Real-world Applications Research Articles

Decomposed Prototype Learning for Few-Shot Scene Graph Generation

Today's scene graph generation (SGG) models typically require abundant manual annotations to learn new predicate types. Therefore, it is difficult to apply them to real-world applications with massive uncommon predicate categories whose annotations are hard to collect. In this paper, we focus on Few-Shot SGG (FSSGG) , which encourages SGG models to be able to quickly transfer previous knowledge and recognize unseen predicates well with only a few examples. However, current methods for FSSGG are hindered by the high intra-class variance of predicate categories in SGG: On one hand, each predicate category commonly has multiple semantic meanings under different contexts. On the other hand, the visual appearance of relation triplets with the same predicate differs greatly under different subject-object compositions. Such great variance of inputs makes it hard to learn generalizable representation for each predicate category with current few-shot learning (FSL) methods. However, we found that this intra-class variance of predicates is highly related to the composed subjects and objects. To model the intra-class variance of predicates with subject-object context, we propose a novel Decomposed Prototype Learning (DPL) model for FSSGG. Specifically, we first construct a decomposable prototype space to capture diverse semantics and visual patterns of subjects and objects for predicates by decomposing them into multiple prototypes. Afterwards, we integrate these prototypes with different weights to generate query-adaptive predicate representation with more reliable semantics for each query sample. We conduct extensive experiments and compare with various baseline methods to show the effectiveness of our method.

From Wet to Protective: Film Formation in Waterborne Coatings.

The importance of anticorrosive coatings cannot be overstated, as they play a vital role in safeguarding assets and infrastructure across various industries. Within this context, the emergence of waterborne (WB) coatings stands out for their paramount significance, offering a sustainable alternative to traditional solvent-based (SB) coatings and addressing pressing environmental concerns. Despite their eco-friendliness, the complexity of their film formation mechanism and the lack of understanding present challenges in enhancing the performance of waterborne coatings for corrosion protection. These coatings are created by dispersing polymers in water, which then form a protective film on a substrate. The process involves stages like water evaporation, particle packing, and polymer interdiffusion, often leading to films with defects and inferior protection against extremely corrosive environments compared to SB coatings. This Review scrutinizes the interplay of factors affecting film formation in application, including coalescing agents, environmental factors, and application conditions. A comparative analysis between SB and WB coatings is also featured to shed light on the performance gap under harsh conditions. This Review discusses analytical techniques for studying film formation, aiming to guide future research toward improving WB coatings' durability and effectiveness. In compiling this collective wisdom, this Review emphasizes the translation of theoretical understanding into practical knowledge, equipping formulators with actionable insights to optimize WB coatings for real-world application and performance.

Artificial intelligence enhances the management of esophageal squamous cell carcinoma in the precision oncology era

Esophageal squamous cell carcinoma (ESCC) is the most common histological type of esophageal cancer with a poor prognosis. Early diagnosis and prognosis assessment are crucial for improving the survival rate of ESCC patients. With the advancement of artificial intelligence (AI) technology and the proliferation of medical digital information, AI has demonstrated promising sensitivity and accuracy in assisting precise detection, treatment decision-making, and prognosis assessment of ESCC. It has become a unique opportunity to enhance comprehensive clinical management of ESCC in the era of precision oncology. This review examines how AI is applied to the diagnosis, treatment, and prognosis assessment of ESCC in the era of precision oncology, and analyzes the challenges and potential opportunities that AI faces in clinical translation. Through insights into future prospects, it is hoped that this review will contribute to the real-world application of AI in future clinical settings, ultimately alleviating the disease burden caused by ESCC.

Unveiling the heterojunction of Fe3O4@MXene nanocatalyst for ultrafast degradation of antibiotics via non-radical pathway

The transfer of advanced oxidation processes (AOPs) to real-world applications faces several technology challenges involving oxidation performance, catalyst cost, poor reactive oxygen species (ROS) yield, and long-term stability. This work aims to develop a reactive, efficient, and stable MXene-based nanocatalyst, namely Fe3O4@Ti3C2Tx, for ultrafast removal of the selected sulfonamide antibiotics. As revealed by extensive nanocatalyst characterization, the MXene surface not only anchors Fe/Fe3O4 nanoparticles (NPs) but also prevents NP agglomeration. Notably, the conductive interface between MXene and Fe3O4 promotes the Fe(II)/Fe(III) redox cycle and accelerates peroxymonosulfate (PMS) activation. The Fe3O4@Ti3C2Tx demonstrates excellent catalytic activity by achieving complete sulfamethoxazole (SMX) degradation within 6 min and maintained its catalytic activity after 5 continuous cycles. In the Fe3O4@Ti3C2Tx/PMS system, multiple ROS (e.g., •OH, SO4•-, O2•-) are involved but 1O2 was found to be the primary ROS for SMX degradation. The presence of co-exiting inorganic impurities and pH changes may hinder SMX degradation. Finally, elucidating possible degradation pathways of SMX would greatly aid in treating antibiotic-containing wastewater.

Covariance Control for Uncrewed Aircraft Systems Under Correlated Uncertainty

Uncrewed aircraft systems and advanced air mobility are associated with uncertain conditions, such as wind prediction errors, and they may deviate from dedicated airspaces referred to as corridors. However, further studies on these aspects are required. This study focuses on the covariance control for a stochastic discrete-time linear system under correlated uncertainties. A covariance control problem with chance constraints can be formulated as a convex-programming problem. Numerical simulations of uncrewed aircraft systems path planning can be used to compare two control laws: covariance control laws that consider correlated and noncorrelated noise. Numerical simulations were conducted to verify that the covariance control law considering correlated noise generates a trajectory that appropriately satisfies the constraints. Consequently, the effect of considering correlated noise in the covariance control law was clarified. In addition, covariance control has been demonstrated to generate appropriate trajectories for various state constraints. Moreover, numerical simulations indicated that covariance control facilitated the control of uncertainty, and the effectiveness of the covariance control was evaluated. The insights gained from the results of this study can inform enhanced uncertainty management for improving reliability and safety in real-world applications of uncrewed aircraft systems and advanced air mobility.

Explainability of convolutional neural networks for damage diagnosis using transmissibility functions

Structural health monitoring (SHM) is crucial for ensuring the safety and efficiency of engineering systems. Vibration-based signals have been widely employed in numerous studies for damage detection, localization and quantification, thanks to their high sensitivity to structural damages and versatile applicability across various structures. Among these, transmissibility functions (TFs) have emerged as particularly promising because they can be derived from output-only measurements, making them practical for real-world applications. Vibration-based features, including those captured by TFs, provide a wealth of damage-related information, but extracting these features can be challenging with traditional methods and often requires a priori knowledge. To address this challenge, deep learning has gained significant attention for its capacity to handle intricate patterns and extract hidden features. However, deep learning algorithms are often perceived as black-box models due to their complex and heterogeneous layers, which hinder transparency and raise concerns, especially in critical domains. Although numerous studies have highlighted the diagnostic potential of deep learning in SHM with vibration-based data, there has been limited exploration of the rationale behind using deep neural networks (DNNs) for processing TFs. This paper addresses this gap by introducing explainable artificial intelligence (XAI) methods in two distinct TF-based damage diagnostic case studies: a numerical aluminium structural beam and a large-scale experimental steel frame. In the former, damage is simulated as local stiffness reductions, while in the latter, damage is introduced by loosening joint bolts. In both cases, a convolutional neural network (CNN) is used to process raw TF data for damage detection, localization, and quantification in the beam, and for detection and localization in the frame. Subsequently, an XAI method based on the layer-wise relevance propagation (LRP) algorithm is employed to identify the most relevant input features. A pattern analysis is presented to explain recurrent patterns, and a comparison with other XAI methods is proposed for further validation. The results demonstrate that the CNN provides accurate predictions for both case studies, with its focus on damage-sensitive features aligning closely with findings in the existing literature.

Selective and highly sensitive measurement of H2O2 and organic hydroperoxides with PtNP/Poly(Brilliant Green)/SPCE

This research presents a novel electrochemical approach for the selective measurement of hydrogen peroxide and organic hydroperoxides, which is pivotal in many fields. The study details the development of an advanced sensor using a one-pot, one-step synthesis to embed platinum nanoparticles within a 3D-polymeric matrix of poly (brilliant green) on screen-printed carbon electrodes. The modified surfaces were characterized using scanning and transmission electron microscopy, Raman spectroscopy, cyclic voltammetry and electrochemical impedance spectroscopy. The results obtained by amperometry showed that, at 0.1 V, only H2O2 produced an electrochemical signal, while, at higher potential (0.5 V), all the hydroperoxides tested exhibited an electrochemical signal. Sensitivities obtained for H2O2 by flow injection analysis were 431 ± 3 and 465 ± 4 μC mM−1 at 0.1 and 0.5 V, respectively, with detection limits (S/N = 3) 116 and 30 nM, respectively. For organic hydroperoxides, sensitivities ranged from 22.3 to 32.6 μC mM−1 at 0.5 V, and limits of detection from 1.15 to 5.95 μM. Chemometric analysis indicated the sensor can satisfactorily measure H2O2 in the presence of the organic hydroperoxides herein analysed. The proposed sensor showed excellent properties in terms of repeatability, reproducibility and stability, with minimal interference. The reliability of the sensor was verified by measuring hydroperoxides spiked in aqueous extracts from real air quality monitoring filters. These features highlight the suitability of the sensor for hydroperoxide measurement and underscore its reliability as a practical tool for real-world applications.

Sinusoidal PWM Generation for 3 Phase Inverter and RPM Measurement using Hercules TMS570LC43xx Launchpad

This project focuses on implementing a 3 phase Sinusoidal PWM generation using the Hercules TMS570LC43xx Launchpad Development Kit (Launchpad). The primary objective is to generate synchronized Sinusoidal nature PWM signals using the onboard High-End Timer (HET) and Enhanced Pulse Width Modulation (ePWM) module, which can be given to the Inverter for conversion of Direct current (DC) power into Alternating current (AC) power. This conversion is essential in various applications where AC power is required but the power source provides DC power. It is used in Solar photovoltaic (PV) systems and other renewable energy installations. These systems generate power suitable for powering household appliances or feeding into the electrical grid. Itis alsoused in electric vehicles (EVs) to drive the electric motor with variable speed. For verification of the wave nature, we have used an external lowpass filter (LPF) to transform the dynamic PWM signals into sinusoidal waveforms, ensuring compatibility with various applications like Inverters which can be further used in equipment and machinery such as Brushless DC motors, pumps and compressors. With the addon functionality to control the signal’s frequency which will be given to the Inverter to control the speed of the motor. Additionally, the project incorporates RPM measurement of the motor using an optical encoder setup interfaced with the Enhanced Quadrature Encoder Pulse (eQEP) module on the Launchpad. This feature enables accurate measurement of rotational speeds, position and Revolution per minute (RPM), enhancing the functionality of the system in real world applications like the speed of conveyor belts and other automated transport systems. Through successful implementation, this project demonstrates wide control for the Inverter, achieving reliable synchronized 3 phase signals with its variable speed having 120-degree phase shift signals alongside precise RPM measurement. The project highlights the Launchpad’s capabilities in handling complex signal processing tasks essential for modern power electronics applications. Looking forward, this project establishes a foundation for future enhancements and innovations in power electronics.

Development of a 6-degrees-of-freedom measurement system using a robotic total station for a free-running model test

The measurement of a model ship's six degrees of freedom (6-DOF) using sensors such as Marvelmind beacons and robotic total stations (RTS) has previously demonstrated substantial error margins. To address these limitations, this study introduces a refined approach for accurate and reliable 6-DOF measurement through free-running model tests (FRMT). The tests were meticulously designed, with three RTS strategically positioned to meet the specific requirements of the measurement scenario. Given the potential constraints in an FRMT environment, we proposed feasible installation ranges for the RTS and prisms, ensuring the stable measurement of the model ship's 6-DOF. To achieve high-precision motion measurement, even in complex environments, we employed the prism queue algorithm, which identifies and processes overlapping or untracked prisms monitored by the RTS. The proposed methodology was validated through simulations and FRMT conducted in the ocean engineering basin of the Korea Research Institute of Ships and Ocean Engineering and by developing an actual implementation program. Comparative analysis using high-precision navigation equipment, such as fiber optic gyros, demonstrated that the method provided unbiased and stable data. Consequently, the proposed method shows significant promise as a viable alternative for precision navigation in real-world applications.

Encoding human activities using multimodal wearable sensory data

Human Activity Recognition (HAR) is the task to automatically analyze and recognize human body gestures or actions. HAR using time-series multi-modal sensory data is a challenging and important task in the field of machine learning and feature engineering due to its increasing demands in numerous real-world applications such as healthcare, sports and surveillance. Numerous daily wearable devices e.g., smartphones, smartwatches, and smart glasses can be used to collect and analyze the human activities on an unprecedented scale. This paper presents a generic framework to recognize the different human activities using continuous time-series multimodal sensory data of these smart gadgets. The proposed framework follows the channel of Bag-of-Features which consists of four steps: (i) Data acquisition and pre-processing, (ii) codebook computation, (iii) feature encoding, and (iv) classification. Each step in the framework plays a significant role to generate an appropriate feature representation of raw sensory data for efficient activity recognition. In the first step, we employed a simple overlapped-window sampling approach to segment the continuous time-series sensory data to make it suitable for activity recognition. Secondly, we build a codebook using k-means clustering algorithm to group the similar sub-sequences. The center of each group is known as codeword and we assume that it represents a specific movement in the activity sequence. The third step consists of feature encoding which transform the raw sensory data of activity sequence into its respective high-level representation for the classification. Specifically, we presented three reconstruction-based encoding techniques to encode sensory data, namely: Sparse Coding, Local Coordinate Coding, and Locality-constrained Linear Coding. The segmented activity sub-sequences are transformed to high-level representation using these techniques and earlier computed codebook. Finally, the encoded features are classified using a simple Random Forest classifier. The proposed HAR framework using three different encoding techniques is evaluated on three large benchmark datasets: UniMiB-SHAR dataset, MHEALTH dataset and WISDM dataset and their results are compared with most recent state-of-the-art techniques. It outperforms the existing techniques and achieves 98.39%, 99.5%, and 99.4% recognition scores on UniMiB-SHAR dataset, MHEALTH dataset, and WISDM dataset respectively. The excellent recognition results and computational analysis confirms the effectiveness and efficiency of the proposed framework.

In-Depth Study of Ambipolar Charge-Transport Regime through External Triggers and Gas Sensing.

Ambipolar devices are a hot topic in research tables due to their unique advantage in reducing the size of the electrical system and enhancing its efficiency. Here, we report a bilayer heterojunction device constructed using octafluoro-vanadyl-phthalocyanine (VOF8Pc) and lutetium bisphthalocyanine (LuPc2), which exhibits both p- and n-type behaviors under oxidizing (NO2 and O3) and reducing gas (NH3) species depending on the humidity level and temperature variations. The initial polarity of the device is identified as n-type by measuring a current decrease under oxygen exposure. Most interestingly, we were capable of observing the zero state (no response) where both opposite charge carriers fight for the majority to dominate the electrical properties of the device when it goes from n- to p-type or vice versa. The inversion in the nature of the majority charge carriers in this ambipolar device was achieved by optimizing the external trigger. The unique property of controllable polarity inversion in a VOF8Pc/LuPc2-based bilayer heterojunction device makes it the most effective ambipolar device for real-world applications.

Fixed Point Results for Fuzzy Enriched Contraction in Fuzzy Banach Spaces with Applications to Fractals and Dynamic Market Equillibrium

We introduce fuzzy enriched contraction, which extends the classical notion of fuzzy Banach contraction and encompasses specific fuzzy non-expansive mappings. Our investigation establishes both the presence and uniqueness of fixed points considering this broad category of operators using a Krasnoselskij iterative scheme for their approximation. We also show the graphical representation of fuzzy enriched contraction and analyze its graph for different values of beta. The implications of these findings extend to significant results within fuzzy fixed-point theory, enriching the understanding of iterative processes in fuzzy metric spaces. To demonstrate the versatility of our innovative concepts and the associated fixed-point theorems, we provide illustrative examples that showcase their applicability across diverse domains, including the generation of fractals. This demonstrates the relevance of fuzzy enriched contraction to iterated function systems, enabling the study of fractal structures under various contractive conditions. Additionally, we explore practical applications of fuzzy enriched contraction in dynamic market equilibrium, offering new insights into stability and convergence in economic models. Through this unified framework, we open new avenues for both theoretical advancements and real world applications in fuzzy systems.

True pinch mode of magnetorheological fluids

Abstract Magnetorheological (MR) fluids are representatives of smart materials. They react to magnetic fields by developing a yield stress. The effect has been employed in real-world applications such as automotive chassis systems or optical finishing. By convention, MR devices can be operated in at least one of the fundamental modes: flow, shear, squeeze, gradient pinch of which the former has been the least studied and understood. In pinch mode, the material in the flow channel is exposed to non-uniform magnetic fields in the direction parallel to fluid flow. As a result, only the volume of MR fluid near the channel walls are energized to modify the particular material property (yield stress). The result is the channel’s effective diameter change. The behavior of the material in pinch mode is unique and unseen in the other controllable fluids. To study the material’s characteristics in the specific mode, the authors developed a novel circuit concept for energizing the material in an effort to achieve the ’true-zero’ pinch mode magnetic behavior. Contrary to the existing pinch mode valve concepts, the concept valve allows to achieve zero magnetic flux density in the center of the flow channel regardless of the current level. To test the hypothesis a prototype valve was modeled, manufactured and tested across a range of external (flow rate, current/magnetic flux) stimuli. The obtained results yield sufficient evidence proven by results of magnetic simulations to support the underlying hypothesis. The experimental results illustrate the pinch mode type behaviour, i.e. the slope change in the pressure vs flow rate characteristics.

Modelling ℤ-valued time series with Skellam thinning-based INAR(1) process

ABSTRACT Non-stationary count time series, which are small in value and show a trend having relatively large fluctuation, are commonly encountered in real-world applications. Differencing is a commonly employed method to transform a non-stationary time series into a stationary one. However, the differenced count time series is defined on Z and cannot be fitted by the mainstream count time series models, which are defined on N 0 . Although some integer-valued autoregressive (INAR) models based on the signed thinning operator have been proposed for fitting Z -valued time series, these models still encounter difficulties in terms of data generation mechanism, model selection, and parameter estimation. These difficulties may pose challenges for practitioners in practical applications. Therefore, this article utilizes the reparameterized version of the Skellam (Poisson difference) distribution to construct a new thinning operator, thereby introducing a class of Z -valued time series models. Strict stationarity, ergodicity, statistical properties, and parameter estimation methods of the processes are detailedly discussed. An application to the monthly counts of newly listed companies illustrates that the new model is superior in data analysis of finance and economics.

Sequential Task Allocation of More Scalable Artificial Dragonfly Swarms Considering Dubins Trajectory

With the rapid advancement of UAV technology and the increasing complexity of tasks, multi-UAV systems face growing challenges in task execution. Traditional task allocation algorithms often perform poorly when dealing with issues such as local optima, slow convergence speed, and low convergence accuracy, making it difficult to meet the demands for efficiency and practicality in real-world applications. To address these problems, this paper focuses on collaborative task allocation technology for multi-UAV. It proposes a collaborative task allocation strategy for multi-UAV in a multi-target environment, which comprehensively considers various complex constraints in practical application scenarios. The strategy utilizes Dubins curves for trajectory planning and constructs a multi-UAV collaborative task allocation model, with targets including the shortest total distance index, the minimum time index, and the trajectory coordination index. Each UAV is set as an artificial dragonfly by modifying the traditional dragonfly algorithm, incorporating differential evolution algorithms and their crossover, mutation, and selection operations to bring UAV swarms closer to the characteristics of biological dragonflies. The modifications can enhance the global scalability of artificial dragonfly swarms (ADSs), including wider search capacity, wider speed range, and more diverse search accuracy. Meanwhile, potential solutions with global convergence properties are stored to better support real-time adjustments to task allocation. The simulation results show that the proposed strategy can generate a conflict-free task execution scheme and plan the trajectory, which has advantages in changing the data scale of the UAV and the target and improves the reliability of the system to a certain extent.

Integrating ChatGPT with Salesforce for Real-Time Market Insights on Accounts

This paper explores the integration of ChatGPT, an AI-driven natural language processing model, with Salesforce to enhance sales teams' access to real-time market information about specific accounts. By integrating ChatGPT with Salesforce through API calls, users can fetch the latest market trends, news articles, stock prices, financial data, legal filings, and acquisitions related to specific accounts directly on the Account detail page. This integration not only improves decision-making and customer engagement but also offers a scalable, automated solution that saves time and enhances data accuracy. The paper includes coding examples, architecture diagrams, and use cases to demonstrate the implementation, and it discusses key performance indicators (KPIs) for assessing its effectiveness in real-world applications. Keywords: Salesforce, ChatGPT, AI Integration, Real-Time Data, Account Insights, API, Market Trends, CRM, Natural Language Processing, Automation

A Cloud-Based Solution to a Radar Data Deluge

An open-science tool built to support NASA missions is making synthetic aperture radar, once the domain only of subject matter experts, more accessible for nonspecialists and real-world applications.

Optimization and Efficiency of Novel Magnetic-Resin-Based Approaches for Enhanced Nickel Removal from Water

Nickel contamination in water is a critical issue due to its toxicity and persistence. This study presents a novel magnetic resin, developed by modifying Lewatit® MonoPlus TP 207 with magnetite nanoparticles, to enhance adsorption capacity and facilitate efficient separation. A Definitive Screening Design (DSD) was employed to identify and optimize key parameters affecting nickel adsorption, including pH, resin dosage, initial nickel concentration, and the presence of competing ions (calcium and magnesium). The DSD analysis revealed that pH and magnesium concentration were the most significant factors influencing nickel removal. Optimal conditions were determined as pH 7, 270 min contact time, resin dosage of 0.5 mL/L, initial nickel concentration of 110 µg/L, calcium concentration of 275 mg/L, and magnesium concentration of 52.5 mg/L, achieving a maximum removal efficiency of 99.21%. The magnetic resin exhibited enhanced adsorption capacity and faster kinetics compared to the unmodified resin, leading to more efficient nickel removal. Moreover, its magnetic properties facilitated rapid separation from treated water, offering practical advantages for real-world applications. This study demonstrates the effective use of DSD in optimizing adsorption parameters and underscores the potential of magnetic resin as a sustainable and efficient adsorbent for water treatment.

Equity in Education: How Classroom Practices Can Promote Social Justice

This paper investigates how classroom practices can actively promote equity and social justice in education by focusing on three core strategies: inclusive teaching, differentiated instruction, and fostering teacher-student relationships. Using a case study approach, the research examines classrooms in underfunded urban schools where these practices have been implemented. Social justice theory serves as the foundation for this analysis, emphasizing how classroom practices directly address disparities in academic outcomes among students from different socioeconomic backgrounds. Findings show that inclusive teaching practicessuch as using diverse curricular materialsencourage greater participation from marginalized students. Differentiated instruction tailored to individual learning needs reduces performance gaps, while teacher expectations significantly impact student motivation and success. Prior studies are reviewed to contextualize these findings, highlighting a gap in the literature concerning real-world application of these strategies in disadvantaged communities. This research uniquely contributes by providing a concrete analysis of how these practices are applied in practice, offering practical insights for educators and policymakers aiming to reduce inequality in education.

Urban freight distribution with electric vehicles: comparing some solution procedures

The Vehicle Routing Problem (VRP) is a well-known discrete optimization problem that has an impact on theoretical and practical applications. In this paper, a freight distribution model that includes a charging system located at the depot, making it feasible for real world-implementation, is proposed. Two different solution methods are proposed and compared: a genetic algorithm (GA) and a population-based simulated annealing (PBSA) with the number of moves increasing during the iterations. Among the variety of algorithm used to solve the VRP, population-based search methods are the most useful, due to the ability to update the memory at each iteration. To demonstrate the practical aspects of the proposed solution a case study is solved using travel time on a real network to evaluate the potentiality for a real-world application.