Structure Tensor Research Articles

Heavy quarkonia in a baryon asymmetric strongly magnetized hot quark matter

Recently there is a resurrection in the study of heavy quark bound states in a hot and baryonless matter with an ambient magnetic field but the matter produced at heavy-ion collider experiments is not perfectly baryonless, so we wish to explore the effect of small baryon asymmetry on the in-medium properties of the heavy quark bound states submerged in a strongly magnetized hot quark matter. Therefore, we have first given a revisit to the general covariant tensor structure of gluon self-energy in the above environment to compute the resummed propagator for gluons. This resummed propagator embodies the properties of medium, which gets translated into the (complex) potential between Q and Q¯ placed in the medium. We observe that the baryon asymmetry makes the real-part of potential slightly more attractive and weakens the imaginary-part. This opposing effects thus lead to the enhancement of binding energies and the reduction of thermal widths of QQ¯ ground states, respectively. Finally, the properties of quarkonia thus deciphered facilitate to compute the dissociation points of J/ψ and ϒ, which are found to have slightly larger values in the presence of baryon asymmetry. For example, J/ψ gets dissociated at 1.64Tc, 1.69Tc, and 1.75Tc, whereas ϒ is dissociated at 1.95Tc, 1.97Tc and 2.00Tc, for μ=0,60 and 100 MeV, respectively. This observation prevents early dissociation of quarkonia in the matter produced at ultrarelativistic heavy ion collisions with a small net baryon number, compared to the ideal baryonless matter.

Fiber enhancement and 3D orientation analysis in label-free two-photon fluorescence microscopy

Fluorescence microscopy can be exploited for evaluating the brain’s fiber architecture with unsurpassed spatial resolution in combination with different tissue preparation and staining protocols. Differently from state-of-the-art polarimetry-based neuroimaging modalities, the quantification of fiber tract orientations from fluorescence microscopy volume images entails the application of specific image processing techniques, such as Fourier or structure tensor analysis. These, however, may lead to unreliable outcomes as they do not isolate myelinated fibers from the surrounding tissue. In this work, we describe a novel image processing pipeline that enables the computation of accurate 3D fiber orientation maps from both grey and white matter regions, exploiting the selective multiscale enhancement of tubular structures of varying diameters provided by a 3D implementation of the Frangi filter. The developed software tool can efficiently generate orientation distribution function maps at arbitrary spatial scales which may support the histological validation of modern diffusion-weighted magnetic resonance imaging tractography. Despite being tested here on two-photon scanning fluorescence microscopy images, acquired from tissue samples treated with a label-free technique enhancing the autofluorescence of myelinated fibers, the presented pipeline was developed to be employed on all types of 3D fluorescence images and fiber staining.

Infrared small target detection based on local-image construction and maximum correntropy

Infrared small target detection has been widely applied in the civil and military fields, but is hard to deal with complex background and highlight edge. To achieve fast and robust target detection, a novel strategy is proposed in this paper. Initially, the local-image method is used to construct a single-frame infrared image into a low-rank sparse matrix with close spatial relevance. Then, the sparse matrix based on maximum correntropy is used to express the small target, which is applied to the improved bilateral random projection method to achieve more robust and faster decomposition for small target. Finally, in order to achieve better convergence result, the prominent edge structure tensor for infrared image is designed and separated from the sparse matrix of target. Extensive experimental results show that the proposed algorithm is obviously superior to other methods in running time and detection rate, especially for highlight cluster.

Cascaded structure tensor for robust baggage threat detection

Cascaded structure tensor for robust baggage threat detection

Hyperspectral image mixed noise removal via tensor robust principal component analysis with tensor-ring decomposition

ABSTRACT Due to the instability of sensors and other factors, hyperspectral images (HSIs) are inevitably polluted by various types of mixed noise. To explore a better denoising method based on the existing research, combining the denoising advantages of tensor-ring (TR) decomposition and tensor robust principal component analysis (TRPCA), a mixed-noise removal method for HSIs is proposed in this paper. First, TRPCA maintains the tensor structure of the image itself, accurately recovers the low-rank part and the sparse part from their sum and separates the sparse noise in the form of sparse tensors. Then, TR decomposition is introduced to denoise the low-rank tensors. To verify the effectiveness and superiority of this method, experiments are carried out on two simulated data sets and two real data sets. Compared with the traditional denoising methods and several existing improved denoising methods from both visual and quantitative aspects, the proposed TRPCA-TR method provides better denoising results.

Mechanical modeling of the maturation process for textile reinforced tissue‐engineered heart valves

AbstractBiohybrid tissue‐engineered implants offer promising possibilities to treat cardiovascular diseases. The ability of tissue‐engineered implants to grow and remodel can be utilized to produce implants that can adapt to changes in the physiological conditions within the patient body. To improve the mechanical properties of the implant, a fibre‐reinforced textile scaffold is used. The embedment of a load‐oriented textile scaffold in the implant acts as biomimetic reinforcement and guides the direction of the extracellular matrix (ECM) growth. Our main objective is to design biohybrid heart valves that can withstand tough physiological loading conditions for decades. That makes it necessary to develop accurate and efficient models to support the design process.In this paper, the focus is on the macro‐mechanical modeling of the maturation process for textile reinforced heart valves. A biohybrid heart valve is modeled as a composite structure. The constitutive model is constructed by defining the total Helmholtz free energy as the sum of energies of the valve constituents. The two main constituents are (i) the extracellular matrix (ECM) and (ii) the fiber‐reinforced textile scaffold. The density of protein fibers such as collagen and elastin fibers are treated as internal variables. We introduce a new approach for modeling the densification of protein fibers during the cultivation process which is capable of considering both static and dynamic cultivation processes. The model considers the density change caused by the immunological response as well as the density change due to mechanical stimulation. The textile scaffold anisotropic behavior is introduced using structural tensors. The constitutive models are then embedded into a special finite element technology with reduced integration that enables efficient computations. The finite element simulation results are then validated using the experimental data provided by BioTex (Institute of Applied Medical Engineering, RWTH Aachen University, Germany). In the end, a finite element simulation for an exemplary valve is presented.

A Novel Method of Efficient Max-min Metric for Classification

Distance metric learning is an important method to study distance metrics that reflect the interaction between features and labels. Because of the high computational complexity and the fact that existing studies on algorithms that measure the similarities with Euclidean distances cannot reflect the real correlations between pairs of samples, learning a suitable distance metric is quite demanding for many data mining tasks. This paper innovatively proposes an extended efficient max-min metric (EMM) that maximizes the total distance between different pairs and minimizes the total distance between similar pairs as much as possible. Simultaneously, the adoption of the local preserving projection framework changes the solution process of the algorithm and improves the speed of the algorithm without losing accuracy. Because traditional EMM only considers pairwise constraints and ignores sample distribution, this study extends EMM based on sample distribution and successfully solves the multi-manifold problem. In the process of data realization, compared with the vector representation method, the use of high-order tensors will make the image representation more accurate and natural. To maintain the structure of higher-order tensors, a tensor-efficient max-min metric (TEMM) is proposed. In order to prove the accuracy and superiority of the research method in this paper, a large number of experiments have been carried out on image processing. The experimental results show that the method proposed in this paper has a good effect.

Modeling the anisotropic elasto‐plastic material behavior of paper and paperboard at finite deformations

AbstractAccurately modeling the material behavior of paper and paperboard is proven to be highly demanding. Most difficulties arise from the pronounced anisotropic nature of the material. Additionally, differences in the in‐plane and out‐of‐plane behavior as well as non‐isochoric plasticity, which affects the elastic behavior, further deepen the issue at hand. Therefore, an material model is proposed, capable of capturing the anisotropic elasto‐plastic behavior of paper and paperboard. The constitutive framework was derived in a thermodynamically consistent manner, valid for finite deformations, and accounting for the coupling of in‐plane and out‐of‐plane behavior. The influence of the orthotropic anisotropy on the elasto‐plastic material response was introduced through the concept of structural tensors. A slight generalization of the elastic part of the free energy was made in order to incorporate the densification effect. A well known yield criterion was modified in order to account for some inconsistencies. Non‐isotropic hardening was considered by defining a set of coupled internal variables. The model's capabilities were demonstrated by simulating the process of folding and creasing.

UV divergence and tensor reduction

We present an efficient algorithm to decompose the ultraviolet (UV) divergences of Feynman integrals to local divergences and various types of sub-divergences. With some reasonable assumptions the local divergences of Feynman integrals can be uniquely defined in dimensional regularization scheme. By an asymptotic expansion in the hard momenta, the computation of local and sub-divergences is reduced to the computation of local divergences of massless vacuum integrals. In theories with spin $$\le \frac{1}{2}$$ , the beta functions and anomalous dimensions can be extracted directly from the local divergence of integrals. We also propose two methods to reduce the tensor structures which can be used in the computation of local divergence. The first method is based on dimensional shift and is extremely powerful for integrals with loop number $$L\le 3$$ . The second method is based on a PV reduction in a $$d_{\infty }$$ dimension subspace, and it is more suited in four and more loops.

Module intersection and uniform formula for iterative reduction of one-loop integrals

In this paper, we develop an iterative sector-level reduction strategy for Feynman integrals, which bases on module intersection in the Baikov representation and auxiliary vector for tensor structure. Using this strategy we have studied the reduction of general one-loop integrals, i.e., integrals having arbitrary tensor structures and arbitrary power for propagators. Inspired by these studies, a uniform and compact formula that iteratively reduces all one-loop integrals has been written down, where messy polynomials in integration-by-parts (IBP) relations have organized themselves to Gram determinants.

Global Conharmonic Concept Type Nearly Kahler Manifold

The concept of permanence conharmonic type Nearly Kahler manifold conditions are obtained when the Nearly Kahler manifold is a manifold conharmonic constant type. Proved that the local conharmonic constancy of type Nearly Kahler manifold is equivalent to its global conharmonic constancy of type. And also proved that the Nearly Kahler manifold conharmonic constant type is a manifold of constant scalar curvature.The notion of constancy of type Nearly Kahler manifolds was introduced A.Greem ([1], [2], [3]) and has proved very useful in the study of the geometry of nearly kahler manifolds. Next nearly kahler manifolds of constant type considered various mathematicians (see. [1], [3], [4], [5], [6], [7]). Comprehensive description of Nearly Kahler manifolds of constant type was obtained V.F.Kirichenko [6]. In [6] it is shown that Nearly Kahler manifold of pointwise constant type description by the identity , where B - a function on Nearly Kahler manifold, . Wherein local constancy of type Nearly Kahler manifold is equivalent to the local constancy of its type. Moreover, the covariant constancy of structure tensors of the first and second kind it follows immediately that, the constancy of type at only one point Nearly Kahler space implies global consistency of its type.

Two-loop QCD corrections to the V → qoverline{q}g helicity amplitudes with axial-vector couplings

We compute the two-loop corrections to the helicity amplitudes for the coupling of a massive vector boson to a massless quark-antiquark pair and a gluon, accounting for vector and axial-vector couplings of the vector boson and distinguishing isospin non-singlet and singlet contributions. A new four-dimensional basis for the decomposition of the amplitudes into 12 invariant tensor structures is introduced. The associated form factors are then computed up to two loops in QCD using dimensional regularization. After performing renormalization and infrared subtraction, the finite parts of the renormalized non-singlet vector and axial-vector form factors are shown agree with each other, and to reproduce the previously known two-loop amplitudes. The singlet axial-vector amplitude receives a contribution from the axial anomaly from two loops onwards. This amplitude is computed for massless and massive internal quarks. Our results provide the last missing two-loop amplitudes entering the NNLO QCD corrections of vector-boson-plus-jet production at hadron colliders.

Deep Hybrid 2-D-3-D CNN Based on Dual Second-Order Attention With Camera Spectral Sensitivity Prior for Spectral Super-Resolution.

A largely ignored fact in spectral super-resolution (SSR) is that the subsistent mapping methods neglect the auxiliary prior of camera spectral sensitivity (CSS) and only pay attention to wider or deeper network framework design while ignoring to excavate the spatial and spectral dependencies among intermediate layers, hence constraining representational capability of convolutional neural networks (CNNs). To conquer these drawbacks, we propose a novel deep hybrid 2-D-3-D CNN based on dual second-order attention with CSS prior (HSACS), which can excavate sufficient spatial-spectral context information. Specifically, dual second-order attention embedded in the residual block for more powerful spatial-spectral feature representation and relation learning is composed of a brand new trainable 2-D second-order channel attention (SCA) or 3-D second-order band attention (SBA) and a structure tensor attention (STA). Concretely, the band and channel attention modules are developed to adaptively recalibrate the band-wise and interchannel features via employing second-order band or channel feature statistics for more discriminative representations. Besides, the STA is promoted to rebuild the significant high-frequency spatial details for enough spatial feature extraction. Moreover, the CSS is first employed as a superior prior to avoid its effect of SSR quality, on the strength of which the resolved RGB can be calculated naturally through the super-reconstructed hyperspectral image (HSI); then, the final loss consists of the discrepancies of RGB and the HSI as a finer constraint. Experimental results demonstrate the superiority and progressiveness of the presented approach in terms of quantitative metrics and visual effect over SOTA SSR methods.

Cardiac structure discontinuities revealed by ex-vivo microstructural characterization. A focus on the basal inferoseptal left ventricle region

BackgroundWhile the microstructure of the left ventricle (LV) has been largely described, only a few studies investigated the right ventricular insertion point (RVIP). It was accepted that the aggregate cardiomyocytes organization was much more complex due to the intersection of the ventricular cavities but a precise structural characterization in the human heart was lacking even if clinical phenotypes related to right ventricular wall stress or arrhythmia were observed in this region. MethodsMRI-derived anatomical imaging (150 µm3) and diffusion tensor imaging (600 µm3) were performed in large mammalian whole hearts (human: N = 5, sheep: N = 5). Fractional anisotropy, aggregate cardiomyocytes orientations and tractography were compared within both species. Aggregate cardiomyocytes orientation on one ex-vivo sheep whole heart was then computed using structure tensor imaging (STI) from 21 µm isotropic acquisition acquired with micro computed tomography (MicroCT) imaging. Macroscopic and histological examination were performed. Lastly, experimental cardiomyocytes orientation distribution was then compared to the usual rule-based model using electrophysiological (EP) modeling. Electrical activity was modeled with the monodomain formulation. ResultsThe RVIP at the level of the inferior ventricular septum presented a unique arrangement of aggregate cardiomyocytes. An abrupt, mid-myocardial change in cardiomyocytes orientation was observed, delimiting a triangle-shaped region, present in both sheep and human hearts. FA's histogram distribution (mean ± std: 0.29 ± 0.06) of the identified region as well as the main dimension (22.2 mm ± 5.6 mm) was found homogeneous across samples and species. Averaged volume is 0.34 cm3 ± 0.15 cm3. Both local activation time (LAT) and morphology of pseudo-ECGs were strongly impacted with delayed LAT and change in peak-to-peak amplitude in the simulated wedge model. ConclusionThe study was the first to describe the 3D cardiomyocytes architecture of the basal inferoseptal left ventricle region in human hearts and identify the presence of a well-organized aggregate cardiomyocytes arrangement and cardiac structural discontinuities. The results might offer a better appreciation of clinical phenotypes like RVIP-late gadolinium enhancement or uncommon idiopathic ventricular arrhythmias (VA) originating from this region.

Toward Compact Transformers for End-to-End Object Detection With Decomposed Chain Tensor Structure

DEtection TRansformer (DETR) is a recently proposed method that streamlines the detection pipeline and achieves competitive results against two-stage detectors such as Faster-RCNN. The DETR models get rid of complex anchor generation and post-processing procedures thereby making the detection pipeline more intuitive. However, the numerous redundant parameters in transformers make the computation and storage of the DETR models intensive, which seriously hinder them to be deployed on the resources-constrained devices. In this paper, to obtain a compact end-to-end detection framework, we propose to deeply compress the transformers with low-rank tensor decomposition. The basic idea of our tensor-based compression method is to represent the large-scale weight matrix in one network layer with a chain of low-order matrices. Furthermore, we show that redundant attention heads will hinder the performance of detection transformers. We thus propose a gated multi-head attention (GMHA) module to suppress the redundant attention information by normalizing the attention heads. In GMHA, each attention head has an independent gate to determine the passed attention value, thereby down-weighting the uninformative heads. The accuracy drop of the tensor-compressed DETR models can be mitigated by applying GMHA modules. Lastly, to obtain fully compressed DETR models, a low-bitwidth quantization technique is introduced for further reducing the model storage size. Based on the proposed methods, we can achieve significant parameter and model size reduction while maintaining high detection performance. We conduct extensive experiments on the COCO and PASCAL VOC datasets to validate the effectiveness of our tensor-compressed (tensorized) DETR models. The experimental results on the COCO benchmark show that we can attain <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$3.7\times $ </tex-math></inline-formula> full model compression with <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$482\times $ </tex-math></inline-formula> feed forward network (FFN) parameter reduction and only 0.6 points accuracy drop.

Exploiting Tensor-Based Bayesian Learning for Massive Grant-Free Random Access in LEO Satellite Internet of Things

With the rapid development of Internet of Things (IoT), low earth orbit (LEO) satellite IoT is expected to provide low power, massive connectivity and wide coverage IoT applications. In this context, this paper provides a massive grant-free random access (GF-RA) scheme for LEO satellite IoT. This scheme does not need to change the transceiver, but transforms the received signal to a tensor decomposition form. By exploiting the characteristics of the tensor structure, a Bayesian learning algorithm for joint active device detection and channel estimation during massive GF-RA is designed. Theoretical analysis shows that the proposed algorithm has fast convergence and low complexity. Finally, extensive simulation results confirm its better performance in terms of error probability for active device detection and normalized mean square error for channel estimation over baseline algorithms in LEO satellite IoT. Especially, it is found that the proposed algorithm requires short preamble sequences and support massive connectivity with a low power, which is appealing to LEO satellite IoT.

Mechanical investigations of the peltate leaf of Stephania japonica (Menispermaceae): Experiments and a continuum mechanical material model.

Stephania japonica is a slender climbing plant with peltate, triangular-ovate leaves. Not many research efforts have been devoted to investigate the anatomy and the mechanical properties of this type of leaf shape. In this study, displacement driven tensile tests with three cycles on different displacement levels are performed on petioles, venation and intercostal areas of the Stephania japonica leaves. Furthermore, compression tests in longitudinal direction are performed on petioles. The mechanical experiments are combined with light microscopy and X-ray tomography. The experiments show, that these plant organs and tissues behave in the finite strain range in a viscoelastic manner. Based on the results of the light microscopy and X-ray tomography, the plant tissue can be considered as a matrix material reinforced by fibers. Therefore, a continuum mechanical anisotropic viscoelastic material model at finite deformations is proposed to model such behavior. The anisotropy is specified as the so-called transverse isotropy, where the behavior in the plane perpendicular to the fibers is assumed to be isotropic. The model is obtained by postulating a Helmholtz free energy, which is split additively into an elastic and an inelastic part. Both parts of the energy depend on structural tensors to account for the transversely isotropic material behavior. The evolution equations for the internal variables, e.g. inelastic deformations, are chosen in a physically meaningful way that always fulfills the second law of thermodynamics. The proposed model is calibrated against experimental data, and the material parameters are identified. The model can be used for finite element simulations of this type of leaf shape, which is left open for the future work.

Nonabelian embedding tensors

In this paper, first we introduce the notion of a nonabelian embedding tensor, which is a nonabelian generalization of an embedding tensor. Then, we introduce the notion of a Leibniz–Lie algebra, which is the underlying algebraic structure of a nonabelian embedding tensor, and can also be viewed as a nonabelian generalization of a Leibniz algebra. Next using the derived bracket, we construct a differential graded Lie algebra, whose Maurer–Cartan elements are exactly nonabelian embedding tensors. Consequently, we obtain the differential graded Lie algebra that governs deformations of a nonabelian embedding tensor. Finally, we define the cohomology of a nonabelian embedding tensor and use the second cohomology group to characterize linear deformations.

Symmetry-based identification and enumeration of independent tensor properties in nonlinear and chiral optics.

For many laser-based methods of material characterization and analysis, a tensor formulation of theory is necessary, especially in techniques that exploit nonlinear or chiral optics. The fundamental interactions that underpin such methods offer various levels of approach to theory, but the most rigorous often lead to equations of considerable complexity. To compute the values for individual material parameters frequently demands making assumptions of extreme simplicity, overly dependent on calculational method, yet still providing unsatisfactory results. A pragmatic and entirely rigorous symmetry-based approach to the irreducible tensorial structures circumvents many of these problems, securing reliable results and guiding the pathway to applications. Instead of focusing on individual tensor components, such an approach can rapidly determine the number of linearly independent quantities-and hence the number of operationally different setups necessary for full characterization. By such means, one can directly ascertain how variations of optical polarization and beam geometry can reliably capture the response of any material system. The use of an irreducible tensor method operates independently of any means that might be chosen to calculate material properties. It removes the need for common simplifying assumptions, such as the approximation of tensorial structure by a scalar representation, adoption of a two-state model, or disregarding near-resonance damping. It also obviates any dependence on a choice of simulation package or quantum-calculational software. In this paper, the principles are set down and illustrated by application to experiments of varying degrees of complexity, including interactions of growing significance in the realm of chiral nonlinear optics. Limitations of this approach are also critically assessed.

3D mesh saliency from local spiral hop descriptors

Mesh saliency, the process of detecting visually important regions in 3D meshes, is a significant component in computer graphics, that can be used in various applications such as denoising and simplification. In this paper, we propose a new 3D mesh saliency measure that can identify sharp geometric features in meshes. A local normal-based descriptor is built for each vertex thanks to a spiral path within a 2-hop neighborhood. First, a geometric-based saliency is computed as the mean local alignment between the spiral descriptors within a 1-hop, and weighted by a vertex roughness measure. Second, a spectral-based saliency is computed from the spectral energy of each vertex structure tensor with the gradient defined from the spiral descriptor alignments. The final saliency is then defined as a weighted sum of both. This single-scale saliency can be extended to a multi-scale saliency by decimating the mesh at several scales and averaging back the obtained saliencies after mapping them between decimated meshes. The approach presents competitive results with state-of-the-art.