Preservation Of Fine Details Research Articles

ToonCrafter: Generative Cartoon Interpolation

We introduce ToonCrafter, a novel approach that transcends traditional correspondence-based cartoon video interpolation, paving the way for generative interpolation. Traditional methods, that implicitly assume linear motion and the absence of complicated phenomena like dis-occlusion, often struggle with the exaggerated non-linear and large motions with occlusion commonly found in cartoons, resulting in implausible or even failed interpolation results. To overcome these limitations, we explore the potential of adapting live-action video priors to better suit cartoon interpolation within a generative framework. ToonCrafter effectively addresses the challenges faced when applying live-action video motion priors to generative cartoon interpolation. First, we design a toon rectification learning strategy that seamlessly adapts live-action video priors to the cartoon domain, resolving the domain gap and content leakage issues. Next, we introduce a dual-reference-based 3D decoder to compensate for lost details due to the highly compressed latent prior spaces, ensuring the preservation of fine details in interpolation results. Finally, we design a flexible sketch encoder that empowers users with interactive control over the interpolation results. Experimental results demonstrate that our proposed method not only produces visually convincing and more natural dynamics, but also effectively handles dis-occlusion. The comparative evaluation demonstrates the notable superiority of our approach over existing competitors. Code and model weights are available at https://doubiiu.github.io/projects/ToonCrafter

A Comprehensive Method for Example-Based Color Transfer with Holistic-Local Balancing and Unit-Wise Riemannian Information Gradient Acceleration.

Color transfer, an essential technique in image editing, has recently received significant attention. However, achieving a balance between holistic color style transfer and local detail refinement remains a challenging task. This paper proposes an innovative color transfer method, named BHL, which stands for Balanced consideration of both Holistic transformation and Local refinement. The BHL method employs a statistical framework to address the challenge of achieving a balance between holistic color transfer and the preservation of fine details during the color transfer process. Holistic color transformation is achieved using optimal transport theory within the generalized Gaussian modeling framework. The local refinement module adjusts color and texture details on a per-pixel basis using a Gaussian Mixture Model (GMM). To address the high computational complexity inherent in complex statistical modeling, a parameter estimation method called the unit-wise Riemannian information gradient (uRIG) method is introduced. The uRIG method significantly reduces the computational burden through the second-order acceleration effect of the Fisher information metric. Comprehensive experiments demonstrate that the BHL method outperforms state-of-the-art techniques in both visual quality and objective evaluation criteria, even under stringent time constraints. Remarkably, the BHL method processes high-resolution images in an average of 4.874 s, achieving the fastest processing time compared to the baselines. The BHL method represents a significant advancement in the field of color transfer, offering a balanced approach that combines holistic transformation and local refinement while maintaining efficiency and high visual quality.

DuINet: A dual-branch network with information exchange and perceptual loss for enhanced image denoising

Image denoising is a fundamental task in image processing and low-level computer vision, often necessitating a delicate balance between noise removal and the preservation of fine details. In recent years, deep learning approaches, particularly those utilizing various neural network architectures, have shown significant promise in addressing this challenge. In this study, we propose DuINet, a novel dual-branch network specifically designed to capture complementary aspects of image information. DuINet integrates an information exchange module that facilitates effective feature sharing between the branches, and it incorporates a perceptual loss function aimed at enhancing the visual quality of the denoised images. Extensive experimental results demonstrate that DuINet surpasses existing dual-branch models and several state-of-the-art convolutional neural network (CNN)-based methods, particularly under conditions of severe noise where preserving fine details and textures is critical. Moreover, DuINet maintains competitive performance in terms of the LPIPS index when compared to deeper or larger networks such as Restormer and MIRNet, underscoring its ability to deliver high visual quality in denoised images.

A Hybrid Approach for CT Image Noise Reduction Combining Method Noise-CNN and Shearlet Transform

The presence of gaussian noise commonly weakens the diagnostic precision of low-dose CT imaging. A novel CT image denoising technique that integrates the non-subsampled shearlet transform (NSST) with Bayesian thresholding, and incorporates a modern method noise Deep Convolutional neural network (DCNN) based post-processing operation on denoised images to strengthen low-dose CT imaging quality. The hybrid method commences with NSST and Bayesian thresholding to mitigate the initial noise while preserving crucial image features, such as corners and edges. The novel aspect of the proposed approach is its successive application of a DnCNN on initial denoised image, which learns and removes residual noise patterns from denoised images, thereby enhancing fine detail preservation. This dual-phase strategy addresses both noise suppression and image-detail preservation. The proposed technique is evaluated through the use of metrics, such as PSNR, SNR, SSIM, ED, and UIQI. The results demonstrate that the hybrid approach outperforms standard denoising techniques in preserving image quality and fine details.

Unsupervised low-dose CT denoising using bidirectional contrastive network

Background and Objective:Low-dose computed tomography (LDCT) scans significantly reduce radiation exposure, but introduce higher levels of noise and artifacts that compromise image quality and diagnostic accuracy. Supervised learning methods have proven effective in denoising LDCT images, but are hampered by the need for large, paired datasets, which pose significant challenges in data acquisition. This study aims to develop a robust unsupervised LDCT denoising method that overcomes the reliance on paired LDCT and normal-dose CT (NDCT) samples, paving the way for more accessible and practical denoising techniques. Methods:We propose a novel unsupervised network model, Bidirectional Contrastive Unsupervised Denoising (BCUD), for LDCT denoising. This model innovatively combines a bidirectional network structure with contrastive learning theory to map the precise mutual correspondence between the noisy LDCT image domain and the clean NDCT image domain. Specifically, we employ dual encoders and discriminators for domain-specific data generation, and use unique projection heads for each domain to adaptively learn customized embedded representations. We then align corresponding features across domains within the learned embedding spaces to achieve effective noise reduction. This approach fundamentally improves the model’s ability to match features in latent space, thereby improving noise reduction while preserving fine image detail. Results:Through extensive experimental validation on the AAPM-Mayo public dataset and real-world clinical datasets, the proposed BCUD method demonstrated superior performance. It achieved a peak signal-to-noise ratio (PSNR) of 31.387 dB, a structural similarity index measure (SSIM) of 0.886, an information fidelity criterion (IFC) of 2.305, and a visual information fidelity (VIF) of 0.373. Notably, subjective evaluation by radiologists resulted in a mean score of 4.23, highlighting its advantages over existing methods in terms of clinical applicability. Conclusions:This paper presents an innovative unsupervised LDCT denoising method using a bidirectional contrastive network, which greatly improves clinical applicability by eliminating the need for perfectly matched image pairs. The method sets a new benchmark in unsupervised LDCT image denoising, excelling in noise reduction and preservation of fine structural details.

Denoising of poisson-corrupted microscopic biopsy images using fourth-order partial differential equation with ant colony optimization

Microscopic biopsies play a critical role in the identification of lung cancer. The readable features of lung such as tissue and cell are usually affected by Poisson noise. Moreover, the existence of noise reduces the quality of the image that limits the interpretation of the disease by medical specialists. To address this, an ant colony optimization based reformed fourth order partial differential equation i.e., ACO-RFPDE is proposed. In order to develop RFPDE, two major advancements have been incorporated. In first advancement, the restoration of edges and fine details preservation has been achieved by maximum likelihood estimation. In second advancement, instead of user based manual value, a mean absolute deviation of the image has been employed for automatic calculation of the edge threshold constraint of diffusion coefficient of FPDE. Furthermore, ACO is utilized for tuning of the controlling parameters of the proposed method where peak signal to noise ratio is used as a fitness function. The performance is analyzed comparatively with existing methods as well as qualitatively and quantitatively for the LC25000 dataset. The quantitative analysis includes mean square error, correlation, normalized absolute error, no reference, universal quality index, peak signal-to-noise ratio, and structural similarity index. The following values 0.175, 0.964, 0.111, 42.45, 0.946, 55.68 and 0.985 of these metrics have been obtained by the proposed method for the images of lung squamous carcinoma cell. The results of the proposed method demonstrate the successful elimination of Poisson noise while preserving the image features like edges, boundaries and enhancing the abnormal region.

EDSR: Empowering super-resolution algorithms with high-quality DIV2K images

Remarkable strides have been made in Enhanced-resolution image reconstruction in recent times. In this research, we propose EDSR, a novel algorithm leveraging high-quality DIV2K images to enhance super-resolution algorithms’ performance. Our approach addresses challenges associated with capturing and reconstructing fine details and textures in low-resolution images. The training process utilizes a deep neural network fed with the extensive DIV2K dataset, comprising meticulously selected high-resolution images for super-resolution tasks. The network learns from these high-quality images to improve super-resolved outputs’ accuracy and fidelity. Extensive experimentation evaluates the effectiveness of the EDSR algorithm in achieving superior results. Experimental outcomes exhibit the remarkable progress made by EDSR compared to other leading super-resolution algorithms. The mean PSNR values obtained using EDSR on various test datasets indicate substantial improvement in image quality. The SSIM scores show enhanced preservation of fine details and textures. In the Nature category, the EDSR algorithm exhibits impressive performance, achieving a mean Peak Signal-to-Noise Ratio (PSNR) of 35.92 dB and a Structural Similarity Index (SSIM) of 0.9321. When analyzing Portrait images, the algorithm consistently maintains high quality, delivering a mean PSNR of 34.17 dB and an SSIM of 0.9125. For Cityscape images, EDSR showcases excellent capabilities with a mean PSNR of 36.05 dB and an SSIM of 0.9382, highlighting its exceptional handling in this category. Lastly, in the Still Life category, the algorithm demonstrates competence, achieving a mean PSNR of 33.79 dB and an SSIM of 0.9024. In conclusion, the proposed EDSR algorithm empowers super-resolution algorithms by leveraging high-quality DIV2K images. The empirical findings validate that EDSR is highly efficient in enhancing image quality and preserving fine details.

D2Net: Deep Denoising Network in Frequency Domain for Hyperspectral Image

Dear Editor, Since the existing hyperspectral image denoising methods suffer from excessive or incomplete denoising, leading to information distortion and loss, this letter proposes a deep denoising network in the frequency domain, termed D2Net. Our motivation stems from the observation that images from different hyperspectral image (HSI) bands share the same structural and contextual features while the reflectance variations in the spectra are mainly fallen on the details and textures. We design the D2Net in three steps: 1) spatial decomposition, 2) spatial-spectral denoising, and 3) refined reconstruction. It achieves multi-scale feature learning without information loss by adopting the rigorous symmetric discrete wavelet transform (DWT) and inverse discrete wavelet transform (IDWT). In particular, the specific design for different frequency components ensures complete noise removal and preservation of fine details. Experiment results demonstrate that our D2Net can attain a promising denoising performance.

Conditional Generative Adversarial Networks for low-dose CT image denoising aiming at preservation of critical image content.

X-ray Computed Tomography (CT) is an imaging modality where patients are exposed to potentially harmful ionizing radiation. To limit patient risk, reduced-dose protocols are desirable, which inherently lead to an increased noise level in the reconstructed CT scans. Consequently, noise reduction algorithms are indispensable in the reconstruction processing chain. In this paper, we propose to leverage a conditional Generative Adversarial Networks (cGAN) model, to translate CT images from low-to-routine dose. However, when aiming to produce realistic images, such generative models may alter critical image content. Therefore, we propose to employ a frequency-based separation of the input prior to applying the cGAN model, in order to limit the cGAN to high-frequency bands, while leaving low-frequency bands untouched. The results of the proposed method are compared to a state-of-the-art model within the cGAN model as well as in a single-network setting. The proposed method generates visually superior results compared to the single-network model and the cGAN model in terms of quality of texture and preservation of fine structural details. It also appeared that the PSNR, SSIM and TV metrics are less important than a careful visual evaluation of the results. The obtained results demonstrate the relevance of defining and separating the input image into desired and undesired content, rather than blindly denoising entire images. This study shows promising results for further investigation of generative models towards finding a reliable deep learning-based noise reduction algorithm for low-dose CT acquisition.

Video Vectorization via Bipartite Diffusion Curves Propagation and Optimization.

We propose a new video vectorization approach for converting videos in the raster format to vector representation with the benefits of resolution independence and compact storage. Through classifying extracted curves in each video frame into salient ones and non-salient ones, we introduce a novel bipartite diffusion curves (BDCs) representation in order to preserve both important image features such as sharp boundaries and regions with smooth color variation. This bipartite representation allows us to propagate non-salient curves across frames such that the propagation, in conjunction with geometry optimization and color optimization of salient curves, ensures the preservation of fine details within each frame and across different frames, and meanwhile, achieves good spatial-temporal coherence. Thorough experiments on a variety of videos show that our method is capable of converting videos to the vector representation with low reconstruction errors, low computational cost, and fine details, demonstrating our superior performance over the state of the art. We also show that, when used for video upsampling, our method produces results comparable to video super-resolution.

A Review on SAR Image and its Despeckling

The method of speckle reduction is widely used in synthetic aperture radar (SAR) imagery over the last three decades. The SAR images are inherently speckled in nature. Speckle noise is a granular pattern distribution, usually modeled as a multiplicative noise that affects the SAR images, as well as all coherent images. Other SAR related problems are also discussed in this paper. Therefore, despeckling approaches are needed to improve the quality of SAR images. However, there is a trade-off between speckle reduction and the preservation of fine details in the despeckled SAR image. The reduction of the speckle noise without losing the fine details of the SAR image is a diffucult task. However, many despeckling methods have been discussed to reduce the speckle noise from the SAR images. Each method has their own norms, advantages and disadvantages. This article contains a review of some major work in the field of SAR image despeckling. Often, scientists and scholars have faced the struggle to understand the pattern distribution of the speckle noise in SAR images. Hence, a brief details about radar, SAR imaging, speckle noise in SAR images and the prevalent approaches of SAR image despeckling are reviewed here. The advantages and disadvantages of SAR image despeckling approaches are also analysed and discussed.

De-Noising of Corrupted Fluoroscopy Images Based on a New Multi-Line Algorithm

Fluoroscopic images are a field of medical images that depends on the quality of image for correct diagnosis; the main trouble is the de-nosing and how to keep the poise between degradation of noisy image, from one side, and edge and fine details preservation, from the other side, especially when fluoroscopic images contain black and white type noise with high density. The previous filters could usually handle low/medium black and white type noise densities, that expense edge, =fine details preservation and fail with high density of noise that corrupts the images. Therefore, this paper proposed a new Multi-Line algorithm that deals with high-corrupted image with high density of black and white type noise. The experiments achieved images with a high quality and effectively preserved edge and fine details against black and white type noise densities depending on Peak Signal to Noise Ratio (PSNR), Mean Squared Error (MSE), and Image Enhancement Factor (IEF) measures.

Total Variation-Based Noise Reduction Image Processing Algorithm for Confocal Laser Scanning Microscopy Applied to Activity Assessment of Early Carious Lesions

The confocal laser scanning microscopy (CLSM) system has been widely used to analyze early carious lesions with fluorescent ligands in dental imaging. This system can be used to examine the physiological condition of cellular colonization in the tooth structure. However, the undesirable noise in CLSM images hinders accurate activity assessment of early carious lesions. To address this limitation, a total variation (TV)-based noise reduction algorithm with good edge preservation was developed, and its applicability to medical tooth specimen images obtained with CLSM was verified. To evaluate the imaging performance, the proposed algorithm was compared with conventional filtering methods in terms of the normalized noise power spectrum, contrast-to-noise ratio, and coefficient of variation. The results indicate that the proposed algorithm achieved better noise performance and fine-detail preservation, in comparison with the conventional methods.

Adaptive multi-threshold based de-noising filter for medical image applications

Medical image processing is the emerging research area and many researchers contributed to medical image processing by proposing new techniques for medical image enhancement and abnormality detection. Interpretation of medical images is a challenging problem because of the unavoidable noise produced by the medical imaging devices and interference. In this work, a new framework is proposed for noise detection and reduction. This framework comprises two phases. First phase is the noise detection phase which is performed using the newly proposed adaptive multi-threshold scheme (AMT). In second phase, modification of noisy pixel is done using edge preserving median filter (EPM), which conserves the edge component and controls the blurring effect with preservation of fine details of interior region. The proposed work is tested with benchmark images and few medical images. It produces promising result and the results are compared with existing two-stage noise reduction techniques. Popular performance metrics such PSNR and SSIM are used for evaluation. Quantitative analysis and experimental results demonstrate that the proposed method is more efficient and suitable for medical image pre-processing.

Adaptive multi-threshold based de-noising filter for medical image applications

Medical image processing is the emerging research area and many researchers contributed to medical image processing by proposing new techniques for medical image enhancement and abnormality detection. Interpretation of medical images is a challenging problem because of the unavoidable noise produced by the medical imaging devices and interference. In this work, a new framework is proposed for noise detection and reduction. This framework comprises two phases. First phase is the noise detection phase which is performed using the newly proposed adaptive multi-threshold scheme (AMT). In second phase, modification of noisy pixel is done using edge preserving median filter (EPM), which conserves the edge component and controls the blurring effect with preservation of fine details of interior region. The proposed work is tested with benchmark images and few medical images. It produces promising result and the results are compared with existing two-stage noise reduction techniques. Popular performance metrics such PSNR and SSIM are used for evaluation. Quantitative analysis and experimental results demonstrate that the proposed method is more efficient and suitable for medical image pre-processing.

FIB-SEM imaging properties of Drosophila melanogaster tissues embedded in Lowicryl HM20.

Lowicryl resins enable processing of biological material for electron microscopy at the lowest temperatures compatible with resin embedding. When combined with high-pressure freezing and freeze-substitution, Lowicryl embedding supports preservation of fine structural details and fluorescent markers. Here, we analysed the applicability of Lowicryl HM20 embedding for focused ion beam (FIB) scanning electron microscopy (SEM) tomography of Drosophila melanogaster embryonic and larval model systems. We show that the freeze-substitution with per-mill concentrations of uranyl acetate provided sufficient contrast and an image quality of SEM imaging in the range of similar samples analysed by transmission electron microscopy (TEM). Preservation of genetically encoded fluorescent proteins allowed correlative localization of regions of interest (ROI) within the embedded tissue block. TEM on sections cut from the block face enabled evaluation of structural preservation to allow ROI ranking and thus targeted, time-efficient FIB-SEM tomography data collection. The versatility of Lowicryl embedding opens new perspectives for designing hybrid SEM-TEM workflows to comprehensively analyse biological structures. LAY DESCRIPTION: Focused ion beam scanning electron microscopy is becoming a widely used technique for the three-dimensional analysis of biological samples at fine structural details beyond levels feasible for light microscopy. To withstand the abrasion of material by the ion beam and the imaging by the scanning electron beam, biological samples have to be embedded into resins, most commonly these are very dense epoxy-based plastics. However, dense resins generate electron scattering which interferes with the signal from the biological specimen. Furthermore, to improve the imaging contrast, epoxy embedding requires chemical treatments with e.g. heavy metals, which deteriorate the ultrastructure of the biological specimen. In this study we explored the applicability of an electron lucent resin, Lowicryl HM 20, for focused ion beam scanning electron microscopy. The Lowicryl embedding workflow operates at milder chemical treatments and lower temperatures, thus preserving the sub-cellular and sub-organellar organization, as well as fluorescent markers visible by light microscopy. Here we show that focus ion beam scanning electron microscopy of Lowicryl-embedded fruit flies tissues provides reliable imaging revealing fine structural details. Our workflow benefited from use of transmission electron microscopy for the quality control of the ultrastructural preservation and fluorescent light microscopy for localization of regions of interest. The versatility of Lowicryl embedding opens up new perspectives for designing hybrid workflows combining fluorescent light, scanning, and transmission electron microscopy techniques to comprehensively analyze biological structures.

Video Denoising with Optical Flow Estimation

In this paper we describe the implementation of state-of-the-art video denoising algorithm SPTWO [A. Buades, J.L. Lisani, M. Miladinovic, Patch Based Video Denoising with Optical Flow Estimation, IEEE Transactions on Image Processing 25 (6), 2573--2586]. This algorithm, inspired by image fusion techniques, uses motion compensation by regularized optical flow methods, which permits robust patch comparison in spatiotemporal volumes. Groups of similar patches are denoised using Principal Component Analysis, which ensures the correct preservation of fine texture and details.

An efficient nonlinear approach for removing fixed-value impulse noise from grayscale images

Removal of salt and pepper noise has been one of the most interesting researches in the field of image preprocessing tasks; it has two simultaneous stringent demands: the suppression of impulses and the preservation of fine details. To address this problem, a scheme based on nonlinear filters is proposed; it consists of the introduction of a redescending M-estimator within the modified nearest neighbor filter. In order to analyze all pixels in the neighborhood, as well as to reduce the magnitude of the existing impulses, a redescending M-estimator is used; the remaining pixels are processed by the modified nearest neighbor filter to obtain the best estimation of a noise-free pixel. The impulsive suppression is applied on the entire image by using a sliding window; the local information obtained by this one also allows to calculate the thresholds that characterize the influence function tested in the redescending M-estimator. To suppress high density fixed-value impulse noise in large-size grayscale images, the proposal is implemented on a heterogeneous CPU–GPU architecture. The noise reduction and the processing time of the proposed approach are evaluated by extensive simulations; its effectiveness is verified by quantitative and qualitative results.

A Effectual Technique of Impulse Noise Suppression for Assessing the Impact of Brain Disorders in MRI by applying Selective Weighted Median Filter

An accurate analysis of medical images is progressively demanding in providing the absolute detection and diagnosis of diseases in medical imaging. The significant pre-processing step in MRI data processing is noise elimination. Noise deletion is essential step to increase image quality and performance of all the tasks desirable for quantitative imaging analysis. In this paper a new scheme for impulse noise removal in corrupted MRI brain images is introduced. The proposed scheme is a simple & efficient filtering technique that effectively detects and removes the salt and pepper noise. The experimental results of suggested noise purifying process executed on standard set of assessment images shows that algorithm provides a very good results with low mean-squared-error and high signal-to-noise ratio values for noise density up to 95% and outperforms significant tradeoff between fine detail preservation and noise removal in brain MRI images.

Poisson Wiener filtering with non-local weighted parameter estimation using stochastic distances

The Wiener filter is a classical approach for signal denoising and it is known to be the optimum filter in the sense of the linear minimum mean square error (LMMSE). In general, knowledge of some statistical parameters of the underlying signal and noise are required to compute the Wiener filter. These parameters are usually estimated from the corrupted image using a local neighborhood, thus assuming that the signal and noise are locally stationary. However, images corrupted by signal-dependent noise, such as Poisson noise, are not locally stationary. In this paper, we are proposing a novel Wiener filtering approach for Poisson corrupted images considering a non-local weighted parameter estimation. In the proposed method, named Poisson Non-Local Wiener filter (PNL-Wiener), filtering parameters are estimated from the degraded image considering a non-local neighborhood, where the weights of the estimation function are computed based on the stochastic distances between image patches. Experimental results show that the proposed method is competitive to other state-of-the-art denoising methods designed specifically for Poisson corrupted images, yet providing better preservation of edges and fine details in the images.