A deep variational Bayesian framework for blind image deblurring

Blind image deconvolution, i.e., estimating both the latent image and the blur kernel from the only observed blurry image, is a severely ill-posed inverse problem. In this paper, we propose a blur kernel estimation method for blind motion deblurring using sparse representation and cross-scale self-similarity of image patches as priors to recover the latent sharp image from a single blurry image. Sparse representation indicates that image patches can always be represented well as a sparse linear combination of atoms in an appropriate dictionary. Cross-scale self-similarity results in that any image patch can in some way be well approximated by a number of other similar patches across different image scales. Our method is based on the observations that almost any image patch in a natural image has multiple similar patches in down-sampled versions of the image, and down-sampling produces image patches that are sharper than those in the blurry image itself. In our method, the dictionary for sparse representation is trained adaptively from sharper patches sampled from the down-sampled latent image estimate to make the similar patches of the latent sharp image well represented sparsely, and meanwhile, all patches from the latent image estimate are optimized to be as close to the sharper similar patches searched from the down-sampled version to enforce the sharp recovery of the latent image by constructing a non-local regularization. Experimental results on both simulated and real blurry images demonstrate that our method outperforms state-of-the-art blind deblurring methods.

Blind image deblurring, i.e., deblurring without knowledge of the blur kernel, is a highly ill-posed problem. The problem can be solved in two parts: i) estimate a blur kernel from the blurry image, and ii) given an estimated blur kernel, de-convolve the blurry input to restore the target image. In this paper, we propose a graph-based blind image deblurring algorithm by interpreting an image patch as a signal on a weighted graph. Specifically, we first argue that a skeleton image-a proxy that retains the strong gradients of the target but smooths out the details-can be used to accurately estimate the blur kernel and has a unique bi-modal edge weight distribution. Then, we design a reweighted graph total variation (RGTV) prior that can efficiently promote a bi-modal edge weight distribution given a blurry patch. Further, to analyze RGTV in the graph frequency domain, we introduce a new weight function to represent RGTV as a graph l1-Laplacian regularizer. This leads to a graph spectral filtering interpretation of the prior with desirable properties, including robustness to noise and blur, strong piecewise smooth (PWS) filtering and sharpness promotion. Minimizing a blind image deblurring objective with RGTV results in a non-convex non-differentiable optimization problem. Leveraging the new graph spectral interpretation for RGTV, we design an efficient algorithm that solves for the skeleton image and the blur kernel alternately. Specifically for Gaussian blur, we propose a further speedup strategy for blind Gaussian deblurring using accelerated graph spectral filtering. Finally, with the computed blur kernel, recent non-blind image deblurring algorithms can be applied to restore the target image. Experimental results demonstrate that our algorithm successfully restores latent sharp images and outperforms state-of-the-art methods quantitatively and qualitatively.

Blind image deblurring is a conundrum because there are infinitely many pairs of latent image and blur kernel. To get a stable and reasonable deblurred image, proper prior knowledge of the latent image and the blur kernel is urgently required. Different from the recent works on the statistical observations of the difference between the blurred image and the clean one, our method is built on the surface-aware strategy arising from the intrinsic geometrical consideration. This approach facilitates the blur kernel estimation due to the preserved sharp edges in the intermediate latent image. Extensive experiments demonstrate that our method outperforms the state-of-the-art methods on deblurring the text and natural images. Moreover, our method can achieve attractive results in some challenging cases, such as low-illumination images with large saturated regions and impulse noise. A direct extension of our method to the non-uniform deblurring problem also validates the effectiveness of the surface-aware prior.

Blind image deblurring is a long-standing challenge in remote sensing image restoration tasks. It aims to recover a latent sharp image from a blurry image while the blur kernel is unknown. To solve this problem, many image priors-based algorithms and learning-based algorithms have been proposed. However, most of these methods are based on a single blurry image. Due to the lack of high frequency information, the images restored by these algorithms still have some deficiencies in edge and texture details. In this work, we propose a novel deep learning model named Reference-Based Multi-Level Features Fusion Deblurring Network (Ref-MFFDN), which registers the reference image and the blurry image in the multi-level feature space and transfers the high-quality textures from registered reference features to assist image deblurring. Comparative experiments on the testing set prove that our Ref-MFFDN outperforms many state-of-the-art single image deblurring approaches in both quantitative evaluation and visual results, which indicates the effectiveness of using reference images in remote sensing image deblurring tasks. More ablation experiments demonstrates the robustness of Ref-MFFDN to the input image size, the effectiveness of multi-level features fusion network (MFFN) and the effect of different feature levels in multi-feature extractor (MFE) on algorithm performance.

Blind image deblurring problem, estimating the blur kernel and latent image from a single blurry image, is the a severely ill-posed problem. Since the blur kernel plays an important role in image deblurring problem, we focus on how to effectively apply Elastic-Net prior(ENP) to estimate blur kernel. During the estimation of kernel blur, the ENP is used to maintain the slight edge and removing insignificant texture of intermediate image. Thus, a kernel estimation method is proposed, based on Maximum a Posteriori(MAP) method with ENP. And then, the half-quadratic splitting method is used to solve the non-convex problem of this model. Finally, the latent image can be estimated by using the hyper-Laplacian priors and the estimated blur kernel. Experimental results show that the proposed method can obtain better blur kernel and improve the quality of the latent image estimation, against with the state-of-the art methods.

Objectives: To develop image deblurring algorithm to effectively recover the original image from the captured blurry and noisy image. Methods: A novel image deblurring algorithm using adaptive priors is proposed. The adaptive priors vary the sparsity induced based on factors such as the type of blur affecting the image, image region, statistical parameters regarding the blur kernel and/or statistical predictions on the blur kernel. The proposed algorithm using adaptive priors improves the quality of the deblurred result, when compared with various recent image deblurring algorithms. Finding: the work contains a case study with regard to certain standard parameters. It is observed that proposed method is better in terms of frequency response, Peak signal-tonoise ratio (PSNR) and Structural Similarity Index (SSIM) values in comparison with other priors. Novelty: the proposed priors lead to the most effective results for image deblurring using the Bayesian framework. The proposed method improves the performance by 30% in PSNR and 45% in SSIM values in dB with uniform kernel size 12 * 12 and improves performance by 30 % in PSNR and 32% SSIM with a standard deviation of 3.5. The proposed method enhances the frequency response of the real-time image restoration process. Applications: Some important applications include the restoration of medical images such as MRI images, CT images where the intensity of these radiations is maintained to avoid damage to human organs. Remote sensing images captured through drones at a specified time cannot be retaken and hence restoring such images is very important. Similarly restoring images from CCTV footage, astronomical images. Keywords: Blind Image Deblurring; Maximum a Posteriori Estimation; Image priors; Regularization; Structural Similarity Index (SSIM) and Peak signal-to-noise ratio (PSNR)

Blind image deblurring, which requires simultaneous estimation of the latent image and blur kernel, constitutes a classic ill-posed problem. To address this, priors based on , , and regularizations have been widely adopted. Based on this foundation and combining successful experiences of previous work, this paper introduces regularization, a novel nonlinear sparse regularization combining the and norms via nonlinear coupling. Statistical probability analysis demonstrates that regularization achieves stronger sparsity than traditional regularizations like , , and regularizations. Furthermore, building upon the regularization, we propose a novel nonlinear sparse model for blind image deblurring. To optimize the proposed regularization, we introduce an Adaptive Generalized Soft-Thresholding (AGST) algorithm and further develop an efficient optimization strategy by integrating AGST with the Half-Quadratic Splitting (HQS) strategy. Extensive experiments conducted on synthetic datasets and real-world images demonstrate that the proposed nonlinear sparse model achieves superior deblurring performance while maintaining completive computational efficiency.

The blind facial-image deconvolution problem is ill-posed so that the estimation of latent image and blur kernel is hard to obtain without additional priors or heuristics. Although data-driven methods with deep neural networks have achieved excellent performance in this issue, most existing methods are designed like black boxes without transparency and interpretability. Thus, a model-guided explainable approach is proposed to address the above issues on blurry-facial images’ recovery, named the model-guided deep neural network (MG-DNN). To break the barrier between the network design and interpretability, we optimize blind image deconvolution based on salient edge regularization (BID-SER) and use it to guide the proposed MG-DNN. First, basic units for processing image and kernel features are designed with convolutional operators. Then, unfolding the optimization of BID-SER, the multiblocks concatenation is designed by several kernel networks and image networks with basic units. Finally, comprehensive experiments evaluated by several metrics demonstrate that the proposed MG-DNN performs favorably with existing state-of-the-art methods while processing on three different face data sets.

_ This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper OTC 35821, “A Mobile Platform for AI-Powered Underwater Image Processing,” by Chenyu Zhou, University of Houston; Lily Walker, William & Mary; and Xin Fu, University of Houston, et al. The paper has not been peer reviewed. Copyright 2025 Offshore Technology Conference. _ Timely inspection of subsea infrastructure—particularly pipelines—is crucial for preventing oil spills and protecting the environment. This paper introduces a technology for offshore pipeline inspection centered on an autonomous robotic system equipped with underwater computer vision and edge-computing capabilities. The aim is to create a fast, cost-effective solution for inspecting underwater structures, enabling early detection of environmental hazards. Introduction Autonomous underwater vehicles (AUVs) have revolutionized the ability to explore and monitor marine environments. The cornerstone of AUV capabilities lies in their imaging systems, although the underwater environment poses unique and complex challenges for these systems. Additional image-processing and correction ability remains a goal for AUVs, creating computational demands that must be balanced against limited onboard power and processing capabilities (Fig. 1). Recently, advancements in deep learning have enabled deep neural networks (DNNs) to achieve remarkable success in solving a wide range of image-processing challenges, significantly surpassing the performance of traditional methods. In this research, the authors build upon foundational work in the literature to present a comprehensive workflow for adapting and deploying pretrained DNN models onto mobile computing platforms suitable for integration with AUVs. The described approach examines and documents the transition from powerful desktop systems to resource-constrained mobile platforms. Underwater Image Enhancement Motion blur is a common challenge in underwater image processing, especially when the camera is mounted on a moving platform such as an AUV. The lack of adequate lighting in underwater environments often requires longer exposure times, further exacerbating motion blur. The blurring process can be mathematically described as a convolution of the sharp image with a blur kernel, plus noise. If the blur kernel is known beforehand, the problem becomes one of nonblind deconvolution. However, when the blur kernel is unknown—common in dynamic underwater settings—it becomes a blind deconvolution problem. This problem is inherently ill-posed because multiple combinations of sharp images and blur kernels can explain the same blurred image. To resolve this ambiguity, assumptions or previous knowledge about the blur kernel or image structure typically are introduced. Traditional blind deconvolution methods often use iterative algorithms, refining the blur kernel and sharp image estimates over multiple steps. These classical approaches, however, often suffer from artifacts such as ringing and struggling with large motions or noise. To address these limitations, researchers have explored regularization techniques that incorporate priors to constrain the solution space. For example, sparsity priors in the temporal domain encourage smooth results, while Bayesian frameworks provide probabilistic formulations of the problem. One promising direction involves leveraging domain-specific priors for underwater imagery. By incorporating this property into optimization frameworks, researchers have developed more-effective and -robust blind deblurring methods tailored to underwater scenarios.

In recent years, deep neural network-based restoration methods have achieved state-of-the-art results in various image deblurring tasks. However, one major drawback of deep learning-based deblurring networks is that large amounts of blurry-clean image pairs are required for training to achieve good performance. Moreover, deep networks often fail to perform well when the blurry images and the blur kernels during testing are very different from the ones used during training. This happens mainly because of the overfitting of the network parameters on the training data. In this work, we present a method that addresses these issues. We view the non-blind image deblurring problem as a denoising problem. To do so, we perform Wiener filtering on a pair of blurry images with the corresponding blur kernels. This results in a pair of images with colored noise. Hence, the deblurring problem is translated into a denoising problem. We then solve the denoising problem without using explicit clean target images. Extensive experiments are conducted to show that our method achieves results that are on par to the state-of-the-art non-blind deblurring works.

Image blurs are a major source of degradation in an imaging system. There are various blur types, such as motion blur and defocus blur, which reduce image quality significantly. Therefore, it is essential to develop methods for recovering approximated latent images from blurry ones to increase the performance of the imaging system. In this paper, an image blur removal technique based on sparse optimization is proposed. Most existing methods use different image priors to estimate the blur kernel but are unable to fully exploit local image information. The proposed method adopts an image prior based on nonzero measurement in the image gradient domain and introduces an analytical solution, which converges quickly without additional searching iterations during the optimization. First, a blur kernel is accurately estimated from a single input image with an alternating scheme and a half-quadratic optimization algorithm. Subsequently, the latent sharp image is revealed by a non-blind deconvolution algorithm with the hyper-Laplacian distribution-based priors. Additionally, we analyze and discuss its solutions for different prior parameters. According to the tests we conducted, our method outperforms similar methods and could be suitable for dealing with image blurs in real-life applications.

Solving blind image deblurring usually requires defining a data fitting function and image priors. While existing algorithms mainly focus on developing image priors for blur kernel estimation and non-blind deconvolution, only a few methods consider the effect of data fitting functions. In contrast to the state-of-the-art methods that use a single or a fixed data fitting term, we propose a data-driven approach to learn effective data fitting functions from a large set of motion blurred images with the associated ground truth blur kernels. The learned data fitting function facilitates estimating accurate blur kernels for generic scenes and domain-specific problems with corresponding image priors. In addition, we extend the learning approach for data fitting function to latent image restoration and nonuniform deblurring. Extensive experiments on challenging motion blurred images demonstrate the proposed algorithm performs favorably against the state-of-the-art methods.

The problem of application of deep neural networks (DNNs) for extremely small dataset is considered to find the effect of dataset size and quality on performance metrics. Classification task is considered for the set of problems where images of Lyme disease symptoms were collected and structured in the open-access dataset with the different subsets of the whole dataset: clean images only (C), dirty images only (D), and clean and dirty images in the equal proportions (C_D) (https://www.kaggle.com/yoctoman/lyme-clean-and-dirty) . Two different data preprocessing techniques were applied: normalization without data augmentation (NODA) and normalization with data augmentation (DA). The two versions of the modern NASNet DNN architectures of different sizes were used, namely NASNetLarge and NASNetMobile, with the weights obtained after training on ImageNet dataset. These DNNs were trained in two experiments, namely with frozen layers, aka transfer learning (TL), and with unfrozen layers when all weights were updated (NOTL). After fivefold cross-validated training the best results for mean validation metrics were demonstrated by the models trained in C_NOTL_NODA, C_NOTL_DA, and C_TL_NODA experiments for NASNetLarge, and in C_TL_NODA by NASNetMobile. But the testing on the new unknown subset of Lyme disease images with the equal proportion of clean and dirty images gave quite different results. The models trained in the before-mentioned experiments demonstrated poor performance in comparison with the models trained in C_NOTL_DA, D_NOTL_DA experiments, and, especially in D_C_NOTL_DA ones. The successful training strategy should include tuning ImageNet-trained weights without their freezing (NOTL) and intensive data augmentation (DA), but it would not be enough if some diverse relevant background will not be added. That is why the additional methods of “context-relevant data augmentation” (CRDA) mimicked by inclusion of D_C data configuration should be considered. These findings can be used for the more effective selection of training experiments of DNNs in the healthcare context in the view of the small number of available medical data in practice.KeywordsNeural networksDeep learningSmall dataset problemData augmentationTransfer learningHealth careLyme disease

Image motion deblurring with unknown blur kernel is an ill‐posed problem. This paper proposes a blind motion deblurring approach that solves blur kernel and the latent image robustly. For kernel optimization, an edge mask is used as an image prior to improve kernel update, then an edge selection mask is adopted to improve image update. In addition, an alternative iterative method is introduced to perform kernel optimization under a multiscale scheme. Moreover, for image restoration, a total‐variation‐(TV‐) based algorithm is proposed to recover the latent image via nonblind deconvolution. Experimental results demonstrate that our method obtains accurate blur kernel and achieves better deblurring results than previous works.

Most of blur kernel estimation models may fail when the blurred image contains some complex structures or is contaminated by large blur. In this paper, we propose a hybrid order l 0-regularized blur kernel estimation model for solving the problem. Firstly, we regularize the latent image in a hybrid order case involving both first-order and second-order regularization term, in which l 0 sparse prior is introduced. Secondly, we introduce an improved adaptive adjustment factor into the model for removing detrimental structures and obtaining more useful information. Finally, we develop an efficient optimization algorithm based on the half-quadratic splitting technique to obtain an accurate blur kernel. Extensive experiments results on both synthetic and some challenged real-life images show that proposed model can estimate a more accurate blur kernel and can effectively recover the latent image when it contains complex structures or is contaminated by large blur.