image denoising, additive white noisy-image denoising, real noisy image denoising, blind denoising, hybrid noisy images, state- of-the-art methods for image denoising with deep learning, salt and pepper noise, non-linear filters for digital color images. The main objective of this paper is to provide a comprehensive overview of various approaches used for image denoising, each of which has been explored and developed based on individual research studies. The paper aims to discuss these approaches in a systematic and organized manner, comparing their strengths and weaknesses to provide insights for future research in the field. [ABSTRACT FROM AUTHOR]

Objective. Deep learning reconstruction (DLR) algorithms exhibit object-dependent resolution and noise performance. Thus, traditional geometric CT phantoms cannot fully capture the clinical imaging performance of DLR. This study uses a patient-derived 3D-printed PixelPrint lung phantom to evaluate a commercial DLR algorithm across a wide range of radiation dose levels. Method. The lung phantom used in this study is based on a patient chest CT scan containing ground glass opacities and was fabricated using PixelPrint 3D-printing technology. The phantom was placed inside two different size extension rings to mimic a small- and medium-sized patient and was scanned on a conventional CT scanner at exposures between 0.5 and 20 mGy. Each scan was reconstructed using filtered back projection (FBP), iterative reconstruction, and DLR at five levels of denoising. Image noise, contrast to noise ratio (CNR), root mean squared error, structural similarity index (SSIM), and multi-scale SSIM (MS SSIM) were calculated for each image. Results. DLR demonstrated superior performance compared to FBP and iterative reconstruction for all measured metrics in both phantom sizes, with better performance for more aggressive denoising levels. DLR was estimated to reduce dose by 25%–83% in the small phantom and by 50%–83% in the medium phantom without decreasing image quality for any of the metrics measured in this study. These dose reduction estimates are more conservative compared to the estimates obtained when only considering noise and CNR. Conclusion. DLR has the capability of producing diagnostic image quality at up to 83% lower radiation dose, which can improve the clinical utility and viability of lower dose CT scans. Furthermore, the PixelPrint phantom used in this study offers an improved testing environment with more realistic tissue structures compared to traditional CT phantoms, allowing for structure-based image quality evaluation beyond noise and contrast-based assessments. [ABSTRACT FROM AUTHOR]

This work presents a performance-oriented study of a heterogeneous application developed with Intel OneAPI to solve two well-known diffusion problems: heat diffusion and image denoising. We have explored CPU+iGPU and CPU+FPGA schemes, applying dynamic load balancing and conducting experiments on Intel DevCloud. The results demonstrate that the CPU+iGPU scheme outperforms the execution times achieved by the fastest device when the problem is sufficiently computationally demanding. We also found that the performance of the CPU+FPGA scheme is heavily affected by bandwidth limitations and specific strategies to manage memory efficiently are required. Moreover, it was demonstrated that dynamic workload balancing is crucial due to possible performance fluctuations in any of the implicated devices. In conclusion, Intel OneAPI provides a helpful tool for multi-platform development using a unique high-level language, DPC++. However, developing specific code for each platform is necessary to achieve optimal performance. [ABSTRACT FROM AUTHOR]

A crucial aspect of contemporary image processing systems is image denoising. The anisotropic diffusion function is a feature of the partial differential equation employed for the purpose of noise reduction and the preservation of image characteristics such as edges. A new tangent sigmoid diffusion coefficient and a new adaptive threshold parameter have been proposed in this work, which leads to faster convergence. In comparison to traditional anisotropic diffusion model techniques, the proposed technique performs admirably. As evidenced by the results, which demonstrate that the new anisotropic diffusion technique is not only capable of efficiently removing noise, but also of maintaining content in the denoised image. The performance of the proposed method is evaluated using various metrics, including peak signal-to-noise ratio, convergence rate, structural similarity index, time complexity, and space complexity. When comparing the proposed approach to previous methods, it is evident that the proposed method outperforms in various aspects. These include a higher convergence rate (− 0.1278), a greater peak signal-to-noise ratio (37.9827 dB), a higher structural similarity index (0.97432), a lower time complexity (5.72 s), and a smaller space complexity (15.6 KB). [ABSTRACT FROM AUTHOR]

Next‐generation light source facilities offer extreme spatial and temporal resolving power, enabling multiscale, ultra‐fast and dynamic characterizations. However, a trade‐off between acquisition efficiency and data quality needs to be made to fully unleash the resolving potential, for which purpose powerful denoising algorithms to improve the signal‐to‐noise ratio of the acquired X‐ray images are desirable. Yet, existing models based on machine learning mostly require massive and diverse labeled training data. Here we introduce a self‐supervised pre‐training algorithm with blind denoising capability by exploring the intrinsic physical symmetry of X‐ray patterns without requiring high signal‐to‐noise ratio reference data. The algorithm is more efficient and effective than algorithms without symmetry involved, including an supervised algorithm. It allows us to recover physical information from spatially and temporally resolved data acquired in X‐ray diffraction/scattering and pair distribution function experiments, where pattern symmetry is often well preserved. This study facilitates photon‐hungry experiments as well as in situ experiments with dynamic loading. [ABSTRACT FROM AUTHOR]

Image denoising is regarded as an ill-posed problem in computer vision tasks that removes additive noise from imaging sensors. Recently, several convolution neural network-based image-denoising methods have achieved remarkable advances. However, it is difficult for a simple denoising network to recover aesthetically pleasing images owing to the complexity of image content. Therefore, this study proposes a multi-branch network to improve the performance of the denoising method. First, the proposed network is designed based on a conventional autoencoder to learn multi-level contextual features from input images. Subsequently, we integrate two modules into the network, including the Pyramid Context Module (PCM) and the Residual Bottleneck Attention Module (RBAM), to extract salient information for the training process. More specifically, PCM is applied at the beginning of the network to enlarge the receptive field and successfully address the loss of global information using dilated convolution. Meanwhile, RBAM is inserted into the middle of the encoder and decoder to eliminate degraded features and reduce undesired artifacts. Finally, extensive experimental results prove the superiority of the proposed method over state-of-the-art deep-learning methods in terms of objective and subjective performances. [ABSTRACT FROM AUTHOR]

In this paper, we address the image denoising problem in presence of speckle degradation typically arising in ultra-sound images. Variational methods and Convolutional Neural Networks (CNNs) are considered well-established methods for specific noise types, such as Gaussian and Poisson noise. Nonetheless, the advances achieved by these two classes of strategies are limited when tackling the de-speckle problem. In fact, variational methods for speckle removal typically amounts to solve a non-convex functional with the related issues from the convergence viewpoint; on the other hand, the lack of large datasets of noise-free ultra-sound images has not allowed the extension of the state-of-the-art CNN denoiser methods to the case of speckle degradation. Here, we aim at combining the classical variational methods with the predictive properties of CNNs by considering a weighted total variation regularized model; the local weights are obtained as the output of a statistically inspired neural network that is trained on a small and composite dataset of natural and synthetic images. The resulting non-convex variational model, which is minimized by means of the Alternating Direction Method of Multipliers (ADMM) is proven to converge to a stationary point. Numerical tests show the effectiveness of our approach for the denoising of natural and satellite images. • An hybrid strategy for the despeckling problem combining convolutional neural networks and variational methods is proposed. • The proposed neural architecture is statistically-inspired and not requires the training on a large data-set of images. • A proof of convergence to a stationary point for the alternating direction method is given. [ABSTRACT FROM AUTHOR]

Background: Preclinical low‐count positron emission tomography (LC‐PET) imaging offers numerous advantages such as facilitating imaging logistics, enabling longitudinal studies of long‐ and short‐lived isotopes as well as increasing scanner throughput. However, LC‐PET is characterized by reduced photon‐count levels resulting in low signal‐to‐noise ratio (SNR), segmentation difficulties, and quantification uncertainties. Purpose: We developed and evaluated a novel deep‐learning (DL) architecture—Attention based Residual‐Dilated Net (ARD‐Net)—to generate standard‐count PET (SC‐PET) images from LC‐PET images. The performance of the ARD‐Net framework was evaluated for numerous low count realizations using fidelity‐based qualitative metrics, task‐based segmentation, and quantitative metrics. Method: Patient Derived tumor Xenograft (PDX) with tumors implanted in the mammary fat‐pad were subjected to preclinical [18F]‐Fluorodeoxyglucose (FDG)‐PET/CT imaging. SC‐PET images were derived from a 10 min static FDG‐PET acquisition, 50 min post administration of FDG, and were resampled to generate four distinct LC‐PET realizations corresponding to 10%, 5%, 1.6%, and 0.8% of SC‐PET count‐level. ARD‐Net was trained and optimized using 48 preclinical FDG‐PET datasets, while 16 datasets were utilized to assess performance. Further, the performance of ARD‐Net was benchmarked against two leading DL‐based methods (Residual UNet, RU‐Net; and Dilated Network, D‐Net) and non‐DL methods (Non‐Local Means, NLM; and Block Matching 3D Filtering, BM3D). The performance of the framework was evaluated using traditional fidelity‐based image quality metrics such as Structural Similarity Index Metric (SSIM) and Normalized Root Mean Square Error (NRMSE), as well as human observer‐based tumor segmentation performance (Dice Score and volume bias) and quantitative analysis of Standardized Uptake Value (SUV) measurements. Additionally, radiomics‐derived features were utilized as a measure of quality assurance (QA) in comparison to true SC‐PET. Finally, a performance ensemble score (EPS) was developed by integrating fidelity‐based and task‐based metrics. Concordance Correlation Coefficient (CCC) was utilized to determine concordance between measures. The non‐parametric Friedman Test with Bonferroni correction was used to compare the performance of ARD‐Net against benchmarked methods with significance at adjusted p‐value ≤0.01. Results: ARD‐Net‐generated SC‐PET images exhibited significantly better (p ≤ 0.01 post Bonferroni correction) overall image fidelity scores in terms of SSIM and NRMSE at majority of photon‐count levels compared to benchmarked DL and non‐DL methods. In terms of task‐based quantitative accuracy evaluated by SUVMean and SUVPeak, ARD‐Net exhibited less than 5% median absolute bias for SUVMean compared to true SC‐PET and lower degree of variability compared to benchmarked DL and non‐DL based methods in generating SC‐PET. Additionally, ARD‐Net‐generated SC‐PET images displayed higher degree of concordance to SC‐PET images in terms of radiomics features compared to non‐DL and other DL approaches. Finally, the ensemble score suggested that ARD‐Net exhibited significantly superior performance compared to benchmarked algorithms (p ≤ 0.01 post Bonferroni correction). Conclusion: ARD‐Net provides a robust framework to generate SC‐PET from LC‐PET images. ARD‐Net generated SC‐PET images exhibited superior performance compared other DL and non‐DL approaches in terms of image‐fidelity based metrics, task‐based segmentation metrics, and minimal bias in terms of task‐based quantification performance for preclinical PET imaging. [ABSTRACT FROM AUTHOR]

Background: On enhancing the image quality of low‐dose computed tomography (LDCT), various denoising methods have achieved meaningful improvements. However, they commonly produce over‐smoothed results; the denoised images tend to be more blurred than the normal‐dose targets (NDCTs). Furthermore, many recent denoising methods employ deep learning(DL)‐based models, which require a vast amount of CT images (or image pairs). Purpose: Our goal is to address the problem of over‐smoothed results and design an algorithm that works regardless of the need for a large amount of training dataset to achieve plausible denoising results. Over‐smoothed images negatively affect the diagnosis and treatment since radiologists had developed clinical experiences with NDCT. Besides, a large‐scale training dataset is often not available in clinical situations. To overcome these limitations, we propose locally‐adaptive noise‐level matching (LANCH), emphasizing the output should retain the same noise‐level and characteristics to that of the NDCT without additional training. Methods: We represent the NDCT image as the pixel‐wisely weighted sum of an over‐smoothed output from off‐the‐shelf denoiser (OSD) and the difference between the LDCT image and the OSD output. Herein, LANCH determines a 2D ratio map (i.e., pixel‐wise weight matrix) by locally matching the noise‐level of output and NDCT, where the LDCT‐to‐NDCT device flux (mAs) ratio reveals the NDCT noise‐level. Thereby, LANCH can preserve important details in LDCT, and enhance the sharpness of the noise‐free regions. Note that LANCH can enhance any LDCT denoisers without additional training data (i.e., zero‐shot). Results: The proposed method is applicable to any OSD denoisers, reporting significant texture plausibility development over the baseline denoisers in quantitative and qualitative manners. It is surprising that the denoising accuracy achieved by our method with zero‐shot denoiser was comparable or superior to that of the best training‐based denoisers; our result showed 1% and 33% gains in terms of SSIM and DISTS, respectively. Reader study with experienced radiologists shows significant image quality improvements, a gain of + 1.18 on a five‐point mean opinion score scale. Conclusions: In this paper, we propose a technique to enhance any low‐dose CT denoiser by leveraging the fundamental physical relationship between the x‐ray flux and noise variance. Our method is capable of operating in a zero‐shot condition, which means that only a single low‐dose CT image is required for the enhancement process. We demonstrate that our approach is comparable or even superior to supervised DL‐based denoisers that are trained using numerous CT images. Extensive experiments illustrate that our method consistently improves the performance of all tested LDCT denoisers. [ABSTRACT FROM AUTHOR]

Objectives: The conventional group sparse representation (GSR) model encounters challenges in precisely estimating the sparsity of individual image groups, because the influence of noise results in distortion in the restoration of original image group. Methods: This paper introduces a strategy denominated as the group sparse residual constraint. This strategy leverages non-local self-similarity priors to derive precise estimate of sparse coefficients for each original image group. Subsequently, the sparse coefficients of corresponding degraded image groups are constrained to approximate this estimate. Furthermore, the intricate nature of noise in area-array satellite images (e. g., Luojia3-01 satellite images), surpasses the simplicity of additive white Gaussian noise (AWGN). Consequently, conventional GSR model tailored for AWGN exhibits a reduction in efficacy when deployed in real image denoising scenarios. To address this, this paper introduces two distinct weight matrices into the data fidelity term and regularization, respectively, with the intent of characterizing the statistical features inherent in both images and noise. This paper adopts the alternating direction method of multipliers for optimizing the proposed image denoising task. Results: The effectiveness of the proposed model is validated by the experiments using simulated data and real images obtained from the Luojia3-01 satellite. In the experiments of simulated data, the proposed model outperforms other methods in removing AWGN and spatially variant noise. In the experiments of real images, the entropy of denoised images by the proposed model shows improvements of 2.03%, 1.18%, 1.26%, 1.24%, and 2.10%, compared to the block-matching and 3-D filtering method, multi-channel weighted nuclear norm minimization method, nonlocally centralized sparse representation method, low-rankness guided group sparse representation method, and trilateral weighted sparse coding method, respectively. Conclusions: The proposed method exhibits good performance in processing area-array satellite images with low signal-to-noise ratios and complex noise structures, significantly enhancing the quality of the image. By effectively reducing noise while preserving crucial details and edges, it provides a reliable solution for enhancing the usability and interpretability of area-array satellite images, particularly in challenging conditions. [ABSTRACT FROM AUTHOR]

A cavity ring-down spectroscopy (CRDS) system based on the time-division multiplexing (TDM) and an improved Whittaker-Henderson (W-H) filtering algorithm was proposed to synchronous detect of methane (CH4) and carbon dioxide (CO2) gas absorption spectroscopy. The mechanical optical switch was used to quickly switch distributed feedback (DFB) lasers with different wavelengths in the optical path.And the adaptive W-H filtering algorithm combined with convolutional neural network (CNN) could quickly and accurately obtain the best weight parameters to achieve effective denoising of absorption spectrum.In addition, the gradient concentration of CH4 and CO2weredetected and the limit of detection (LOD) of CRDS system was studied.The results of experiment demonstrated that the system could detect CO2 and CH4 in real time and had good stability. Allan deviation analysis shows that the LOD of CH4 and CO2 are 8 ppb and 0.85 ppm under the average number of 1. The LOD of CH4 and CO2 can be optimized to 2 ppb and 0.16 ppm under the optimal average number. In short, the proposed system had high sensitivity, good stability and could measure various gases synchronously, which had important application potential in the field of gas monitoring. [ABSTRACT FROM AUTHOR]