Explicit Boundaries Research Articles

Spatiotemporally derived agricultural field delineations for species effects assessments and environmental decision support.

Rural landscapes are strongly defined by the spatial distribution of agricultural fields. GIS layers that capture this information have much utility in many decision support contexts, particularly with regards to the intersection of agricultural pesticide use and endangered species habitat. The United States Department of Agriculture's Cropland Data Layer (CDL) is a georeferenced, annual resource that often serves a crucial role in pesticide risk-related decision support applications. However, CDL agriculture timeseries data are not mapped to explicit field boundaries, contributing to increased uncertainty regarding differentiated crop type spatial homogeneity and geographic extent, inherently adding complexity to multi-temporal crop monitoring and analyses efforts. We describe the development and testing of an approach for field delineation based on timeseries information from the 2008-2021 CDL at spatial scales relevant for endangered species risk assessment. We validate and test the approach against quantitative crop information and contextualize the outputs as part of a case study reconstructing past agricultural pesticide exposures to non-target species to demonstrate the utility of the method for ecological risk assessment decision support. The approach resulted in delineated field unit boundaries that effectively incorporated the unmodified CDL crop type generalized spatial distribution patterns; derived metrics closely corresponded with reported crop metrics for landscapes with proportionally significant agriculture use. When modified to reflect areas of mixed/small crop acreages, the method can provide a useful framework for large-scale field delineation of the CDL, which can complement ongoing environmental risk assessment and conservation efforts in agricultural landscapes.

A corpus of Chinese word segmentation agreement

The absence of explicit word boundaries is a distinctive characteristic of Chinese script, setting it apart from most alphabetic scripts, leading to word boundary disagreement among readers. Previous studies have examined how this feature may influence reading performance. However, further investigations are required to generate more ecologically valid and generalizable findings. In order to advance our understanding of the impact of word boundaries in Chinese reading, we introduce the Chinese Word Segmentation Agreement (CWSA) corpus. This corpus consists of 500 sentences, comprising 9813 character tokens and 1590 character types, and provides data on word segmentation agreement at each character position. The data revealed a high level of overall segmentation agreement (92%). However, participants disagreed on the position of word boundaries in 8.96% of the cases. Moreover, about 85% of the sentences contained at least one ambiguous word boundary. The character strings with high levels of disagreement were tentatively classified into three categories, namely the morphosyntactic type (e.g., “反映–了”), modifier–head type (e.g., “科學–教育”), and others (e.g., “大力–支持”). Finally, the agreement scores also significantly influenced reading behaviors, as evidenced by analyses with published eye movement data. Specifically, a high level of disagreement was associated with longer single fixation durations. We discuss the implications of these results and highlight how the CWSA corpus can facilitate future research on word segmentation in Chinese reading.

Research of anomaly detection based on dynamic anomaly detection enhancement framework

Abstract Anomaly detection plays a crucial role in various fields, from industrial defect inspection to geological detection. However, traditional approaches often struggle with insufficient discriminability and an inability to generalize to unseen anomalies. These limitations stem from the practical difficulty in gathering a comprehensive set of anomalies and the tendency to overlook anomalous instances in favor of normal samples. To address these challenges, we propose a novel Dynamic Anomaly Detection Enhancement Framework, integrating three key innovations: (1) SaliencyAug: An adaptive saliency-guided augmentation method that generates realistic pseudo-samples to enhance learning of rare anomalies, improving model generalization. (2) DynAB: A dynamic attention block that achieves effective multi-level feature fusion while minimizing redundant information, enhancing detection accuracy. (3) DualOM: A dual-head optimization module which employs separate heads for normal and anomalous sample learning, creating more explicit and discriminative decision boundaries. Extensive experiments across multiple real-world datasets demonstrate our framework's superior performance in detecting a wide range of anomalies, demonstrating 2.4% improvement over state-of-the-art methods.

Hierarchical bound states in heat transport

Higher-order topological phases in non-Hermitian photonics revolutionize the understanding of wave propagation and modulation, which lead to hierarchical states in open systems. However, intrinsic insulating properties endorsed by the lattice symmetry of photonic crystals fundamentally confine the robust transport only at explicit system boundaries, letting alone the flexible reconfiguration in hierarchical states at arbitrary positions. Here, we report a dynamic topological platform for creating the reconfigurable hierarchical bound states in heat transport systems and observe the robust and nonlocalized higher-order states in both the real- and imaginary-valued bands. Our experiments showcase that the hierarchical features of zero-dimension corner and nontrivial edge modes occur at tailored positions within the system bulk states instead of the explicit system boundaries. Our findings uncover the mechanism of non-localized hierarchical non-trivial topological states and offer distinct paradigms for diffusive transport field management.

Bidirectional consistency with temporal-aware for semi-supervised time series classification

Semi-supervised learning (SSL) has achieved significant success due to its capacity to alleviate annotation dependencies. Most existing SSL methods utilize pseudo-labeling to propagate useful supervised information for training unlabeled data. However, these methods ignore learning temporal representations, making it challenging to obtain a well-separable feature space for modeling explicit class boundaries. In this work, we propose a semi-supervised Time Series classification framework via Bidirectional Consistency with Temporal-aware (TS-BCT), which regularizes the feature space distribution by learning temporal representations through pseudo-label-guided contrastive learning. Specifically, TS-BCT utilizes time-specific augmentation to transform the entire raw time series into two distinct views, avoiding sampling bias. The pseudo-labels for each view, generated through confidence estimation in the feature space, are then employed to propagate class-related information into unlabeled samples. Subsequently, we introduce a temporal-aware contrastive learning module that learns discriminative temporal-invariant representations. Finally, we design a bidirectional consistency strategy by incorporating pseudo-labels from two distinct views into temporal-aware contrastive learning to construct a class-related contrastive pattern. This strategy enables the model to learn well-separated feature spaces, making class boundaries more discriminative. Extensive experimental results on real-world datasets demonstrate the effectiveness of TS-BCT compared to baselines.

On the Role of Freshwater Budgets in the Formation of Salinity Extremes in the Ocean Interior

We propose interpreting the vertical structure of waters not only in space but also over time, taking into account the continuous variability of the ocean and the independence of temperature and salinity parameters. This approach allows us to consider, in real conditions, the entire spectrum of short- and long-term extrema of both characteristics, either separately or together at the same depth. Vertical distributions of temperature and salinity transform separately and independently of each other under the influence of heat and freshwater budgets, which change at different time scales. Each of the parameters has its own active layers with corresponding time scales. Changes in the signs of heat and freshwater budgets and subsequent changes in the characteristics of the surface layer cause the appearance and disappearance of separate extrema of temperature and salinity in the water column. Using a salinity field as an example, we have shown that each of the extrema, in their vertical distribution from the near-surface to the intermediate depths, is a temporary phenomenon with various lifetime scales and is located at the lower boundary of the corresponding active layers. In some areas of the ocean, temperature and salinity extrema exist together at the same depth. Volumes of water with such characteristics and explicit boundaries should be considered water masses.

The role of semantic information in Chinese word segmentation

ABSTRACT Word segmentation is crucial for reading in Chinese, where the absence of explicit word boundaries poses a distinct challenge. Previous studies in Chinese have examined how lexical and sub-lexical variables affect word segmentation. The present study investigated whether higher-level semantic information affects word segmentation using a primed word segmentation task with Overlapping Ambiguous Strings (OAS). An OAS is a three-character string in Chinese (e.g. ABC [in Latin letters]) where the middle character can constitute a word with both the left (word AB) and right (word BC) characters. The OAS was preceded by a semantic or repetition prime (presented for 42, 83, or 200 ms, across participants), priming either AB or BC. The semantic priming effect occurred at the 200-ms Stimulus Onset Asynchrony (SOA), whereas the repetition priming effect occurred at both 83 and 200-ms SOAs. These findings demonstrate that semantic information can affect word segmentation in Chinese within 200 ms.

Word separation in continuous sign language using isolated signs and post-processing

Continuous Sign Language Recognition (CSLR) is a long challenging task in Computer Vision due to the difficulties in detecting the explicit boundaries between the words in a sign sentence. To deal with this challenge, we propose a two-stage model. In the first stage, the predictor model, which includes a combination of CNN, SVD, and LSTM, is trained with the isolated signs. In the second stage, we apply a post-processing algorithm to the Softmax outputs obtained from the first part of the model in order to separate the isolated signs in the continuous signs. While the proposed model is trained on the isolated sign classes with similar frame numbers, it is evaluated on the continuous sign videos with a different frame length per each isolated sign class. Due to the lack of a large dataset, including both the sign sequences and the corresponding isolated signs, two public datasets in Isolated Sign Language Recognition (ISLR), RKS-PERSIANSIGN and ASLLVD, are used for evaluation. Results of the continuous sign videos confirm the efficiency of the proposed model to deal with isolated sign boundaries detection. The intuition behind the proposed post-processing methodology is to improve the recognition accuracy by removing the untrained and repetitive signs using the sliding window approach during the inference phase. This marks the first instance of such a mechanism within this domain. So, we present our methodology as a baseline for research community to enrich the methodology as well as evaluating on the other real data.

Mean Value of Probing Depth and Tooth Mobility of Abutment Teeth in Patients using Removable Partial Denture

Removable partial dentures (RPDs) are a typical treatment for supplanting missing teeth. Notwithstanding, concerns exist in regards to their effect on the health of the abutment teeth supporting the dental replacement. Objective: To assess the impact of RPDs on the periodontal health of abutment teeth. Methods: This cross-sectional study included patients from the Prosthodontics department, Bacha Khan Medical College (Medical Teaching Institute, Mardan) between January 6th, 2020, to June 6th, 2020. Examining profundity (pocket depth among gum and tooth) and tooth versatility were assessed in patients utilizing RPDs. The probing depth was estimated at six focuses around every tooth, and a profundity of 1-3 mm was viewed as typical. Results: The study found no massive contrasts in testing profundity or tooth versatility in view old enough, sort of tooth (front or back), or orientation. Notwithstanding, a genuinely critical affiliation was seen between probing depth and tooth mobility, proposing a possible connection between these two proportions of periodontal wellbeing. Conclusions: This study recommends that while RPD plan itself could not straightforwardly impact explicit periodontal boundaries like probing depth and tooth mobility, keeping up with great oral cleanliness and guaranteeing fitting RPD configuration are pivotal for forestalling expected periodontal issues.

Topology optimization for stationary fluid–structure interaction problems with turbulent flow via sequential integer linear programming and smooth explicit boundaries

Topology optimization methods face serious challenges when applied to structural design with fluid–structure interaction (FSI) loads, especially for high Reynolds fluid flow, i.e., considering turbulence. In this problem, the information at the fluid–structure interface is crucial for the modeling and convergence of the turbulent fluid flow analysis. This paper devises a new explicit boundary method that generates two- and three-dimensional smooth surfaces to be used in topology optimization with binary {0,1} design variables. A phase-field function is obtained after nodal spatial filtering of the design variables. The 0.5 isoline defines a smooth surface to construct the topology. The FSI problem can then be modeled with accurate physics and explicitly defined regions. The Finite Element Method is used to solve the fluid and structural domains. This is the first work to consider a turbulent flow in the fluid–structure topology optimization framework. The fluid flow is solved considering the k−ɛ turbulence model including standard wall functions at the fluid and fluid–structure boundaries. The structure is considered to be linearly elastic. Semi-automatic differentiation is employed to compute sensitivities and the optimization problem is solved via sequential integer linear programming. Results show that the proposed methodology is able to provide structural designs with smooth boundaries considering loads from low and high Reynolds flow.

Equivalent Radius for Well Inflow Calculations at Different Regimes in Reservoir Flow Simulations

A problem of well representation in numerical simulation of fluid flows in porous media is considered. When modelling hydrocarbon reservoirs, underground gas storages and other natural porous media, there’s a large typical difference in scales of wells (about 10 cm) and inter-well distances (from hundreds to thousands meters), so wells aren’t resolved on the grid as explicit boundaries in large-scale reservoir simulations. Instead, wells are represented by sources/sinks. The problem is that the average pressure in a grid block containing the well (well block pressure, WBP) isn’t equal to the well pressure (bottomhole pressure, BHP). D. Peaceman showed that WBP can be interpreted as the pressure at equivalent radius inside the well grid block. He derived the classical formula for the equivalent radius for steady-state (SS) Darcy flow. Though Peaceman showed it could be used for pseudo steady-state (PSS) flow as well, his derivation for PSS was not accurate. In this paper we compute the equivalent radius for PSS and boundary dominated (BD) regimes of Darcy flow and examine the impact of flow regime and distance to external boundary (or inter-well distance) on it. The results are important for accurate well representation in numerical flow simulations of underground reservoirs.

Contesting the conventional wisdom of periodontal risk assessment.

Over the years, several reviews of periodontal risk assessment tools have been published. However, major misunderstandings still prevail in repeated attempts to use these tools for prognostic risk prediction. Here we review the principles of risk prediction and discuss the value and the challenges of using prediction models in periodontology. Most periodontal risk prediction models have not been properly developed according to guidance given for the risk prediction model development. This shortcoming has led to several problems, including the creation of arbitrary risk scores. These scores are often labelled as 'high risk' without explicit boundaries or thresholds for the underlying continuous risk estimates of patient-important outcomes. Moreover, it is apparent that prediction models are often misinterpreted as causal models by clinicians and researchers although they cannot be used as such. Additional challenges like the critical assessment of transportability and applicability of these prediction models, as well as their impact on clinical practice and patient outcomes, are not considered in the literature. Nevertheless, these instruments are promoted with claims regarding their ability to deliver more individualized and precise periodontitis treatment and prevention, purportedly resulting in improved patient outcomes. However, people with or without periodontitis deserve proper information about their risk of developing patient-important outcomes such as tooth loss or pain. The primary objective of disseminating such information should not be to emphasize assumed treatment efficacy, hype individualization of care, or promote business interests. Instead, the focus should be on providing individuals with locally validated and regularly updated predictions of specific risks based on readily accessible and valid key predictors (e.g. age and smoking).

The effect of surface nanostructuring of the reinforcing phase on the mechanical properties of a nickel composite

The most important task of modern mechanical engineering is the combination of high strength with a sufficient margin of the material plasticity. The solution to this problem is the use of metal matrix composite materials. One of the ways to strengthen a metal matrix is its reinforcement, i.e. the introduction of structures with high hardness and strength into the matrix. At the same time, Ni-based materials are of particular interest due to their increased heat resistance. However, the introduction of a dispersed reinforcing phase into the matrix leads to embrittlement of the material, which leads to difficulties in processing it. As a result of our research, an approach was proposed to obtain a metal matrix composite material using the surface structuring process and the powder metallurgy method. The developed approach made it possible to obtain a composite material where titanium carbide (TiC) nanostructures of about 2 nm in size are evenly distributed in the Ni matrix bulk. An important feature of the composite being developed is the absence of explicit interface boundaries between the metal matrix and the reinforcing element. This ensures the binding of the matrix and the reinforcing phase into a single whole. The resulting composite effectively resists plastic deformation and stresses. This allows not only to influence the strength properties of the material as effectively as possible, but also to preserve its plasticity.

DRIP: Segmenting individual requirements from software requirement documents

AbstractNumerous academic research projects and industrial tasks related to software engineering require individual requirements as input. Unfortunately, according to our observation, several requirements may be packed in one paragraph without explicit boundaries in specification documents. To understand this problem's prevalence, we performed a preliminary study on the open requirement documents widely used in the academic community over the last 10 years, and found that 26% of them include this phenomenon. Several text segmentation approaches have been reported; however, they tend to identify topically coherent units which may contain more than one requirement. What is more, they do not take the constitutions of semantic units of requirements into consideration. Here we report a two‐phase learning‐based approach named DRIP to segment individual requirements from paragraphs. To be specific, we first propose a Requirement Segmentation Siamese framework, which models the similarity of sentences and their conjunction relations, and then detects the initial boundaries between individual requirements. Then, we optimize the boundaries heuristically based on the semantic completeness validation of the segments. Experiments with 1132 paragraphs and 6826 sentences show that DRIP outperforms the popular unsupervised and supervised text segmentation algorithms with respect to processing different documents (with accuracy gains of 57.65%–187.53%) and processing paragraphs of different complexity (with average accuracy gains of 54.46%–158.68%). We also show the importance of each component of DRIP to the segmentation.

A Dirichlet Process Mixture of Robust Task Models for Scalable Lifelong Reinforcement Learning.

While reinforcement learning (RL) algorithms are achieving state-of-the-art performance in various challenging tasks, they can easily encounter catastrophic forgetting or interference when faced with lifelong streaming information. In this article, we propose a scalable lifelong RL method that dynamically expands the network capacity to accommodate new knowledge while preventing past memories from being perturbed. We use a Dirichlet process mixture to model the nonstationary task distribution, which captures task relatedness by estimating the likelihood of task-to-cluster assignments and clusters the task models in a latent space. We formulate the prior distribution of the mixture as a Chinese restaurant process (CRP) that instantiates new mixture components as needed. The update and expansion of the mixture are governed by the Bayesian nonparametric framework with an expectation maximization (EM) procedure, which dynamically adapts the model complexity without explicit task boundaries or heuristics. Moreover, we use the domain randomization technique to train robust prior parameters for the initialization of each task model in the mixture; thus, the resulting model can better generalize and adapt to unseen tasks. With extensive experiments conducted on robot navigation and locomotion domains, we show that our method successfully facilitates scalable lifelong RL and outperforms relevant existing methods.

On the multi-objective perspective of discrete topology optimization in fluid-structure interaction problems

Fluid-structure interaction is a challenging topic that addresses fluid and solid physics, as well as the stress coupling between them. Traditional topology optimization methods are performed with coupling load interpolation schemes in order to have some information on fluid flow sensitivity analysis and improve compliance minimization into design-dependent problems. In this paper, we propose a strategy to design fluid-structure systems by combining solid and fluid objectives without the need for the coupling load interpolation scheme. Therefore, we investigate topology optimization applied to fluid-structure interaction problems via a multi-objective formulation. We combine structural compliance with some fluid flow objective functions (energy dissipation, downforce, and drag) subject to a volume constraint. We assume linear elasticity for the solid and the steady incompressible Navier-Stokes equations for the fluid. The optimization problem is solved by using sequential integer linear programming via the TOBS (Topology Optimization of Binary Structures) method with a geometry trimming (GT) technique. It is a gradient-based method that produces explicit (discrete) boundaries that are convenient for coupled physics problems. Numerical results are presented for two and three-dimensional problems considering low to moderate Reynolds numbers. We demonstrate that the proposed multi-objective approach yields physically meaningful designs with improved performance, highlighting that the incorporation of coupling load interpolation into fluid-structure interaction optimization is redundant.

A compatible boundary condition-based topology optimization paradigm for static mechanical cloak design

We propose a novel compatible boundary condition-based topology optimization formulation for realizing static mechanical cloaking of linear elastic structures. The key idea is to make the actual boundary conditions along the exterior of the cloaking region consistent with the reference boundary conditions, including the displacement and the nodal force, by designing the cloak structure solely. This compatible boundary condition-based optimization paradigm is not only applicable to various shaped voids (e.g., circle, ellipse, star, Mickey, square, “DUT” logo) and diverse loading conditions (e.g., pressure-sliding, biaxial-stretching, linear-stretching, random-stretching, stretching-sliding) but also demonstrates insensitivity to element quantities and slight variations of external load, ensuring consistent and robust elastostatic cloaking performance. Additionally, adopting the Moving Morphable Voids (MMV) method guarantees the optimized structures have the merits of explicit geometric boundaries and easy fabrication. Numerical examples and experimental validation demonstrate the effectiveness of the proposed approach. The present approach can be easily extended for cloak design in three dimensions and multiphysics.

Numerical simulation of shock wave propagation over a dense particle layer using the Baer–Nunziato model

The present study examines the possibility of numerical simulation of a strong shock wave propagating over the surface of a dense layer of particles poured onto an impermeable wall using the Baer–Nunziato two-phase flow model. The setting of the problem follows the full-scale experiment. The mathematical model is based on a two-dimensional system of Baer–Nunziato equations and takes into account intergranular stresses arising in the solid phase of particles. The computational algorithm is based on the Harten–Lax–van Leer–Contact method with a pressure relaxation procedure. The developed algorithm proved to be workable for two-phase problems with explicit interfacial boundaries and strong shock waves. These issues are typical of problems arising from the interaction of a shock wave with a bed or a layer of particles. A comparison with the simulations and full-scale experiments of other authors is carried out. A reasonable agreement with the experiment is obtained for the angles of the transmitted compaction wave and granular contact, including their dependency on the intensity of the propagating shock wave. The granular contact angle increases with the incident shock wave Mach number, while the transmitted compaction wave angle decreases. An explanation is given of the phenomenon of the decrease in thickness of the compacted region in the layer with the increase in intensity of the propagating shock wave. The main reason is that the maximal value of the particle volume fraction in the plug of compacted particles in the layer rises with the increase in shock wave intensity.

Multi-scale infrared and visible image fusion framework based on dual partial differential equations

Image fusion aims to integrate multiple images that contain complementary information into a single image using an appropriate strategy. The fusion of infrared (IR) and visible (VIS) images can effectively integrate thermal radiation and texture information, which fuses foreground and background information into a fused image. An effective image fusion framework should transfer most of the information from the IR and VIS images to the fused image while minimizing artifacts. This paper introduces a novel multi-scale decomposition framework named Dual PDEs, which decomposes the source images into multiple layers using two different types of partial differential equations (PDEs) and fuses the feature representations at each scale layer separately with an adaptive fusion scheme. The target image is obtained from the multi-scale fused layers. Qualitative and quantitative evaluations show that the proposed framework effectively highlights the prominent areas with fewer artifacts, more explicit boundaries, and better visual effects, which outperforms existing fusion methods.

Quantitative Evaluation Methodology for Chassis-Domain Dynamics Performance of Automated Vehicles.

Thorough performance evaluation of automated vehicles (AVs) is an essential prerequisite for AVs' release and deployment. The challenges posed by dynamics performance appraisal of AVs are centered around the complexity of chassis dynamics, performance diversity, and lack of unified quantitative metrics. Therefore, this article proposes a novel quantitative evaluation metric for AVs' chassis-domain performance. We reveal mathematically explicit chassis steady boundaries of various vehicle maneuvers based on the modeling of chassis-domain dynamics and vehicle spatiotemporal signal analysis for safety-critical AVs. By defining and analyzing the multiperformance appraisal problem, this article gives mathematically prerequisites for evaluation metrics. Then, a rigorous metric is developed to quantify AVs' safety and comfort performance comprehensively. Wherein, the steady boundaries are leveraged to the metric normalization. We demonstrate the effectiveness of the proposed quantitative evaluation methodology in various scenarios. Test results illustrate that the proposed method provides a quantitative way to test AVs' integrated dynamics performance.