Imaging Problems Research Articles

A deep neural network approach with attention mechanism to improve the quality of target observation for UAVs

Owing to the limitations of onboard equipment, the direct acquisition of high-resolution images in unmanned aerial vehicle (UAV) target observation tasks is a challenge. However, with the development of deep learning, super-resolution reconstruction has become a popular technique for enhancing image resolution. Though, current super-resolution reconstruction methods can enhance UAV target observation images to some extent, most of them are unable to strike a balance between the number of parameters, computational cost, and reconstruction effect. Furthermore, limited research has been conducted for overcoming the cascade coupling problem in UAV target observation images. To solve these problems and provide a lightweight, fast, and high-quality super-resolution method for UAV target observation images, an enhanced attention dense cross network (EADCN) is proposed. This method includes excellent attention mechanisms, such as the double large kernel attention module (D-LKA) and multi-aware channel attention distillation block (MAADB). In addition to benchmark datasets, a UAV target observation dataset (UAV-TOD) is created to evaluate the performance of EADCN. The experimental results demonstrate that the EADCN has fewer parameters, low computational complexity, high accuracy, and fast processing speed, making it suitable for deployment in UAVs.

On mathematical problems of two-coefficient inverse problems of ultrasonic tomography

This paper proves the theorem of uniqueness for the solution of a coefficient inverse problem for the wave equation in (with two unknown coefficients: speed of sound and absorption. The original nonlinear coefficient inverse problem is reduced to an equivalent system of two uniquely solvable linear integral equations of the first kind with respect to the sound speed and absorption coefficients. Estimates are made, substantiating the multistage method for two unknown coefficients. These estimates show that given sufficiently low frequencies and small inhomogeneities, the residual functional for the nonlinear inverse problem approaches a convex one. This solution method for nonlinear coefficient inverse problems is not linked to the limit approach as frequency tends to zero, but assumes solving the inverse problem using sufficiently low, but not zero, frequencies at the first stage. For small inhomogeneities that are typical, for instance, for medical tasks, carrying out real experiments at such frequencies does not present major difficulties. The capabilities of the method are demonstrated on a model inverse problem with unknown sound speed and absorption coefficients. The method effectively solves the nonlinear problem with parameter values typical for tomographic diagnostics of soft tissues in medicine. A resolution of approximately 2 mm was achieved using an average sounding pulse wavelength of 5 mm.

Artifact-suppressing reconstruction of strongly interacting objects in X-ray near-field holography without a spatial support constraint.

The phase problem is a well known ill-posed reconstruction problem of coherent lens-less microscopic imaging, where only the squared magnitude of a complex wavefront is measured by a detector while the phase information of the wave field is lost. To retrieve the lost information, common algorithms rely either on multiple data acquisitions under varying measurement conditions or on the application of strong constraints such as a spatial support. In X-ray near-field holography, however, these methods are rendered impractical in the setting of time sensitive in situ and operando measurements. In this paper, we will forego the spatial support constraint and propose a projected gradient descent (PGD) based reconstruction scheme in combination with proper preprocessing and regularization that significantly reduces artifacts for refractive reconstructions from only a single acquired hologram without a spatial support constraint. We demonstrate the feasibility and robustness of our approach on different data sets obtained at the nano imaging endstation of P05 at PETRA III (DESY, Hamburg) operated by Helmholtz-Zentrum Hereon.

Solution of the Interpretation Tomography Problem in Geophysics under the Linear Integral Representation Method and Theories of Discrete Gravity and the Magnetic Potential

Solution of the Interpretation Tomography Problem in Geophysics under the Linear Integral Representation Method and Theories of Discrete Gravity and the Magnetic Potential

On seismic gradiometric wave equation inversion for density

SUMMARY Material density remains poorly constrained in seismic imaging problems, yet knowledge of density would provide important insight into physical material properties for the interpretation of subsurface structures. We test the sensitivity to subsurface density contrasts of spatial and temporal gradients of seismic ambient noise wavefields, using wave equation inversion (WEI), a form of seismic gradiometry. Synthetic results for 3-D acoustic media suggest that it is possible to estimate relative density structure with WEI by using a full acoustic formulation for wave propagation and gradiometry. We show that imposing a constant density assumption on the medium can be detrimental to subsurface seismic velocity images. By contrast, the full acoustic formulation allows us to estimate density as an additional material parameter, as well as to improve phase velocity estimates. In 3-D elastic media, severe approximations in the governing wave physics are necessary in order to invert for density using only an array of receivers on the Earth's free surface. It is then not straightforward to isolate the comparatively weak density signal from the influence of phase velocity using gradiometric WEI. However, by using receivers both at the surface and in the shallow subsurface we show that it is possible to estimate density using fully elastic volumetric WEI.

An analytical form of ring artifact correction for computed tomography based on directional gradient domain optimization.

Ring artifact is a common problem in Computed Tomography (CT), which can lead to inaccurate diagnoses and treatment plans. It can be caused by various factors such as detector imperfections, anti-scatter grids, or other nonuniform filters placed in the x-ray beam. Physics-based corrections for these x-ray source and detector non-uniformity, in general cannot completely get rid of the ring artifacts. Therefore, there is a need for a robust method that can effectively remove ring artifacts in the image domain while preservingdetails. This study aims to develop an effective method for removing ring artifacts from reconstructed CTimages. The proposed method starts by converting the reconstructed CT image containing ring artifacts into polar coordinates, thereby transforming these artifacts into stripes. Relative Total Variation is used to extract the image's overall structural information. For the efficient restoration of intricate details, we introduce Directional Gradient Domain Optimization (DGDO) and design objective functions that make use of both the image's gradient and its overall structure. Subsequently, we present an efficient analytical algorithm to minimize these objective functions. The image obtained through DGDO is then transformed back into Cartesian coordinates, finalizing the ring artifact correctionprocess. Through a series of synthetic and real-world experiments, we have effectively demonstrated the prowess of our proposed method in the correction of ring artifacts while preserving intricate details in reconstructed CT images. In a direct comparison, our method has exhibited superior visual quality compared to several previous approaches. These results underscore the remarkable potential of our approach for enhancing the overall quality and clinical utility of CTimaging. The proposed method offers an analytical solution for removing ring artifacts from CT images while preserving details. As ring artifacts are a common problem in CT imaging, this method has high practical value in the medical field. The proposed method can improve image quality and reduce the difficulty of disease diagnosis, thereby contributing to better patientcare.

Dual defect detection and generation system in planar surface

The main problem of applying neural networks in the task of defect detection is that there is a limitation in the size of the set of annotated defect images compared to the set of defect-free images. This imbalance in the dataset poses a challenge for the model to effectively learn and generalize patterns associated with defects. The scarcity of annotated defect samples hinders the model's ability to capture the diverse characteristics of anomalies, making it prone to misdetection and reduced performance on real-world data. To solve this problem, the authors have proposed a neural network architecture based on two neural networks U-Net++ and PatchCore trained in the student-teacher paradigm. In this paradigm, the student model undergoes initial training exclusively on defect-free images. Consequently, disparities emerge in the learned features between the student and teacher models, facilitating the computation of an anomaly metric and streamlining the anomaly localization process. The aim of the research is to improve the efficiency of defect detection by creating methods and algorithms, and dataset augmentation systems, as an option to solve the problem of defect-free images. The overarching goal of this research is to enhance the efficacy of defect detection through the development of methodologies, algorithms, and dataset augmentation systems. This serves as a strategic approach to address the inherent challenge posed by the scarcity of annotated defect images in comparison to their defect-free counterparts. By creating methods and augmenting datasets, we aim to mitigate the limitations imposed by the uneven distribution of defect and defect-free samples, ultimately advancing the robustness and performance of defect detection models. A dual system of defect detection based on the U-Net++ pre-trained network and feature estimation of individual fragments based on the PatchCore network is presented. The research incorporates computational experiments designed to assess the effectiveness of the proposed method in comparison to existing counterparts. An in-depth efficiency analysis has been conducted, focusing on dataset enrichment using the generative-adversarial network ResStyleGAN. This examination aims to provide empirical evidence of the method's performance and its potential advantages over established approaches in the context of defect detection. The method developed, centered around the dual system PatchNet++ and preliminary generation of new defects using ResStyleGAN, is positioned for practical application in real production scenarios. Subsequent stages involve a detailed analysis to evaluate its performance and suitability in real-world settings.

Wavelet-Improved Score-Based Generative Model for Medical Imaging.

The score-based generative model (SGM) has demonstrated remarkable performance in addressing challenging under-determined inverse problems in medical imaging. However, acquiring high-quality training datasets for these models remains a formidable task, especially in medical image reconstructions. Prevalent noise perturbations or artifacts in low-dose Computed Tomography (CT) or under-sampled Magnetic Resonance Imaging (MRI) hinder the accurate estimation of data distribution gradients, thereby compromising the overall performance of SGMs when trained with these data. To alleviate this issue, we propose a wavelet-improved denoising technique to cooperate with the SGMs, ensuring effective and stable training. Specifically, the proposed method integrates a wavelet sub-network and the standard SGM sub-network into a unified framework, effectively alleviating inaccurate distribution of the data distribution gradient and enhancing the overall stability. The mutual feedback mechanism between the wavelet sub-network and the SGM sub-network empowers the neural network to learn accurate scores even when handling noisy samples. This combination results in a framework that exhibits superior stability during the learning process, leading to the generation of more precise and reliable reconstructed images. During the reconstruction process, we further enhance the robustness and quality of the reconstructed images by incorporating regularization constraint. Our experiments, which encompass various scenarios of low-dose and sparse-view CT, as well as MRI with varying under-sampling rates and masks, demonstrate the effectiveness of the proposed method by significantly enhanced the quality of the reconstructed images. Especially, our method with noisy training samples achieves comparable results to those obtained using clean data. Our code at https://zenodo.org/record/8266123.

Some Discrete Tomography Problems in Hypergraph Model Interpretation

Abstract In this paper we consider discrete tomography problems with an additional requirement of non-repeatability of rows of the binary matrix to be reconstructed; as well as discrete tomography problems with given pairwise projections. Representing the problems in the hypergraph model and pointing out their equivalence to the basic definitions, we state the following results: (i) nonconvexity of the set of hypergraphic sequences of simple hypergraphs with $$n$$ vertices and $$m$$ hyperedges in the $$n$$-dimensional $$m + 1$$-valued lattice, (ii) characterization of monotone Boolean functions associated with degree sequences of 3-/(n–3)-uniform hypergraphs, (iii) formulation of discrete tomography problems with paired projections, their connection to hypergraph degree sequence problem with generalized degrees, a solution for a particular case.

Quantitative passive imaging by iterative holography: the example of helioseismic holography

In passive imaging, one attempts to reconstruct some coefficients in a wave equation from correlations of observed randomly excited solutions to this wave equation. Many methods proposed for this class of inverse problem so far are only qualitative, e.g. trying to identify the support of a perturbation. Major challenges are the increase in dimensionality when computing correlations from primary data in a preprocessing step, and often very poor pointwise signal-to-noise ratios. In this paper, we propose an approach that addresses both of these challenges: it works only on the primary data while implicitly using the full information contained in the correlation data, and it provides quantitative estimates and convergence by iteration. Our work is motivated by helioseismic holography, a well-established imaging method to map heterogenities and flows in the solar interior. We show that the back-propagation used in classical helioseismic holography can be interpreted as the adjoint of the Fréchet derivative of the operator which maps the properties of the solar interior to the correlation data on the solar surface. The theoretical and numerical framework for passive imaging problems developed in this paper extends helioseismic holography to nonlinear problems and allows for quantitative reconstructions. We present a proof of concept in uniform media.

Mixed-integer linear programming formulations and column generation algorithms for the Minimum Normalized Cuts problem on networks

This paper deals with the k-way normalized cut problem in complex networks. It presents a methodology that uses mathematical optimization to provide mixed-integer linear programming formulations for the problem. The paper also develops a branch-and-price algorithm for the above-mentioned problem which scales better than the compact formulations. Additionally, a heuristic algorithm which is able to approximate large-scale image problems in those cases where the exact methods are not applicable is presented. Extensive computational experiments assess the usefulness of these methods to solve the k-way normalized cut problem. Finally, we have applied the minimum normalized cut objective function to the segmentation of actual images, showing the applicability of the introduced methodology.

The affective and spatiotemporal benefits of podcasting for teaching social policy practice: learning to ‘love’ social policy

ABSTRACT Social policy, or policy practice, in social work education suffers from an image problem among students. Internationally, students prefer direct practice and find policy practice intimidating and boring. In response to provocations to adopt modern technologies, some academics in social work are employing podcasts in teaching and are making claims about the potential to enhance affective connection, demonstrate the relevance of difficult topics, and improve access to education. Yet, there is little student experience data supporting these claims, and there are critical questions about the extent to which this method achieves the profession’s social justice aims or contributes to the commodification of learning. In this paper, I reflexively thematically analyze qualitative insights from students about their experience of a podcast as a central teaching resource in a social policy practice class. I argue that podcasting offers unique affective and spatiotemporal—that is, social space and time—affordances that begin to salvage policy practice’s image problem. Caution is required, however, if this method is to contribute to critical pedagogical practices and not replicate or reinvent barriers to education.

Sparse optimization over polyhedral sets based on weighted ℓ-norm minimization

Finding the sparsest solutions over a polyhedral set is a challenging problem that has received a great deal of attention in the research community. This problem is typically NP-hard owing to the non-convexity and non-continuity of the ℓ 0 -norm. In this study, it is demonstrated that any solution of ℓ 0 -minimization over a polyhedral set must be one of the ‘extreme efficient solutions’ (EESs) of a special multiobjective optimization problem. Moreover, a new weighted ℓ 1 -norm minimization method using the Goldman–Tucker Theorem for linear optimization problems is established. Some essential features of this algorithm are examined, and numerical tests are performed to illustrate the efficiency of the proposed strategy. The numerical performance of the proposed method is compared with conventional reweighted ℓ 1 -algorithms and several compressed sensing solvers. Additionally, the method is applied to address the compressed sensing problem in magnetic resonance imaging (MRI) reconstruction.

Efforts to Improve Student Learning Outcomes on the Diversity of Living Creatures in My Environment With a Problem Based Learning Model for Class IV State Primary School 020273 Binjai Utara

This research aims to improve learning outcomes for Natural Sciences (Science) in class IV of SD Negeri 020273 North Binjai. The low student learning outcomes in Natural Sciences (Science) subjects led to this research being carried out. In the classroom, this research uses problem-based learning (PBL) strategies. The research subjects were fifteen fourth grade students at SD Negeri 020273 North Binjai. The purpose of this research is to provide the results of my use of image media and Problem Based Learning (PBL) methodology to study the diversity of animals that inhabit my area. After giving students a five-question comprehension test, researchers gathered information about their learning goals. Based on the research results, an average of 47% of students in cycle I and 80% of students in cycle II met the KKM. To improve the learning outcomes of class IV Natural Sciences (IPA) students which include the diversity of living creatures around me, this learning completeness value was created. Improving student learning outcomes is the goal of initiatives that apply the problem-based learning (PBL) paradigm. Because scientific research is taught through lectures, researchers often choose this approach.

A Bayesian approach for consistent reconstruction of inclusions

This paper considers a Bayesian approach for inclusion detection in nonlinear inverse problems using two known and popular push-forward prior distributions: the star-shaped and level set prior distributions. We analyze the convergence of the corresponding posterior distributions in a small measurement noise limit. The methodology is general; it works for priors arising from any Hölder continuous transformation of Gaussian random fields and is applicable to a range of inverse problems. The level set and star-shaped prior distributions are examples of push-forward priors under Hölder continuous transformations that take advantage of the structure of inclusion detection problems. We show that the corresponding posterior mean converges to the ground truth in a proper probabilistic sense. Numerical tests on a two-dimensional quantitative photoacoustic tomography problem showcase the approach. The results highlight the convergence properties of the posterior distributions and the ability of the methodology to detect inclusions with sufficiently regular boundaries.

Stop moving: MR motion correction as an opportunity for artificial intelligence.

Subject motion is a long-standing problem of magnetic resonance imaging (MRI), which can seriously deteriorate the image quality. Various prospective and retrospective methods have been proposed for MRI motion correction, among which deep learning approaches have achieved state-of-the-art motion correction performance. This survey paper aims to provide a comprehensive review of deep learning-based MRI motion correction methods. Neural networks used for motion artifacts reduction and motion estimation in the image domain or frequency domain are detailed. Furthermore, besides motion-corrected MRI reconstruction, how estimated motion is applied in other downstream tasks is briefly introduced, aiming to strengthen the interaction between different research areas. Finally, we identify current limitations and point out future directions of deep learning-based MRI motion correction.

A novel generative adversarial network‐based super‐resolution approach for face recognition

AbstractFace recognition is an essential feature required for a range of computer vision applications such as security, attendance systems, emotion detection, airport check‐in, and many others. The super‐resolution of subject images is an important and challenging element in numerous scenarios. At times the images are low resolution and need to be processed through super‐resolution techniques to gain more accurate results. For the problem of image super‐resolution, deep learning‐based face recognition systems have been explored in recent years; however, low‐resolution face recognition remains an arduous task. Generative adversarial network (GAN) based models are a promising approach to address this challenge. However, conventional GAN‐based models may generate images that differ significantly from an original high‐resolution image in the test set to the point that the identity of the target face may be changed. To address this shortcoming, we propose a novel U‐Net style generator architecture, where skip‐connections between the encoder and decoder layer can help in preserving the facial characteristics of the input image in the generated image, thus curbing the generator's ability to generate an entirely new image and training it to generate an image more similar in characteristics to the original image. In addition to statistical metrics like structural similarity index measure and Fréchet inception distance, we compute the pixel‐wise distance between the original and model‐generated images to ascertain that our model generates as close to the original images as possible. While we train the model for 4× super‐resolution (64 × 64 images to 256 × 256), our architecture can also be trained for an arbitrary resizing scale. Finally, the number of faces detected over high‐resolution images generated by our model is shown to be higher than state‐of‐the‐art high‐resolution image creation models for face recognition tasks.

The recognition and analysis of fake images based on deep neural network

The generation of fake images and the process of technology is developing quickly, which has generated great concern about its impact on society. To effectively identify the fake image, this article introduces a deep learning-based identification method. This article uses a convolutional neural network (CNN) neural network to construct a model for machine learning image recognition and lets it learn a dataset of hundreds of real and fake pictures to let it complete tasks that are difficult for humans to complete. the use of deep learning to solve this problem has the following advantages: the results can be observed after data input, which is conservative and fast, and there is no need to manually design rules as Deep learning methods can optimise the loss function to learn the rules, mining potential features of data, with strong representation ability. Using this method, the accuracy result is over 60%. This paper proves a machine can learn and recognise real and fake pictures, which is specifically inspiring that learning-based methods can also solve the challenging problem of images.

Posterior-Variance–Based Error Quantification for Inverse Problems in Imaging

Posterior-Variance–Based Error Quantification for Inverse Problems in Imaging

Progressive Histogram Modification for JPEG Reversible Data Hiding

With the development of social network, a large number of private JPEG images are stored in social cloud platform. Correspondingly, the platform embeds user ID or authentication labels to manage these privacy images, preventing them from being arbitrarily accessed or tampered by illegal persons. However, data embedding in JPEG domain inevitably produces irreversible modifications to DCT coefficients, thus resulting in obvious or even serious distortion in the host JPEG images. To address this problem, this paper proposes an efficient JPEG reversible data hiding (RDH) method by constructing progressive two-dimensional histogram mappings. We firstly design distortion function to calculate the cost of each DCT frequency band, and then sort them to build histogram mapping containing a series of coefficient pairs. Subsequently, a progressive mapping mechanism is introduced to maintain most of AC coefficients unchanged. According to the given capacity, this mechanism can adaptively generate an optimum two-dimensional histogram mapping to embed secret messages. Our scheme can achieve an effective balance among embedding capacity, visual quality of the marked image and file size expansion, while keeping high cost-performance complexity. Extensive experiments demonstrate that our method outperforms existing JPEG RDH schemes in terms of visual quality and file size increment of the marked image, and provides an efficient solution for the confidentiality and security access problem of sensitive private image in cloud environment.