Search

Document Layout Analysis (DLA) is the process of automatically identifying and categorizing the structural components (e.g. Text, Figure, Table, etc.) within a document to extract meaningful content and establish the page’s layout structure. It is a crucial stage in document parsing, contributing to their comprehension. However, traditional DLA approaches often demand a significant volume of labeled training data, and the labor-intensive task of generating high-quality annotated training data poses a substantial challenge. In order to address this challenge, we proposed a semi-supervised setting that aims to perform learning on limited annotated categories by eliminating exhaustive and expensive mask annotations. The proposed setting is expected to be generalizable to novel categories as it learns the underlying positional information through a support set and class information through Co-Occurrence that can be generalized from annotated categories to novel categories. Here, we first extract features from the input image and support set with a shared multi-scale feature acquisition backbone. Then, the extracted feature representation is fed to the transformer encoder as a query. Later on, we utilize a semantic embedding network before the decoder to capture the underlying semantic relationships and similarities between different instances, enabling the model to make accurate predictions or classifications with only a limited amount of labeled data. Extensive experimentation on competitive benchmarks like PRIMA, DocLayNet, and Historical Japanese (HJ) demonstrate that this generalized setup obtains significant performance compared to the conventional supervised approach.

Advanced observational astronomy suggests that the density inside neutron stars is several times higher than the saturation density of nuclear matter. At such high densities, physicists speculate on the presence of quark matter inside the neutron stars. However, obtaining the exact equation of state of quark matter reliably remains elusive. This motivates us to consider a generic MIT bag model for the quark matter equation of state to investigate the internal composition of compact stars in these extreme environments. For the background static and spherically symmetric solutions, we describe quark stars within the framework of Rastall gravity theory. Specifically, we analyze the mass–radius relation and the corresponding stability of quark stars depending on model parameters such as a2, a4, and the Rastall parameter η. We conclude by discussing the predictability of such a model with existing results from different high-precision timing datasets of three-millisecond pulsars and GW events, such as GW170817 associated with a binary neutron star merger.

We present a stable model for quark stars in Rastall-Rainbow (R-R) gravity. The structure of this configuration is obtained by utilizing an interacting quark matter equation of state. The R-R gravity theory is developed as a combination of two distinct theories, namely, the Rastall theory and the gravity's rainbow formalism. Depending on the model parameters (λ¯,η,Σ,Beff), the mass-radius relations are numerically computed for modified Tolman-Oppenheimer-Volkoff (TOV) equations with proper boundary conditions. The stability of equilibrium configuration has been checked through the static stability criterion, adiabatic index and the sound velocity. Our calculations predict larger maximum masses for quark stars, and the obtained results are compatible with accepted masses and radii values, including constraints from GW190814 and GW170817 events in all the studied cases.

We report a study of thickness-dependent interband and intraband magnetic breakdown by thermoelectric quantum oscillations in ZrSiSe nanoplates. Under high magnetic fields of up to 30 T, quantum oscillations arising from degenerated hole pockets were observed in thick ZrSiSe nanoplates. However, when decreasing the thickness, plentiful multifrequency quantum oscillations originating from hole and electron pockets are captured. These multiple frequencies can be explained by the emergent interband magnetic breakdown enclosing individual hole and electron pockets and intraband magnetic breakdown within spin–orbit coupling (SOC) induced saddle-shaped electron pockets, resulting in the enhanced contribution to thermal transport in thin ZrSiSe nanoplates. These experimental frequencies agree well with theoretical calculations of the intriguing tunneling processes. Our results introduce a new member of magnetic breakdown to the field and open up a dimension for modulating magnetic breakdown, which holds fundamental significance for both low-dimensional topological materials and the physics of magnetic breakdown.

In quest of enhanced machinability, Nitronic-50 has been subjected to dry-turning operation with TiCN/Al2O3/TiCN coated MT-CVD and TiAlN/TiN coated MT-PVD inserts under three cutting speeds (50, 75 and 100 m/min), three feed rates (0.08, 0.1 and 0.12 mm/rev), and three depths of cut (0.1, 0.3, 0.5 mm) using Taguchi's L27 experimental design. Findings indicate that, compared to the highest values obtained with MT-CVD tools, use of MT-PVD tools in conjunction with an optimal parametric setting of low cutting speed-low feed-low depth of cut has resulted in lowering of surface roughness and tangential cutting force by 71.92% and 78.46% respectively. In contrast, reductions of 65.48%, 59.43%, and 75.42% are reported under the same optimized configuration with MT-CVD tools, in relation to the highest values of tool-tip temperature, chip-reduction coefficient, and flank wear depths respectively, obtained with MT-PVD inserts. Apparent coefficient of friction decreased by 46.83% upon using MT-CVD inserts at an optimal parametric configuration of moderate cutting speed, low feed, and low depth of cut, as compared to the highest values with MT-PVD inserts at higher settings. Overall, turning of Nitronic-50 in the specified domain is found to be significantly affected by cutting velocity and depth of cut.

Bioinspired organic waveguides of mesoscopic length scales have established a new paradigm in photonics with possible applications in precision bioimaging, sensing, and diagnostics. However, the waveguides typically have low-quality factors (Q), which limits their applications significantly. We use optical tweezers to trap and configure spherical micro-robots so that they are attached to a single stationary organic waveguide. Fascinatingly, we observe that a microtube which was erstwhile supporting longitudinal cavity modes now supports transverse modes that propagate along its surface, and have a quality factor higher by a factor $\sim$ 7. Attachment of more micro-robots does not increase the Q due to disruptions in the momentum matching conditions that tend to inhibit nearfield evanescent coupling-based excitation of the transverse modes into the robots, and consequently, damping losses tend to dominate the final output. The experimental results were verified with FDTD-based simulations and achieved good agreement, thus establishing the robustness of the coupling mechanism. To the best of our knowledge, such enhancement of the performance of bio-inspired optical resonators using optical engineering is unprecedented, and may improve their capabilities and application space significantly.

At super-saturation densities, quarks in the core of neutron stars (NSs) have been a subject of intense investigation for last few decades. In particular, an intriguing conjecture has emerged suggesting that the true ground state of hadronic matter may be strange quark matter (SQM). In the present manuscript, we study internal structure and the physical properties of quark stars (QSs) in energy-momentum squared gravity (EMSG) with MIT bag model equations of state (EoS). An important feature of this theory is the addition of the term f(TμνTμν)to the Einstein-Hilbert action. This modification in the action lead to changes in the right-hand side of Einstein's field equations (EFE). We obtain mass-radius relations for a spherically symmetric configuration supported by the anisotropic fluid matter distribution and solved the hydrostatic equilibrium equations numerically in the framework of EMSG. In addition, we present and discuss the stability of stellar structure depending on the model parameters.

The quantum Hall effect is one of the exclusive properties displayed by Dirac Fermions in topological insulators, which propagates along the chiral edge state and gives rise to quantized electron transport. However, the quantum Hall effect formed by the nondegenerate Dirac surface states has been elusive so far. Here, we demonstrate the nondegenerate integer quantum Hall effect from the topological surface states in three-dimensional (3D) topological insulator β-Ag2Te nanostructures. Surface-state dominant conductance renders quantum Hall conductance plateaus with a step of e2/h, along with typical thermopower behaviors of two-dimensional (2D) massless Dirac electrons. The 2D nature of the topological surface states is proven by the electrical and thermal transport responses under tilted magnetic fields. Moreover, the degeneracy of the surface states is removed by structure inversion asymmetry (SIA). The evidenced SIA-induced nondegenerate integer quantum Hall effect in low-symmetry β-Ag2Te has implications for both fundamental study and the realization of topological magneto-electric effects.

Due to the high density of neutron stars, there is still an open question on the complete structure of the inner core of compact object. It is a great challenge to propose a model describing the highly-dense matter of neutron stars since there are neither experimentally nor observationally adequate results. In theoretical physics, Quantum Chromodynamics (QCD) predicts that at high densities and low temperatures the inner core of neutron stars could be made of exotic quark matter in the color–flavor-locked (CFL) phase of color superconductivity. Regarding the equation-of-state, we predict the existence of electrically charged strange quark stars in the background of the modified theory of gravity f(R,T). In our discussion, we assume a linear relation between charge density and the fluid energy density, and also utilize the f(R,T)=R+2βTmodel to numerically compute the generalized Tolman–Oppenheimer–Volkoff (TOV) structure equations. In order to compute the properties of quark stars, we discuss the mass–radius profile, the effect of the charge in the stellar interior, factor of compactness, and mass-central energy density relation for stellar stability. Finally, we provide a set of basic differences between the standard Einstein–Maxwell gravity and the modified f(R,T)gravity theory.

The formation of continuous patterns of nanostructured materials using directed self-assembly under external fields has generated considerable current research interest. We demonstrate for the first time such continuous patterning by inducing irreversible self-assembly leading to nucleation in mesocopic materials (inorganic, organic, and nanoparticles) using a tightly focused laser beam in an optical tweezers apparatus. A dense aqueous dispersion or solution of the material which has high absorption at the laser wavelength is taken in a sample holder so that some material is adsorbed on the top surface. A hot spot is created on the top surface when the adsorbed material absorbs the high intensity at the focus of the laser beam (a submicrometer sized spot), due to which a water vapor bubble is formed. This causes self-assembly of material around the bubble due to Gibbs–Marangoni convection and capillary flow after which the material eventually nucleates into a crystalline state. The bubble is “trapped” at the hot spot due to the temperature gradient around it and can be manipulated by thermal forces generated optically, so that the system may be described as a “thermo-optical” tweezers. We translate the trapped bubble using the microscope sample holder stage of the apparatus so that the nucleation site of the material is simultaneously translated generating continuous patterns. We have demonstrated the technique using exotic inorganic materials such as soft oxometalates, an organic material such as glycine, and a fluorescent dye such as perylene as well as with carbon nanotubes. We have written patterns over lengths of nearly 1 mm at the rate of 1 Hz, with best resolution of about 4 μm. The technique has potential for a wide range of applications ranging from solution processed printable electronics to controlled catalysis.

This paper investigates the internal structure and equation of state (EoS) of quark stars (QS) with pressure anisotropy, incorporating recent astrophysical observations. Utilizing perturbative QCD corrections and color superconductivity, the EoS is expressed as a dimensionless function dependent on a single parameter, providing a comprehensive exploration of strong interaction effects. By rescaling the EoS and introducing dimensionless variables, the study encompasses scenarios ranging from noninteracting quark matter to extreme stiffness characterized by a parameter, λ̄. Anisotropic QS configurations are constructed using a spherically symmetric metric, with locally anisotropic matter described by a generalized EoS for tangential pressure. The Tolman–Oppenheimer–Volkoff equations governing star structure are presented in dimensionless form, and numerical solutions explore the parameter space defined by effective bag constants and λ̄. Mass–radius relations and the compactness of anisotropic QSs are analyzed under various model parameter variations, considering constraints from pulsar masses and gravitational wave phenomena. The study highlights the influence of pressure anisotropy on the maximum mass of QSs and discusses the impact of model parameters on internal properties. Additionally, the paper considers static stability, adiabatic index, and sound velocity profiles, providing a comprehensive understanding of QS behavior. Overall, this investigation contributes valuable insights into the properties of QSs and their compatibility with observational constraints, offering a detailed exploration of their EoS and internal structure.

A microbubble nucleated due to the absorption of a tightly focused laser at the interface of a liquid–solid substrate enables directed and irreversible self-assembly of mesoscopic particles dispersed in the liquid at the bubble base. This phenomenon has facilitated a new microlithography technique which has grown rapidly over the past decade and can now reliably pattern a vast range of soft materials and colloids, ranging from polymers to metals to proteins. In this review, we discuss the science behind this technology and the present state-of-the-art. Thus, we describe the physics of the self-assembly driven by the bubble, the techniques for generating complex mesoarchitectures, both discrete and continuous, and their properties, and the various applications demonstrated in plastic electronics, site-specific catalysis, and biosensing. Finally, we describe a roadmap for the technique to achieve its potential of successfully patterning “everything” mesoscopic and the challenges that lie therein.

The authors propose a web-based adaptive categorization and authenticated recommendation system, based on teacher performance. Distinct layers of the proposed framework have been operated from many geographically distributed locations. The system contains multiple entry points such as a student attendance module, a teacher categorization module, and a teacher recommendation module, strictly accessed by the administrative authority of an academic organization. The student attendance module is required for achieving better results on the teacher categorization module, and the teacher recommendation module. The reliability factor has been incorporated for realizing the accuracy of the proposed system. The administration authorities communicate with the server to categorize and recommend teachers by using teachers' performance. The replication and re-allocation transparencies have been maintained throughout the servers. Lightweight system performance has been enhanced due to the incorporation of a paperless approach and has provided less data loss. A linear time complexity is achieved due to usage of cellular automata as a tool.

In the present time, it is important to predict superior authors from the huge research community, which determines their performance with their future prospects and opportunities, in the field of scientific research. There is a need to solve the authors’ future prediction problem for equally assessing the performance of the researchers that helps to improve research quality highly, influence other types of research, that will also help to identify the research carrier and other research parameters, that will further influence budding researchers. In our proposed model, to solve the authors’ future prediction problems, five machines learning models i.e., SVM, Logistic Regression, Naive Bayes, Decision Tree, and Random Forest are implemented on a data set of 2750 authors where 1000 authors are the ranked authors. So, we have collected 2750 authors’ data from their bibliographic data sets. Then used feature engineering and feature scaling to build the desired model and prepare the final data set used to build the model. There after varied data sizes and training ratios against the five ML models and measured all the evaluation metrics in different cases. We have seen that the Decision Tree and Random Forest model outperform the other models with an accuracy score of 1.0 and the other scores also performed well.

The DrosophiladNab2 protein is an ortholog of human ZC3H14, a poly(A) RNA binding protein required for intellectual function. dNab2 supports memory and axon projection, but its molecular role in neurons is undefined. Here, we present a network of interactions that links dNab2 to cytoplasmic control of neuronal mRNAs in conjunction with the fragile X protein ortholog dFMRP. dNab2and dfmr1interact genetically in control of neurodevelopment and olfactory memory, and their encoded proteins co-localize in puncta within neuronal processes. dNab2 regulates CaMKII, but not futsch, implying a selective role in control of dFMRP-bound transcripts. Reciprocally, dFMRP and vertebrate FMRP restrict mRNA poly(A) tail length, similar to dNab2/ZC3H14. Parallel studies of murine hippocampal neurons indicate that ZC3H14 is also a cytoplasmic regulator of neuronal mRNAs. Altogether, these findings suggest that dNab2 represses expression of a subset of dFMRP-target mRNAs, which could underlie brain-specific defects in patients lacking ZC3H14.

We trap absorbing micro-particles in air by photophoretic forces generated using a single loosely focused Gaussian trapping beam. We measure a component of the radial Brownian motion of a trapped particle cluster and determine the power spectral density, mean squared displacement, and normalized position and velocity autocorrelation functions to characterize the photophoretic body force in a quantitative fashion for the first time. The trapped particles also undergo spontaneous rotation due to the action of this force. This is evident from the spectral density that displays clear peaks at the rotation and the particles’ inertial resonance frequencies. We fit the spectral density to the well-known analytical function derived from the Langevin equation, measure the resonance and rotation frequencies, and determine the values for particle mass that we verify at different trapping laser powers with reasonable accuracy.

Iris segmentation contains the most significant character in the iris recognition system. An accurate iris segmentation helps to increase the accuracy of iris recognition in any biometric system. However, the robustness and efficiency of conventional iris segmentation methodologies are facing massive challenges in a non-cooperative environment because of unfavorable factors, for instance, blur, occlusion, off-axis, low resolution, motion, and specular reflections. These factors severely affect the accuracy of the iris segmentation approaches. In this article, a novel Iris segmentation approach has been obtained with V-Net architectures to accurately localize the boundaries of the iris image with semantic segmentation mask synthesis. A novel image processing technique: YCrCb and HSV color space, has also been utilized to select saliency point set and recover the iris boundary. A detailed analytical study has been obtained with the U-Net and its modified architecture to understand the drawbacks of the U-Net architectures and how V-Net can overcome them. Experimental Results consolidate that the iris segmentation with V-Net achieves 95.6% mean IOU value, whereas U-Net can only achieve 92.3%. So V-Net can easily outperform the existing state-of-the-art approaches on the challenging UBIRIS database.

We describe a simple yet powerful technique of simultaneously measuring both translational and rotational motion of mesoscopic particles in optical tweezers by measuring the backscattered intensity on a quadrant photodiode (QPD). While the measurement of translational motion by taking the difference of the backscattered intensity incident on adjacent quadrants of a QPD is well known, we demonstrate that rotational motion can be measured very precisely by taking the difference between the diagonal quadrants. The latter measurement eliminates the translational component entirely and leads to a detection sensitivity of around 50 mdeg at S/N of 2 for angular motion of a driven microrod. The technique is also able to resolve the translational and rotational Brownian motion components of the microrod in an unperturbed trap and can be very useful in measuring translation–rotation coupling of micro-objects induced by hydrodynamic interactions.