Edge Constraints Research Articles

A Bayesian Approach To Distributed Anomaly Detection In Edge AI Networks

The challenge of anomaly detection is to obtain an accurate understanding of expected behaviour which is intensified when the data are distributed heterogeneously. Transmitting raw data to a central site incurs high communication overhead and raises privacy issues. The concept of Edge AI allows computation to be performed at the edge site allowing for quick decision making in mission critical scenarios such as self-driving cars. A model is learnt locally and its parameters are transmitted and aggregated. However, existing methods of aggregation do not account for variance and heterogeneous distribution of data. They also do not consider edge constraints such as limited computational, memory and communication capabilities of edge devices. In this work, a fully Bayesian approach is employed by means of a Bayesian Random Vector Functional Link AutoEncoder being incorporated with Expectation Propagation for distributed training. Our anomaly detection system operates without any transmission of raw data, is robust under inhomogeneous network densities and under uneven and biased data distributions. It allows for asynchronous updates to converge in a few iterations and is a relatively simple neural network addressing edge constraints without compromising on performance as compared to existing more complex models.

Onset conspiracy in Upper Sorbian

This article has two goals: descriptive and theoretical. On the descriptive side, the article presents a grammar of gliding and epenthesis of Upper Sorbian. The descriptive goal is worthy because Upper Sorbian has a highly complex but regular and productive system of gliding and epenthesis. Upper Sorbian stands out from a typological point of view because it has ten [sic] different strategies to satisfy Onset.On the theoretical side, the question is whether Optimality Theory that has been designed to solve conspiracies can deal with the complexities of Upper Sorbian. The answer is that it cannot unless it is modified to admit derivational levels. A point of interest is that level 1 in Upper Sorbian must be defined as the root level, not as the expected stem level that includes roots and affixes. Further, it is demonstrated that Itô and Mester’s CrispEdgeconstraint makes wrong predictions for Upper Sorbian, so a new constraint, Multi, is postulated. Also, the analysis bears on the issue of positional markednessversuspositional faithfulness and the question of whether Duke of York derivations should be admitted in phonology.

A Generative Image Inpainting Model Based on Edge and Feature Self-Arrangement Constraints.

At present, the image inpainting method based on deep learning has achieved a better inpainting effect than traditional methods, but the inpainting results still have problems such as local structure disorder and blurred texture when the images involving a large defect area are processed. This paper proposes a second-order generative image inpainting model based on edge and feature self-arrangement constraints. The model consists of two parts: edge repair network and image repair network. Based on the self-encoder, the edge repair network generates the edges in the defect area according to the known information of the image and improves the edge repair effect by minimizing the adversarial loss and feature matching loss. The image inpainting network fills the defect area with the edge repair result as a priori condition. On the basis of U-Net, the feature self-arrangement module (FSM) is proposed to reconstruct the coding features of a specific scale, and the reconstructed feature skips to connect the decoding layer of the same scale, and it is fused with the upper layer underlying features for decoding. Meanwhile, guide loss, adversarial loss, and reconstruction loss are introduced to narrow the difference between the repaired image and the original image. The experimental results show that the inpainting results of the proposed model have stronger structural connectivity and clearer textures and the performance of PSNR, SSIM, and mean L1 loss in the Celeba, Facade, and Places2 is better than other inpainting methods, indicating that the algorithm can produce an inpainting effect with highly connected structure, reasonable semantics, and fine details.

Fire resistance of composite shear walls filled with demolished concrete lumps and self-compacting concrete

The recycling and reuse of waste concrete is conducive to promoting the sustainable development of resources and practicing the concept of green development. Demolished concrete lumps (DCLs) and self-compacting concrete (SCC) made from waste concrete take full advantage of their recyclability. In order to fully utilize the advantages of demolished concrete lumps and self-compacting concrete in construction, applying demolished concrete lumps and self-compacting concrete to composite shear wall structures is an effective way to reduce costs and improve industrial production. In this paper, a new composite shear wall filled with demolished concrete lumps and self-compacting concrete is proposed, and its fire performance is investigated. The shear wall combines edge constraints of concrete-filled steel tubes, demolished concrete lumps and self-compacting concrete. In order to facilitate the mass pouring of demolished concrete lumps, a cavity is formed between the steel tubes at the ends and the precast walls on both sides. A double-sided fire tests was conducted on three composite shear wall specimens filled with demolished concrete lumps and self-compacting concrete at a constant axial compression ratio, where the fire-retardant coating and the wall width-to-depth ratio were considered as the main test parameters. The specimens were able to steadily withstand the predetermined axial loads during the fire resistance time of 180 min and showed good fire resistance. The concrete temperature variation trends of the three specimens in the wall section were basically the same. The fire-retardant coating on the concealed column had a significant effect on the section temperature of the steel tube, while increasing the wall width-to-depth ratio had little effect. When the axial compression ratio was 0.19, the specimens were in the expansion stage within 180 min and no compression occurred. The results show that the expansion deformation of the shear wall sprayed with fire-retardant coating on the outer surface of the concealed column is smaller than that of the shear wall without fire-retardant coating, and the axial deformation of the composite shear wall infilled with demolished concrete lumps and self-compacting concrete is not affected by increasing the wall width-to-depth ratio.

Dynamic Programming BN Structure Learning Algorithm Integrating Double Constraints under Small Sample Condition

The Bayesian Network (BN) structure learning algorithm based on dynamic programming can obtain global optimal solutions. However, when the sample cannot fully contain the information of the real structure, especially when the sample size is small, the obtained structure is inaccurate. Therefore, this paper studies the planning mode and connotation of dynamic programming, restricts its process with edge and path constraints, and proposes a dynamic programming BN structure learning algorithm with double constraints under small sample conditions. The algorithm uses double constraints to limit the planning process of dynamic programming and reduces the planning space. Then, it uses double constraints to limit the selection of the optimal parent node to ensure that the optimal structure conforms to prior knowledge. Finally, the integrating prior-knowledge method and the non-integrating prior-knowledge method are simulated and compared. The simulation results verify the effectiveness of the method proposed and prove that the integrating prior knowledge can significantly improve the efficiency and accuracy of BN structure learning.

Seismic Performance of Composite Shear Walls Filled with Demolished Concrete Lumps and Self-Compacting Concrete after Fire

In order to provide a good solution for the treatment and utilization of construction waste, especially waste concrete and to promote the development of green construction to some extent, in this paper, a new composite shear wall filled with demolished concrete lumps (DCLs) and self-compacting concrete (SCC) was proposed, and its seismic performance after fire was investigated. Based on quasi-static tests of four composite shear walls filled with DCLs and SCC, three after a standard fire and one contrastive specimen without fire, the effects of fire exposure time, fire-retardant coating on the edge constraint steel pipe column, and the width-thickness ratio on seismic performance of composite shear walls after fire were studied. The failure patterns, bearing capacity, hysteretic loops, ductility, skeleton curves, rigidity degradation curves, and energy dissipation of shear walls were analyzed and compared. Test results indicate that the bearing capacity of the shear wall after 60 min of standard fire is slightly lower than the contrastive specimen without fire, but specimens still have good seismic performance. The interlayer displacement angle of the shear wall after a fire still meets the requirement of regulations. The fire-retardant coating on the edge of concealed steel pipe column has a limited effect on the seismic performance of specimens after a fire. The horizontal bearing capacity and energy dissipation capacity of shear walls after a fire can obviously be improved by properly increasing the width-thickness ratio.

Thermo-torsional postbuckling of CNT-reinforced composite toroidal shell segments with surrounding elastic media and tangentially restrained edges

An analytical investigation on the nonlinear stability of toroidal shell segment (TSS) made of carbon nanotube (CNT)-reinforced composite, surrounded by an elastic medium, exposed to elevated temperature and subjected to uniform torsion is presented in this paper. The properties of constituents are assumed to be temperature dependent and effective properties of composite are determined using an extended rule of mixture. CNTs are embedded into matrix phase according to functionally graded or uniform distributions. Basic equations in terms of deflection and stress function are established within the framework of classical shell theory including geometrical nonlinearity in von Kármán–Donnell sense and interactive pressure from surrounding elastic medium. Two boundary edges of the shell are assumed to be simply supported and tangentially restrained. Multi-term analytical solutions are assumed and Galerkin method is used to derive expressions of buckling load and nonlinear relation between torsional load and deflection. Parametric studies are carried out to analyze the effects of material and geometry properties, in-plane boundary condition, elevated temperature and surrounding elastic medium on the buckling resistance and postbuckling behavior of torsionally loaded TSS. Novel finding of this study is that tangential edge constraints have no and negative effects on critical torsional loads at room and elevated temperatures, respectively, and profoundly beneficial influences on postbuckling load capacity of TSSs.

Finding Multidimensional Constraint Reachable Paths for Attributed Graphs

A graph acts as a powerful modelling tool to represent complex relationships between objects in the big data era. Given two vertices, vertex and edge constraints, the multidimensional constraint reachable ( MCR) paths problem finds the path between the given vertices that match the user-specified constraints. A significant challenge is to store the graph topology and attribute information while constructing a reachability index. We propose an optimized hashing-based heuristic search technique to address this challenge while solving the multidimensional constraint reachability queries. In the proposed technique, we optimize hashing and recommend an efficient clustering technique based on matrix factorization. We further extend the heuristic search technique to improve the accuracy. We experimentally prove that our proposed techniques are scalable and accurate on real and synthetic datasets. Our proposed extended heuristic search technique is able to achieve an average execution time of 0.17 seconds and 2.55 seconds on MCR true queries with vertex and edge constraints for Robots and Twitter datasets respectively.

Attention Mask R-CNN with edge refinement algorithm for identifying circulating genetically abnormal cells.

Recent studies have suggested that circulating tumor cells with abnormalities in gene copy numbers in mononuclear cell-enriched peripheral blood samples, such as circulating genetically abnormal cells (CACs), can be used as a non-invasive tool to detect patients with benign pulmonary nodules. These cells are identified through fluorescence signals counting by using 4-color fluorescence in situ hybridization (FISH) technology that exhibits high stability, sensitivity, and specificity. When FISH data are analyzed, the overlapping cells and fluorescence noise is a great challenge for identifying of CACs, thereby seriously affecting the efficiency of clinical diagnosis. To address this problem, in this study, we proposed an end-to-end FISH-based method (CACNET) for CAC identification. CACNET achieved nuclear segmentation and counted 4-color staining signals through improved Mask region-based convolutional neural network (R-CNN), followed by cell category (normal cell, deletion cell, gain cell, or CAC) according to pathological criteria. Firstly, the segmentation accuracy of overlapping nuclei was improved by adding an edge constraint head during training. Then, the interference of fluorescence noise was reduced by fusing non-local module to reconstruct the feature extraction network of Mask R-CNN. We trained and tested the proposed model on a dataset comprising 700 frames with 58,083 nuclei. The Accuracy, Sensitivity, and Specificity (overall performance metric for the algorithm) of CAC identification with CACNET were 94.06%, 92.1%, and 99.8%, respectively. Moreover, the developed method exhibited approximately identification speed of approximately 0.22 s per frames. The results showed that the proposed method outperformed the existing CAC identification methods, making it a promising approach for early screening of lung cancer.

Research on Local Counting and Object Detection of Multiscale Crowds in Video Based on Time-Frequency Analysis

Objective. It has become a very difficult task for cameras to complete real-time crowd counting under congestion conditions. Methods. This paper proposes a DRC-ConvLSTM network, which combines a depth-aware model and depth-adaptive Gaussian kernel to extract the spatial-temporal features and depth-level matching of crowd depth space edge constraints in videos, and finally achieves satisfactory crowd density estimation results. The model is trained with weak supervision on a training set of point-labeled images. The design of the detector is to propose a deep adaptive perception network DRD-NET, which can better initialize the size and position of the head detection frame in the image with the help of density map and RGBD-adaptive perception network. Results. The results show that our method achieves the best performance in RGBD dense video crowd counting on five labeled sequence datasets; the MICC dataset, CrowdFlow dataset, FDST dataset, Mall dataset, and UCSD dataset were evaluated to verify its effectiveness. Conclusion. The experimental results show that the proposed DRD-NET model combined with DRC-ConvLSTM outperforms the existing video crowd counting ConvLSTM model, and the effectiveness of the parameters of each part of the model is further proved by ablation experiments.

DEMA: a distance-bounded energy-field minimization algorithm to model and layout biomolecular networks with quantitative features.

SummaryIn biology, graph layout algorithms can reveal comprehensive biological contexts by visually positioning graph nodes in their relevant neighborhoods. A layout software algorithm/engine commonly takes a set of nodes and edges and produces layout coordinates of nodes according to edge constraints. However, current layout engines normally do not consider node, edge or node-set properties during layout and only curate these properties after the layout is created. Here, we propose a new layout algorithm, distance-bounded energy-field minimization algorithm (DEMA), to natively consider various biological factors, i.e., the strength of gene-to-gene association, the gene’s relative contribution weight and the functional groups of genes, to enhance the interpretation of complex network graphs. In DEMA, we introduce a parameterized energy model where nodes are repelled by the network topology and attracted by a few biological factors, i.e., interaction coefficient, effect coefficient and fold change of gene expression. We generalize these factors as gene weights, protein–protein interaction weights, gene-to-gene correlations and the gene set annotations—four parameterized functional properties used in DEMA. Moreover, DEMA considers further attraction/repulsion/grouping coefficient to enable different preferences in generating network views. Applying DEMA, we performed two case studies using genetic data in autism spectrum disorder and Alzheimer’s disease, respectively, for gene candidate discovery. Furthermore, we implement our algorithm as a plugin to Cytoscape, an open-source software platform for visualizing networks; hence, it is convenient. Our software and demo can be freely accessed at http://discovery.informatics.uab.edu/dema.Supplementary information Supplementary data are available at Bioinformatics online.

Interaction of angular velocity and temperature rise in the thermo-inertia bifurcation buckling of FG laminated nanocomposite annular plates

The present study examines the asymmetric stability of a functionally graded (FG) nanocomposite shear deformable annular plate with a constant angular speed in the thermal environment. For the first time, this phenomenon is investigated for a structure augmented with graphene platelets (GPLs). The displacement field in the annular system is estimated depending on the first-order shear deformation plate assumptions called FSDT. Moreover, the von-Kármán premise of geometrically nonlinear relations is utilized to achieve the equilibrium equations and related edge constraints. To recognize the stability relations in a linearized style, the adjacent equilibrium criterion is utilized. Applying a combined solution procedure consisting of the trigonometric expansion along the hoop axis and the generalized differential quadrature along the radius axis (TE-GDQ), the achieved stability equations are solved, and novel numerical results are obtained. An eigenvalue problem is solved to calculate the critical temperature rise for thermal buckling and the required angular velocity for rotating buckling. Based on the various GPL parameters, geometric properties, and boundary conditions, novel numerical findings are developed to illustrate the interplay of thermal and rotational buckling.

A Multiscale Attention-Guided UNet++ with Edge Constraint for Building Extraction from High Spatial Resolution Imagery

Building extraction from high spatial resolution imagery (HSRI) plays an important role in the remotely sensed imagery application fields. However, automatically extracting buildings from HSRI is still a challenging task due to such factors as large size variations of buildings, background complexity, variations in appearance, etc. Especially, it is difficult to extract both crowded small buildings and large buildings with accurate boundaries. To address these challenges, this paper presents an end-to-end encoder–decoder model to automatically extract buildings from HSRI. The designed network, called AEUNet++, is based on UNet++, attention mechanism and multi-task learning. Specifically, the AEUNet++ introduces the UNet++ as the backbone to extract multiscale features. Then, the attention block is used to effectively fuse different-layer feature maps instead of direct concatenation in the output of traditional UNet++, which can assign adaptive weights to different-layer feature maps as their relative importance to enhance the sensitivity of the mode and suppress the background influence of irrelevant features. To further improve the boundary accuracy of the extracted buildings, the boundary geometric information of buildings is integrated into the proposed model by a multi-task loss using a proposed distance class map during training of the network, which simultaneously learns the extraction of buildings and boundaries and only outputs extracted buildings while testing. Two different data sets are utilized for evaluating the performance of AEUNet++. The experimental results indicate that AEUNet++ produces greater accuracy than U-Net and the original UNet++ architectures and, hence, provides an effective method for building extraction from HSRI.

Thermo-torsional buckling and postbuckling of thin FGM cylindrical shells with porosities and tangentially restrained edges

This paper aims to analyze the effects of porosities, tangential constraints of boundary edges and elevated temperature on the buckling and postbuckling behaviors of thin functionally graded cylindrical shells subjected to uniform torsion. The porosities are assumed to exist within functionally graded material (FGM) according to even and uneven distributions. The properties of constituents are assumed to be temperature dependent and effective properties of porous FGM are determined using a modified rule of mixture. Mathematical formulations are established within the framework of classical shell theory taking into account von Kármán–Donnell nonlinearity and elasticity of tangential edge constraints. Multi-term analytical solutions are assumed to satisfy simply supported boundary conditions and Galerkin method is used to determine buckling torsional load and load–deflection relation for postbuckling analysis. Numerical results show that porosities have remarkably deteriorative influences on the torsional buckling resistance and postbuckling load capacities of FGM cylindrical shells. This study also reveals that tangential edge constraints have pronouncedly beneficial and detrimental effects on the postbuckling strength of torsion-loaded porous FGM cylindrical shells at room and elevated temperatures, respectively.

Static and dynamic response analysis of corrugated core sandwich plates under patch loading

Static and dynamic response analysis of sandwich plates having corrugated core under patch loading is presented in this paper. In this work, the problem formulation is made using finite element-based higher shear deformation theory. The influence of various parameters like aspect ratio, corrugated plate thickness, core thickness, corrugation angle, boundary conditions, ratio of face thickness to corrugation plate thickness on transverse deflection, stresses, and dynamic bending responses are revealed. In the present investigation, a MATLAB program is developed with nine nodded seven degrees of freedom per node of element. The obtained results are verified with available results in the literature to check the program’s efficiency. It is found that transverse deflection increases with increase in corrugation angle, and maximum rotation about the ‘x’ and y-axis increases with increase in the area of the plates exposed to patch loading. Edge constraints have a significant effect on transverse deflection and stress.

Edge Constraint and Location Mapping for Liver Tumor Segmentation from Nonenhanced Images.

As there is no contrast enhancement, the liver tumor area in nonenhanced MRI exists with blurred edges and low contrast, which greatly affects the speed and accuracy of liver tumor diagnosis. As a result, precise segmentation of liver tumor from nonenhanced MRI has become an urgent and challenging task. In this paper, we propose an edge constraint and localization mapping segmentation model (ECLMS) to accurately segment liver tumor from nonenhanced MRI. It consists of two parts: localization network and dual-branch segmentation network. We build the localization network, which generates prior coarse masks to provide position mapping for the segmentation network. This part enhances the ability of the model to localize liver tumor in nonenhanced images. We design a dual-branch segmentation network, where the main decoding branch focuses on the feature representation in the core region of the tumor and the edge decoding branch concentrates on capturing the edge information of the tumor. To improve the ability of the model for capturing detailed features, sSE blocks and dense upward connections are introduced into it. We design the bottleneck multiscale module to construct multiscale feature representations using kernels of different sizes while integrating the location mapping of tumor. The ECLMS model is evaluated on a private nonenhanced MRI dataset that comprises 215 different subjects. The model achieves the best Dice coefficient, precision, and accuracy of 90.23%, 92.25%, and 92.39%, correspondingly. The effectiveness of our model is demonstrated by experiment results, and our model reaches superior results in the segmentation task of nonenhanced liver tumor compared to existing segmentation methods.

Buckling Behavior of Thick Porous Functionally Graded Material Toroidal Shell Segments Under External Pressure and Elevated Temperature Including Tangential Edge Restraint

AbstractOwing to mathematical and geometrical complexities, there is an evident lack of stability analyses of thick closed shell structures with porosity. Based on an effective analytical approach, for the first time, simultaneous effects of porosities, elasticity of edge constraint, and surrounding elastic media on the buckling resistance capacity of thick functionally graded material (FGM) toroidal shell segments (TSSs) subjected to external pressure, elevated temperature, and combined thermomechanical loads are investigated in this paper. The volume fractions of constituents are varied across the thickness according to power law functions, and effective properties of the FGM are determined using a modified rule of mixture. The porosities exist in the FGM through even and uneven distributions. Governing equations are based on a higher-order shear deformation theory (HSDT) taking into account interactive pressure from surrounding elastic media. These equations are analytically solved and closed-form expressions of buckling loads are derived adopting the two-term form of deflection along with Galerkin method. Parametric studies indicate that the porosities have beneficial and deteriorative influences on the buckling resistance capacity of thermally loaded and pressure-loaded porous FGM TSSs, respectively. Furthermore, tangential constraints of edges lower the buckling resistance capacity of the shells, especially at elevated temperatures.

Investigations on the forming characteristics of a novel flexible incremental sheet forming method for low-ductility metals at room temperature

• Proposed a novel flexible ISF method without edge constraint on target blank for low-ductility materials at room temperature. • Proposed analytical model of sheet thickness for undesired severe thinning region. • Investigated deformation characteristics and advantages by comparison with other ISF processes. • Fabricated several freeform panels with typical low-ductility materials to evaluate formability and geometric deviation. As important structural freeform panels (FFPs) for medical implants and aerospace system, lightweight thin wall panels with lower ductility are conventionally produced by hot forming process by utilizing the die on the press, which is not efficient or cost-effective for small batch or customized manufacturing. The dieless incremental sheet forming (ISF) provides a promising solution with increased formability, but making sheet metal panels with low-ductility at room temperature is still a challenging work. Targeting to solve this puzzle, a novel ISF without edge constraint on the target blank between the two edge-fixed supporting layers is developed to get flexible and free plastic deformation for the target blank. To reveal the deformation mechanism of this new method, an analytical prediction model of sheet thickness in ISF process is developed to calculate the deformed thickness in severe thinning region, which indicates that the undesired severe thinning region of conventional ISF process is caused by different stress states near the constrained edge, and can be completely eliminated by this proposed method to obtain uniform plastic deformation and improved formability. Different FFPs of magnesium alloy AZ31 and titanium alloy TC4 are fabricated by the novel flexible free ISF (FFISF) to validate the improved formability, and compared with conventional ISF and three-sheet incremental forming. Experimental studies demonstrate that the proposed method provides potentials for flexible fabrication of FFPs with low ductility at room temperature. More importantly, the promising geometric accuracy without trimming can be obtained by this new method, and multiple axisymmetric parts can also be formed within a cycle time.

Influence of Temperature and Moisture on Free Vibration Behavior of Skew Laminated Composite Sandwich Panels with CNTRC Core

This paper presents the influence of temperature and moisture on the free vibration characteristics of skew laminated composite sandwich (SLCS) panels. The face sheets of the panels are made of graphite–epoxy composite, while the core consists of carbon nanotube-reinforced composite. The coupled hygro-elastic and thermo-elastic relations for the SLCS shells/panels are formulated using first-order shear deformation theory. The nonmechanical stiffness matrices are represented by the initial stress stiffness matrix developed using nonlinear strain–displacement relations. The temperature and moisture-dependent material properties are considered to analyze the laminated composite sandwich spherical, hyperbolic, ellipsoid, cylindrical Shells, and flat plates. Several numerical examples are comprehensively studied to establish the influence of temperature, moisture, the volume fraction of carbon nanotubes in the core material, functional gradation types, skew angle, and edge constraints on the vibration responses of SLCS shells. Further exploration is devoted to studying the combined effect of moisture, temperature, and the geometrical parameters such as length to width ratio, length to thickness ratio, radius-to-length ratio, and the core thickness to face sheet thickness ratios on the natural frequency of the skew laminated composite sandwich panels.

Space-time-resolved measurements of the effect of pinned contact line on the dispersion relation of water waves

Surface wave dynamics in small scale configurations can be substantially modified by edge constraints, in particular a pinned contact line, when the force it exerts is non-negligible with respect to gravity and surface tension. This work develops a hybrid approach, which combines a theoretical model with a complete space-time resolved measurement of the surface deformation and static menisci, allowing an accurate estimation of the contribution of a pinned contact line to shift the dispersion relation towards faster phase velocities.