A GPU-accelerated fuzzy method for real-time CT volume filtering

During acquisition and reconstruction, medical images may become noisy and lose diagnostic quality. In the case of CT scans, obtaining less noisy images results in a higher radiation dose being administered to the patient. Filtering techniques can be utilized to reduce radiation without losing diagnosis capabilities. The objective in this work is to obtain an implementation of a filter capable of processing medical images in real-time. To achieve this we have developed several filter methods based on fuzzy logic, and their GPU implementations, to reduce mixed Gaussian-impulsive noise. These filters have been developed to work in attenuation coefficients so as to not lose any information from the CT scans. The testing volumes come from the Mayo clinic database and consist of CT volumes at full and at simulated low dose. The GPU parallelizations reach speedups of over 2700 and take less than 0.1 seconds to filter more than 300 slices. In terms of quality the filter is competitive with other state of the art algorithmic and AI filters. The proposed method obtains good performance in terms of quality and the parallelization results in real-time filtering.

Neural Degradation Representation Learning for All-in-One Image Restoration

Existing methods have demonstrated effective performance on a single degradation type. In practical applications, however, the degradation is often unknown, and the mismatch between the model and the degradation will result in a severe performance drop. In this paper, we propose an all-in-one image restoration network that tackles multiple degradations. Due to the heterogeneous nature of different types of degradations, it is difficult to process multiple degradations in a single network. To this end, we propose to learn a neural degradation representation (NDR) that captures the underlying characteristics of various degradations. The learned NDR adaptively decomposes different types of degradations, similar to a neural dictionary that represents basic degradation components. Subsequently, we develop a degradation query module and a degradation injection module to effectively approximate and utilize the specific degradation based on NDR, enabling the all-in-one restoration ability for multiple degradations. Moreover, we propose a bidirectional optimization strategy to effectively drive NDR to learn the degradation representation by optimizing the degradation and restoration processes alternately. Comprehensive experiments on representative types of degradations (including noise, haze, rain, and downsampling) demonstrate the effectiveness and generalizability of our method. Code is available at https://github.com/mdyao/NDR-Restore.

Large-volume fully automated cell reconstruction generates a cell atlas of plant tissues

The geometric shape and arrangement of individual cells play a role in shaping organ functions. However, analyzing multicellular features and exploring their connectomes in centimeter-scale plant organs remain challenging. Here, we established a set of frameworks named Large-Volume Fully Automated Cell Reconstruction (LVACR), enabling the exploration of three-dimensional (3D) cytological features and cellular connectivity in plant tissues. Through benchmark testing, our framework demonstrated superior efficiency in cell segmentation and aggregation, successfully addressing the inherent challenges posed by light sheet fluorescence microscopy (LSFM) imaging. Using LVACR, we successfully established a cell atlas of different plant tissues. Cellular morphology analysis revealed differences of cell clusters and shapes in between different poplar (P. simonii Carr. and P. canadensis Moench.) seeds, whereas topological analysis revealed that they maintained conserved cellular connectivity. Furthermore, LVACR spatiotemporally demonstrated an initial burst of cell proliferation, accompanied by morphological transformations at an early stage in developing the shoot apical meristem. During subsequent development, cell differentiation produced anisotropic features, thereby resulting in various cell shapes. Overall, our findings provided valuable insights into the precise spatial arrangement and cellular behavior of multicellular organisms, thus enhancing our understanding of the complex processes underlying plant growth and differentiation.

Real-world video superresolution enhancement method based on the adaptive down-sampling model

With the 5G and the popularity of high-definition and ultrahigh-definition equipment, people have increasingly higher requirements for the resolution of images or videos. However, the transmission pressure on servers is also gradually increasing. Therefore, superresolution technology has attracted much attention in recent years. Simultaneously, with the further development of deep learning techniques, superresolution research is shifting from the calculation of traditional algorithms to the deep learning method, which exhibits a greatly superior final display. First, the traditional block-matching-3D (BM3D) algorithm is formed as the postprocessing module, which can avoid the uneven edge of GAN network recovery, make the picture appear more authentic, and improve the viewer's subjective feelings. Next, the adaptive-downsampling model (ADM) is utilized to train models for specific camera styles. The high-resolution (HR) data sequence is subsequently downsampled to a low-resolution (LR) data sequence, enabling the superresolution algorithm to utilize this training set. This method can obtain better results and improve overall performance by 0.1~0.3 dB.

Universal Image Restoration with Text Prompt Diffusion

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.

Multichannel Image Completion With Mixture Noise: Adaptive Sparse Low-Rank Tensor Subspace Meets Nonlocal Self-Similarity

Multichannel image completion with mixture noise is a common but complex problem in the fields of machine learning, image processing, and computer vision. Most existing algorithms devote to explore global low-rank information and fail to optimize local and joint-mode structures, which may lead to oversmooth restoration results or lower quality restoration details. In this study, we propose a novel model to deal with multichannel image completion with mixture noise based on adaptive sparse low-rank tensor subspace and nonlocal self-similarity (ASLTS-NS). In the proposed model, a nonlocal similar patch matching framework cooperating with Tucker decomposition is used to explore information of global and joint modes and optimize the local structure for improving restoration quality. In order to enhance the robustness of low-rank decomposition to data missing and mixture noise, we present an adaptive sparse low-rank regularization to construct robust tensor subspace for self-weighing importance of different modes and capturing a stable inherent structure. In addition, joint tensor Frobenius and l1 regularizations are exploited to control two different types of noise. Based on alternating directions method of multipliers (ADMM), a convergent learning algorithm is designed to solve this model. Experimental results on three different types of multichannel image sets demonstrate the advantages of ASLTS-NS under five complex scenarios.

Flexible and Generalized Real Photograph Denoising Exploiting Dual Meta Attention

Supervised deep learning techniques have been widely explored in real photograph denoising and achieved noticeable performances. However, being subject to specific training data, most current image denoising algorithms can easily be restricted to certain noisy types and exhibit poor generalizability across testing sets. To address this issue, we propose a novel flexible and well-generalized approach, coined as dual meta attention network (DMANet). The DMANet is mainly composed of a cascade of the self-meta attention blocks (SMABs) and collaborative-meta attention blocks (CMABs). These two blocks have two forms of advantages. First, they simultaneously take both spatial and channel attention into account, allowing our model to better exploit more informative feature interdependencies. Second, the attention blocks are embedded with the meta-subnetwork, which is based on metalearning and supports dynamic weight generation. Such a scheme can provide a beneficial means for self and collaborative updating of the attention maps on-the-fly. Instead of directly stacking the SMABs and CMABs to form a deep network architecture, we further devise a three-stage learning framework, where different blocks are utilized for each feature extraction stage according to the individual characteristics of SMAB and CMAB. On five real datasets, we demonstrate the superiority of our approach against the state of the art. Unlike most existing image denoising algorithms, our DMANet not only possesses a good generalization capability but can also be flexibly used to cope with the unknown and complex real noises, making it highly competitive for practical applications.

Learn From Unpaired Data for Image Restoration: A Variational Bayes Approach

Collecting paired training data is difficult in practice, but the unpaired samples broadly exist. Current approaches aim at generating synthesized training data from unpaired samples by exploring the relationship between the corrupted and clean data. This work proposes LUD-VAE, a deep generative method to learn the joint probability density function from data sampled from marginal distributions. Our approach is based on a carefully designed probabilistic graphical model in which the clean and corrupted data domains are conditionally independent. Using variational inference, we maximize the evidence lower bound (ELBO) to estimate the joint probability density function. Furthermore, we show that the ELBO is computable without paired samples under the inference invariant assumption. This property provides the mathematical rationale of our approach in the unpaired setting. Finally, we apply our method to real-world image denoising, super-resolution, and low-light image enhancement tasks and train the models using the synthetic data generated by the LUD-VAE. Experimental results validate the advantages of our method over other approaches.

Restored texture segmentation using Markov random fields

Texture segmentation plays a crucial role in the domain of image analysis and its recognition. Noise is inextricably linked to images, just like it is with every signal received by sensing, which has an impact on how well the segmentation process performs in general. Recent literature reveals that the research community has started recognizing the domain of noisy texture segmentation for its work towards solutions for the automated quality inspection of objects, decision support for biomedical images, facial expressions identification, retrieving image data from a huge dataset and many others. Motivated by the latest work on noisy textures, during our work being presented here, Brodatz and Prague texture images are contaminated with Gaussian and salt-n-pepper noise. A three-phase approach is developed for the segmentation of textures contaminated by noise. In the first phase, these contaminated images are restored using techniques with excellent performance as per the recent literature. In the remaining two phases, segmentation of the restored textures is carried out by a novel technique developed using Markov Random Fields (MRF) and objective customization of the Median Filter based on segmentation performance metrics. When the proposed approach is evaluated on Brodatz textures, an improvement of up to 16% segmentation accuracy for salt-n-pepper noise with 70% noise density and 15.1% accuracy for Gaussian noise (with a variance of 50) has been made in comparison with the benchmark approaches. On Prague textures, accuracy is improved by 4.08% for Gaussian noise (with variance 10) and by 2.47% for salt-n-pepper noise with 20% noise density. The approach in the present study can be applied to a diversified class of image analysis applications spanning a wide spectrum such as satellite images, medical images, industrial inspection, geo-informatics, etc.

Path-Restore: Learning Network Path Selection for Image Restoration

Very deep Convolutional Neural Networks (CNNs) have greatly improved the performance on various image restoration tasks. However, this comes at a price of increasing computational burden, hence limiting their practical usages. We observe that some corrupted image regions are inherently easier to restore than others since the distortion and content vary within an image. To leverage this, we propose Path-Restore, a multi-path CNN with a pathfinder that can dynamically select an appropriate route for each image region. We train the pathfinder using reinforcement learning with a difficulty-regulated reward. This reward is related to the performance, complexity and "the difficulty of restoring a region". A policy mask is further investigated to jointly process all the image regions. We conduct experiments on denoising and mixed restoration tasks. The results show that our method achieves comparable or superior performance to existing approaches with less computational cost. In particular, Path-Restore is effective for real-world denoising, where the noise distribution varies across different regions on a single image. Compared to the state-of-the-art RIDNet [1], our method achieves comparable performance and runs 2.7x faster on the realistic Darmstadt Noise Dataset [2]. Models and codes are available on the project page: https://www.mmlab-ntu.com/project/pathrestore/.

PID Controller-Guided Attention Neural Network Learning for Fast and Effective Real Photographs Denoising

Real photograph denoising is extremely challenging in low-level computer vision since the noise is sophisticated and cannot be fully modeled by explicit distributions. Although deep-learning techniques have been actively explored for this issue and achieved convincing results, most of the networks may cause vanishing or exploding gradients, and usually entail more time and memory to obtain a remarkable performance. This article overcomes these challenges and presents a novel network, namely, PID controller guide attention neural network (PAN-Net), taking advantage of both the proportional-integral-derivative (PID) controller and attention neural network for real photograph denoising. First, a PID-attention network (PID-AN) is built to learn and exploit discriminative image features. Meanwhile, we devise a dynamic learning scheme by linking the neural network and control action, which significantly improves the robustness and adaptability of PID-AN. Second, we explore both the residual structure and share-source skip connections to stack the PID-ANs. Such a framework provides a flexible way to feature residual learning, enabling us to facilitate the network training and boost the denoising performance. Extensive experiments show that our PAN-Net achieves superior denoising results against the state-of-the-art in terms of image quality and efficiency.

Manifold Modeling in Embedded Space: An Interpretable Alternative to Deep Image Prior

Deep image prior (DIP), which uses a deep convolutional network (ConvNet) structure as an image prior, has attracted wide attention in computer vision and machine learning. DIP empirically shows the effectiveness of the ConvNet structures for various image restoration applications. However, why the DIP works so well is still unknown. In addition, the reason why the convolution operation is useful in image reconstruction, or image enhancement is not very clear. This study tackles this ambiguity of ConvNet/DIP by proposing an interpretable approach that divides the convolution into "delay embedding" and "transformation" (i.e., encoder-decoder). Our approach is a simple, but essential, image/tensor modeling method that is closely related to self-similarity. The proposed method is called manifold modeling in embedded space (MMES) since it is implemented using a denoising autoencoder in combination with a multiway delay-embedding transform. In spite of its simplicity, MMES can obtain quite similar results to DIP on image/tensor completion, super-resolution, deconvolution, and denoising. In addition, MMES is proven to be competitive with DIP, as shown in our experiments. These results can also facilitate interpretation/characterization of DIP from the perspective of a "low-dimensional patch-manifold prior."

Meta PID Attention Network for Flexible and Efficient Real-World Noisy Image Denoising

Recent deep convolutional neural networks for real-world noisy image denoising have shown a huge boost in performance by training a well-engineered network over external image pairs. However, most of these methods are generally trained with supervision. Once the testing data is no longer compatible with the training conditions, they can exhibit poor generalization and easily result in severe overfitting or degrading performances. To tackle this barrier, we propose a novel denoising algorithm, dubbed as Meta PID Attention Network (MPA-Net). Our MPA-Net is built based upon stacking Meta PID Attention Modules (MPAMs). In each MPAM, we utilize a second-order attention module (SAM) to exploit the channel-wise feature correlations with second-order statistics, which are then adaptively updated via a proportional-integral-derivative (PID) guided meta-learning framework. This learning framework exerts the unique property of the PID controller and meta-learning scheme to dynamically generate filter weights for beneficial update of the extracted features within a feedback control system. Moreover, the dynamic nature of the framework enables the generated weights to be flexibly tweaked according to the input at test time. Thus, MPAM not only achieves discriminative feature learning, but also facilitates a robust generalization ability on distinct noises for real images. Extensive experiments on ten datasets are conducted to inspect the effectiveness of the proposed MPA-Net quantitatively and qualitatively, which demonstrates both its superior denoising performance and promising generalization ability that goes beyond those of the state-of-the-art denoising methods.

Blind and Compact Denoising Network Based on Noise Order Learning

A lightweight blind image denoiser, called blind compact denoising network (BCDNet), is proposed in this paper to achieve excellent trade-offs between performance and network complexity. With only 330K parameters, the proposed BCDNet is composed of the compact denoising network (CDNet) and the guidance network (GNet). From a noisy image, GNet extracts a guidance feature, which encodes the severity of the noise. Then, using the guidance feature, CDNet filters the image adaptively according to the severity to remove the noise effectively. Moreover, by reducing the number of parameters without compromising the performance, CDNet achieves denoising not only effectively but also efficiently. Experimental results show that the proposed BCDNet yields state-of-the-art or competitive denoising performances on various datasets while requiring significantly fewer parameters.

Exploiting Non-Local Priors via Self-Convolution for Highly-Efficient Image Restoration

Constructing effective priors is critical to solving ill-posed inverse problems in image processing and computational imaging. Recent works focused on exploiting non-local similarity by grouping similar patches for image modeling, and demonstrated state-of-the-art results in many image restoration applications. However, compared to classic methods based on filtering or sparsity, non-local algorithms are more time-consuming, mainly due to the highly inefficient block matching step, i.e., distance between every pair of overlapping patches needs to be computed. In this work, we propose a novel Self-Convolution operator to exploit image non-local properties in a unified framework. We prove that the proposed Self-Convolution based formulation can generalize the commonly-used non-local modeling methods, as well as produce results equivalent to standard methods, but with much cheaper computation. Furthermore, by applying Self-Convolution, we propose an effective multi-modality image restoration scheme, which is much more efficient than conventional block matching for non-local modeling. Experimental results demonstrate that (1) Self-Convolution with fast Fourier transform implementation can significantly speed up most of the popular non-local image restoration algorithms, with two-fold to nine-fold faster block matching, and (2) the proposed online multi-modality image restoration scheme achieves superior denoising results than competing methods in both efficiency and effectiveness on RGB-NIR images. The code for this work is publicly available at https://github.com/GuoLanqing/Self-Convolution.

Tensor Completion via Complementary Global, Local, and Nonlocal Priors

Completing missing entries in multidimensional visual data is a typical ill-posed problem that requires appropriate exploitation of prior information of the underlying data. Commonly used priors can be roughly categorized into three classes: global tensor low-rankness, local properties, and nonlocal self-similarity (NSS); most existing works utilize one or two of them to implement completion. Naturally, there arises an interesting question: can one concurrently make use of multiple priors in a unified way, such that they can collaborate with each other to achieve better performance? This work gives a positive answer by formulating a novel tensor completion framework which can simultaneously take advantage of the global-local-nonlocal priors. In the proposed framework, the tensor train (TT) rank is adopted to characterize the global correlation; meanwhile, two Plug-and-Play (PnP) denoisers, including a convolutional neural network (CNN) denoiser and the color block-matching and 3 D filtering (CBM3D) denoiser, are incorporated to preserve local details and exploit NSS, respectively. Then, we design a proximal alternating minimization algorithm to efficiently solve this model under the PnP framework. Under mild conditions, we establish the convergence guarantee of the proposed algorithm. Extensive experiments show that these priors organically benefit from each other to achieve state-of-the-art performance both quantitatively and qualitatively.

Thermal Image Restoration Based on LWIR Sensor Statistics

An imaging system has natural statistics that reflect its intrinsic characteristics. For example, the gradient histogram of a visible light image generally obeys a heavy-tailed distribution, and its restoration considers natural statistics. Thermal imaging cameras detect infrared radiation, and their signal processors are specialized according to the optical and sensor systems. Thermal images, also known as long wavelength infrared (LWIR) images, suffer from distinct degradations of LWIR sensors and residual nonuniformity (RNU). However, despite the existence of various studies on the statistics of thermal images, thermal image processing has seldom attempted to incorporate natural statistics. In this study, natural statistics of thermal imaging sensors are derived, and an optimization method for restoring thermal images is proposed. To verify our hypothesis about the thermal images, high-frequency components of thermal images from various datasets are analyzed with various measures (correlation coefficient, histogram intersection, chi-squared test, Bhattacharyya distance, and Kullback-Leibler divergence), and generalized properties are derived. Furthermore, cost functions accommodating the validated natural statistics are designed and minimized by a pixel-wise optimization method. The proposed algorithm has a specialized structure for thermal images and outperforms the conventional methods. Several image quality assessments are employed for quantitatively demonstrating the performance of the proposed method. Experiments with synthesized images and real-world images are conducted, and the results are quantified by reference image assessments (peak signal-to-noise ratio and structural similarity index measure) and no-reference image assessments (Roughness (Ro) and Effective Roughness (ERo) indices).

Residual Dense Network for Image Restoration

Recently, deep convolutional neural network (CNN) has achieved great success for image restoration (IR) and provided hierarchical features at the same time. However, most deep CNN based IR models do not make full use of the hierarchical features from the original low-quality images; thereby, resulting in relatively-low performance. In this work, we propose a novel and efficient residual dense network (RDN) to address this problem in IR, by making a better tradeoff between efficiency and effectiveness in exploiting the hierarchical features from all the convolutional layers. Specifically, we propose residual dense block (RDB) to extract abundant local features via densely connected convolutional layers. RDB further allows direct connections from the state of preceding RDB to all the layers of current RDB, leading to a contiguous memory mechanism. To adaptively learn more effective features from preceding and current local features and stabilize the training of wider network, we proposed local feature fusion in RDB. After fully obtaining dense local features, we use global feature fusion to jointly and adaptively learn global hierarchical features in a holistic way. We demonstrate the effectiveness of RDN with several representative IR applications, single image super-resolution, Gaussian image denoising, image compression artifact reduction, and image deblurring. Experiments on benchmark and real-world datasets show that our RDN achieves favorable performance against state-of-the-art methods for each IR task quantitatively and visually.

A Comparison of Optimized Sentinel-2 Super-Resolution Methods Using Wald’s Protocol and Bayesian Optimization

In the context of earth observation and remote sensing, super-resolution aims to enhance the resolution of a captured image by upscaling and enhancing its details. In recent years, numerous methods for super-resolution of Sentinel-2 (S2) multispectral images have been suggested. Most of those methods depend on various tuning parameters that affect how effective they are. This paper’s aim is twofold. Firstly, we propose to use Bayesian optimization at a reduced scale to select tuning parameters. Secondly, we choose tuning parameters for eight S2 super-resolution methods and compare them using real and synthetic data. While all the methods give good quantitative results, Area-To-Point Regression Kriging (ATPRK), Sentinel-2 Sharpening (S2Sharp), and Sentinel-2 Symmetric Skip Connection convolutional neural network (S2 SSC) perform markedly better on several datasets than the other methods tested in this paper.

Real-World Image Denoising with Deep Boosting

We propose a Deep Boosting Framework (DBF) for real-world image denoising by integrating the deep learning technique into the boosting algorithm. The DBF replaces conventional handcrafted boosting units by elaborate convolutional neural networks, which brings notable advantages in terms of both performance and speed. We design a lightweight Dense Dilated Fusion Network (DDFN) as an embodiment of the boosting unit, which addresses the vanishing of gradients during training due to the cascading of networks while promoting the efficiency of limited parameters. The capabilities of the proposed method are first validated on several representative simulation tasks including non-blind and blind Gaussian denoising and JPEG image deblocking. We then focus on a practical scenario to tackle with the complex and challenging real-world noise. To facilitate leaning-based methods including ours, we build a new Real-world Image Denoising (RID) dataset, which contains 200 pairs of high-resolution images with diverse scene content under various shooting conditions. Moreover, we conduct comprehensive analysis on the domain shift issue for real-world denoising and propose an effective one-shot domain transfer scheme to address this issue. Comprehensive experiments on widely used benchmarks demonstrate that the proposed method significantly surpasses existing methods on the task of real-world image denoising. Code and dataset are available at https://github.com/ngchc/deepBoosting.

Weighted Low-Rank Tensor Recovery for Hyperspectral Image Restoration

Hyperspectral imaging, providing abundant spatial and spectral information simultaneously, has attracted a lot of interest in recent years. Unfortunately, due to the hardware limitations, the hyperspectral image (HSI) is vulnerable to various degradations, such as noises (random noise), blurs (Gaussian and uniform blur), and downsampled (both spectral and spatial downsample), each corresponding to the HSI denoising, deblurring, and super-resolution tasks, respectively. Previous HSI restoration methods are designed for one specific task only. Besides, most of them start from the 1-D vector or 2-D matrix models and cannot fully exploit the structurally spectral-spatial correlation in 3-D HSI. To overcome these limitations, in this article, we propose a unified low-rank tensor recovery model for comprehensive HSI restoration tasks, in which nonlocal similarity within spectral-spatial cubic and spectral correlation are simultaneously captured by third-order tensors. Furthermore, to improve the capability and flexibility, we formulate it as a weighted low-rank tensor recovery (WLRTR) model by treating the singular values differently. We demonstrate the reweighed strategy, which has been extensively studied in the matrix, also greatly benefits the tensor modeling. We also consider the stripe noise in HSI as the sparse error by extending WLRTR to robust principal component analysis (WLRTR-RPCA). Extensive experiments demonstrate the proposed WLRTR models consistently outperform state-of-the-art methods in typical HSI low-level vision tasks, including denoising, destriping, deblurring, and super-resolution.

Noisy-As-Clean: Learning Self-supervised Denoising from Corrupted Image

Supervised deep networks have achieved promising performance on image denoising, by learning image priors and noise statistics on plenty pairs of noisy and clean images. Unsupervised denoising networks are trained with only noisy images. However, for an unseen corrupted image, both supervised and unsupervised networks ignore either its particular image prior, the noise statistics, or both. That is, the networks learned from external images inherently suffer from a domain gap problem: the image priors and noise statistics are very different between the training and test images. This problem becomes more clear when dealing with the signal dependent realistic noise. To circumvent this problem, in this work, we propose a novel "Noisy-As-Clean" (NAC) strategy of training self-supervised denoising networks. Specifically, the corrupted test image is directly taken as the "clean" target, while the inputs are synthetic images consisted of this corrupted image and a second yet similar corruption. A simple but useful observation on our NAC is: as long as the noise is weak, it is feasible to learn a self-supervised network only with the corrupted image, approximating the optimal parameters of a supervised network learned with pairs of noisy and clean images. Experiments on synthetic and realistic noise removal demonstrate that, the DnCNN and ResNet networks trained with our self-supervised NAC strategy achieve comparable or better performance than the original ones and previous supervised/unsupervised/self-supervised networks. The code is publicly available at https://github.com/csjunxu/Noisy-As-Clean.

Cascaded Attention Guidance Network for Single Rainy Image Restoration

Restoring a rainy image with raindrops or rainstreaks of varying scales, directions, and densities is an extremely challenging task. Recent approaches attempt to leverage the rain distribution (e.g., location) as prior to generate satisfactory results. However, concatenation of a single distribution map with the rainy image or with intermediate feature maps is too simplistic to fully exploit the advantages of such priors. To further explore this valuable information, an advanced cascaded attention guidance network, dubbed as CAG-Net, is formulated and designed as a three-stage model. In the first stage, a multitask learning network is constructed for producing the attention map and coarse de-raining results simultaneously. Subsequently, the coarse results and the rain distribution map are concatenated and fed to the second stage for results refinement. In this stage, the attention map generation network from the first stage is used to formulate a novel semantic consistency loss for better detail recovery. In the third stage, a novel pyramidal "whereand- how" learning mechanism is formulated. At each pyramid level, a two-branch network is designed to take the features from previous stages as inputs to generate better attention-guidance features and de-raining features, which are then combined via a gating scheme to produce the final de-raining results. Moreover, the uncertainty maps are also generated in this stage for more accurate pixel-wise loss calculation. Extensive experiments are carried out for removing raindrops or rainstreaks from both synthetic and real rainy images, and CAG-Net is demonstrated to produce significantly better results than state-of-the-art models. Code will be publicly available after paper acceptance.

Deep Graph-Convolutional Image Denoising

Non-local self-similarity is well-known to be an effective prior for the image denoising problem. However, little work has been done to incorporate it in convolutional neural networks, which surpass non-local model-based methods despite only exploiting local information. In this paper, we propose a novel end-to-end trainable neural network architecture employing layers based on graph convolution operations, thereby creating neurons with non-local receptive fields. The graph convolution operation generalizes the classic convolution to arbitrary graphs. In this work, the graph is dynamically computed from similarities among the hidden features of the network, so that the powerful representation learning capabilities of the network are exploited to uncover self-similar patterns. We introduce a lightweight Edge-Conditioned Convolution which addresses vanishing gradient and over-parameterization issues of this particular graph convolution. Extensive experiments show state-of-the-art performance with improved qualitative and quantitative results on both synthetic Gaussian noise and real noise.

Blind Universal Bayesian Image Denoising with Gaussian Noise Level Learning

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

Efficient and Fast Real-World Noisy Image Denoising by Combining Pyramid Neural Network and Two-Pathway Unscented Kalman Filter

Recently, image prior learning has emerged as an effective tool for image denoising, which exploits prior knowledge to obtain sparse coding models and utilize them to reconstruct the clean image from the noisy one. Albeit promising, these prior-learning based methods suffer from some limitations such as lack of adaptivity and failed attempts to improve performance and efficiency simultaneously. With the purpose of addressing these problems, in this paper, we propose a Pyramid Guided Filter Network (PGF-Net) integrated with pyramid-based neural network and Two-Pathway Unscented Kalman Filter (TP-UKF). The combination of pyramid network and TP-UKF is based on the consideration that the former enables our model to better exploit hierarchical and multi-scale features, while the latter can guide the network to produce an improved (a posteriori) estimation of the denoising results with fine-scale image details. Through synthesizing the respective advantages of pyramid network and TP-UKF, our proposed architecture, in stark contrast to prior learning methods, is able to decompose the image denoising task into a series of more manageable stages and adaptively eliminate the noise on real images in an efficient manner. We conduct extensive experiments and show that our PGF-Net achieves notable improvement on visual perceptual quality and higher computational efficiency compared to state-of-the-art methods.

Connecting Image Denoising and High-Level Vision Tasks via Deep Learning

Image denoising and high-level vision tasks are usually handled independently in the conventional practice of computer vision, and their connection is fragile. In this paper, we cope with the two jointly and explore the mutual influence between them with the focus on two questions, namely (1) how image denoising can help improving high-level vision tasks, and (2) how the semantic information from high-level vision tasks can be used to guide image denoising. First for image denoising we propose a convolutional neural network in which convolutions are conducted in various spatial resolutions via downsampling and upsampling operations in order to fuse and exploit contextual information on different scales. Second we propose a deep neural network solution that cascades two modules for image denoising and various high-level tasks, respectively, and use the joint loss for updating only the denoising network via backpropagation. We experimentally show that on one hand, the proposed denoiser has the generality to overcome the performance degradation of different high-level vision tasks. On the other hand, with the guidance of high-level vision information, the denoising network produces more visually appealing results. Extensive experiments demonstrate the benefit of exploiting image semantics simultaneously for image denoising and highlevel vision tasks via deep learning. The code is available online: https://github.com/Ding-Liu/DeepDenoising.

Learning No-Reference Quality Assessment of Multiply and Singly Distorted Images with Big Data

Previous research on no-reference (NR) quality assessment of multiply-distorted images focused mainly on three distortion types (white noise, Gaussian blur, and JPEG compression), while in practice images can be contaminated by many other common distortions due to the various stages of processing. Although MUSIQUE (MUltiply-and Singly-distorted Image QUality Estimator) Zhang et al., TIP 2018 is a successful NR algorithm, this approach is still limited to the three distortion types. In this paper, we extend MUSIQUE to MUSIQUE-II to blindly assess the quality of images corrupted by five distortion types (white noise, Gaussian blur, JPEG compression, JPEG2000 compression, and contrast change) and their combinations. The proposed MUSIQUE-II algorithm builds upon the classification and parameter-estimation framework of its predecessor by using more advanced models and a more comprehensive set of distortion-sensitive features. Specifically, MUSIQUE-II relies on a three-layer classification model to identify 19 distortion types. To predict the five distortion parameter values, MUSIQUE-II extracts an additional 14 contrast features and employs a multi-layer probability-weighting rule. Finally, MUSIQUE-II employs a new most-apparent-distortion strategy to adaptively combine five quality scores based on outputs of three classification models. Experimental results tested on three multiply-distorted and six singly-distorted image quality databases show that MUSIQUE-II yields not only a substantial improvement in quality predictive performance as compared with its predecessor, but also highly competitive performance relative to other state-of-the-art FR/NR IQA algorithms.

DeepLSR: a deep learning approach for laser speckle reduction

Speckle artifacts degrade image quality in virtually all modalities that utilize coherent energy, including optical coherence tomography, reflectance confocal microscopy, ultrasound, and widefield imaging with laser illumination. We present an adversarial deep learning framework for laser speckle reduction, called DeepLSR (https://durr.jhu.edu/DeepLSR), that transforms images from a source domain of coherent illumination to a target domain of speckle-free, incoherent illumination. We apply this method to widefield images of objects and tissues illuminated with a multi-wavelength laser, using light emitting diode-illuminated images as ground truth. In images of gastrointestinal tissues, DeepLSR reduces laser speckle noise by 6.4 dB, compared to a 2.9 dB reduction from optimized non-local means processing, a 3.0 dB reduction from BM3D, and a 3.7 dB reduction from an optical speckle reducer utilizing an oscillating diffuser. Further, DeepLSR can be combined with optical speckle reduction to reduce speckle noise by 9.4 dB. This dramatic reduction in speckle noise may enable the use of coherent light sources in applications that require small illumination sources and high-quality imaging, including medical endoscopy.

High-throughput, high-resolution deep learning microscopy based on registration-free generative adversarial network

We combine a generative adversarial network (GAN) with light microscopy to achieve deep learning super-resolution under a large field of view (FOV). By appropriately adopting prior microscopy data in an adversarial training, the neural network can recover a high-resolution, accurate image of new specimen from its single low-resolution measurement. Its capacity has been broadly demonstrated via imaging various types of samples, such as USAF resolution target, human pathological slides, fluorescence-labelled fibroblast cells, and deep tissues in transgenic mouse brain, by both wide-field and light-sheet microscopes. The gigapixel, multi-color reconstruction of these samples verifies a successful GAN-based single image super-resolution procedure. We also propose an image degrading model to generate low resolution images for training, making our approach free from the complex image registration during training data set preparation. After a well-trained network has been created, this deep learning-based imaging approach is capable of recovering a large FOV (~95 mm2) enhanced resolution of ~1.7 μm at high speed (within 1 second), while not necessarily introducing any changes to the setup of existing microscopes.

Multispectral Image Restoration via Inter- and Intra-block Sparse Estimation based on Physically-induced Joint Spatiospectral Structures

Existing low-level vision algorithms (e.g., those for superresolution, denoising, deblurring etc.) were primarily motivated and optimized for precision in spatial domain. However, high precision in spectral domain is of importance for many applications in scientific and technical fields, such as spectral analysis, recognition, and classification. In quest for both high spectral and spatial fidelity we introduce previously-unexplored, physically-induced, joint spatiospectral sparsities to improve existing methods for multispectral image restoration. The bidirectional image formation model is used to reveal that the discontinuities of a multispectral image tend to align spatially across different spectral bands; in other words, the 2D Laplacians of different bands are not only sparse each, but they also agree with one the other in significance positions. Such strongly structured sparsities give rise to a new inter-and intra-block sparse estimation approach. The estimation is performed on 3D spatiospectral sample blocks, rather than on separate 2D patches, one per spectral band or per luminance and chrominance component as in current practice. Moreover, intra-block and inter-block sparsity priors are combined via an intra-block ℓ1,2-norm minimization term and an inter-block low rank term, strengthening the regularization of the underlying inverse problem. The new approach is tested and evaluated on two concrete applications: superresolving and denoising multispectral images; its validity and advantages over the current state of the art are established by empirical results.

Block-matching sparsity regularization-based image reconstruction for low-dose computed tomography

PURPOSE

Low-dose computed tomography (CT) imaging has been widely explored because it can reduce the radiation risk to human bodies. This presents challenges in improving the image quality because low radiation dose with reduced tube current and pulse duration introduces severe noise. In this study, we investigate block-matching sparsity regularization (BMSR) and devise an optimization problem for low-dose image reconstruction.

METHOD

The objective function of the program is built by combining the sparse coding of BMSR and analysis error, which is subject to physical data measurement. A practical reconstruction algorithm using hard thresholding and projection-onto-convex-set for fast and stable performance is developed. An efficient scheme for the choices of regularization parameters is analyzed and designed.

RESULTS

In the experiments, the proposed method is compared with a conventional edge preservation method and adaptive dictionary-based iterative reconstruction. Experiments with clinical images and real CT data indicate that the obtained results show promising capabilities in noise suppression and edge preservation compared with the competing methods.

CONCLUSIONS

A block-matching-based reconstruction method for low-dose CT is proposed. Improvements in image quality are verified by quantitative metrics and visual comparisons, thereby indicating the potential of the proposed method for real-life applications.

Single Image Super-Resolution Based on Wiener Filter in Similarity Domain

Single image super-resolution (SISR) is an ill-posed problem aiming at estimating a plausible high-resolution (HR) image from a single low-resolution image. Current state-of-the-art SISR methods are patch-based. They use either external data or internal self-similarity to learn a prior for an HR image. External data-based methods utilize a large number of patches from the training data, while self-similarity-based approaches leverage one or more similar patches from the input image. In this paper, we propose a self-similarity-based approach that is able to use large groups of similar patches extracted from the input image to solve the SISR problem. We introduce a novel prior leading to the collaborative filtering of patch groups in a 1D similarity domain and couple it with an iterative back-projection framework. The performance of the proposed algorithm is evaluated on a number of SISR benchmark data sets. Without using any external data, the proposed approach outperforms the current non-convolutional neural network-based methods on the tested data sets for various scaling factors. On certain data sets, the gain is over 1 dB, when compared with the recent method A+. For high sampling rate (x4), the proposed method performs similarly to very recent state-of-the-art deep convolutional network-based approaches.

Block matching sparsity regularization-based image reconstruction for incomplete projection data in computed tomography

In medical imaging many conventional regularization methods, such as total variation or total generalized variation, impose strong prior assumptions which can only account for very limited classes of images. A more reasonable sparse representation frame for images is still badly needed. Visually understandable images contain meaningful patterns, and combinations or collections of these patterns can be utilized to form some sparse and redundant representations which promise to facilitate image reconstructions. In this work, we propose and study block matching sparsity regularization (BMSR) and devise an optimization program using BMSR for computed tomography (CT) image reconstruction for an incomplete projection set. The program is built as a constrained optimization, minimizing the L1-norm of the coefficients of the image in the transformed domain subject to data observation and positivity of the image itself. To solve the program efficiently, a practical method based on the proximal point algorithm is developed and analyzed. In order to accelerate the convergence rate, a practical strategy for tuning the BMSR parameter is proposed and applied. The experimental results for various settings, including real CT scanning, have verified the proposed reconstruction method showing promising capabilities over conventional regularization.

Noise creates polarization artefacts

The accuracy of calculations of both the degree and angle of polarization depend strongly on the noise in the measurements used. The noise in the measurements recorded by both camera based systems and spectrometers can lead to significant artefacts and incorrect conclusions about high degrees of polarization when in fact none exist. Three approaches are taken in this work: firstly, the absolute error introduced as a function of the signal to noise ratio for polarization measurements is quantified in detail. An important finding here is the reason for why several studies incorrectly suggest that black (low reflectivity) objects are highly polarized. The high degree of polarization is only an artefact of the noise in the calculation. Secondly, several simple steps to avoid such errors are suggested. Thirdly, if these points can not be followed, two methods are presented for mitigating the effects of noise: a maximum likelihood estimation method and a new denoising algorithm to best calculate the degree of polarization of natural polarization information.

Category-Specific Object Image Denoising

We present a novel image denoising algorithm that uses external, category specific image database. In contrast to existing noisy image restoration algorithms that search patches either from a generic database or noisy image itself, our method first selects clean images similar to the noisy image from a database that consists of images of the same class. Then, within the spatial locality of each noisy patch, it assembles a set of "support patches" from the selected images. These noisy-free support samples resemble the noisy patch and correspond principally to the identical part of the depicted object. In addition, we employ a content adaptive distribution model for each patch, where we derive the parameters of the distribution from the support patches. We formulate noise removal task as an optimization problem in the transform domain. Our objective function composed of a Gaussian fidelity term that imposes category specific information, and a low-rank term that encourages the similarity between the noisy and the support patches in a robust manner. The denoising process is driven by an iterative selection of support patches and optimization of the objective function. Our extensive experiments on five different object categories confirm the benefit of incorporating category-specific information to noise removal and demonstrate the superior performance of our method over the state-of-the-art alternatives.