Search Results (7,020)

LDANet represents an innovative no-reference quality assessment model specifically engineered to evaluate colorized infrared images. This is a crucial task for various applications, and existing methods often fail to capture color-specific distortions. The proposed model distinguishes itself by uniquely combining color feature extraction through latent Dirichlet allocation (LDA) with spatial feature extraction enhanced by multichannel and spatial attention mechanisms. It employs a dual-feature approach that facilitates thorough assessment of both color fidelity and detail preservation in colorized images. The architecture of LDANet encompasses two critical components: an LDA-based color feature extraction module which meticulously analyzes and learns color distribution patterns, and a spatial feature extraction module that leverages an inception network bolstered by attention mechanisms to effectively capture multiscale spatial characteristics. Rigorous experimental validation conducted on a specialized dataset of colorized infrared images demonstrates that LDANet significantly outperforms existing leading no-reference image quality assessment methods. This study reports the effectiveness of integrating color-specific features within a quality assessment framework tailored for infrared image colorization, representing a meaningful advancement in this domain. These findings emphasize the essential role of color feature integration in the evaluation of colorized infrared images, providing a robust tool for optimizing colorization algorithms and enhancing their practical applications. Full article

The development and electrochemical characteristics of ionic liquid crystal elastomers (iLCEs) are described for use as electrolyte components in lithium-ion batteries. The unique combination of elastic and liquid crystal properties in iLCEs grants them robust mechanical attributes and structural ordering. Specifically, the macroscopic alignment of phase-segregated, ordered nanostructures in iLCEs serves as an ion pathway, which can be solidified through photopolymerization to create ion-conductive solid-state polymer lithium batteries (SSPLBs) with high ionic conductivity (1.76 × 10⁻³ S cm⁻¹ at 30 °C), and a high (0.61) transference number. Additionally, the rubbery state ensures good interfacial contact with electrodes that inhibits lithium dendrite formation. Furthermore, in contrast to liquid electrolytes, the iLCE shrinks upon heating, thus preventing any overheating-related explosions. The Li/LiFePO₄ (LFP) cells fabricated using iLCE-based solid electrolytes show excellent cycling stability with a discharge capacity of ~124 mAh g⁻¹ and a coulombic efficiency close to 100%. These results are promising for the practical application of iLCE-based SSPLBs. Full article

The accurate and reliable localization and tracking of Autonomous Underwater Vehicles (AUVs) are essential for the success of various underwater missions, such as environmental monitoring, subsea resource exploration, and military operations. long-baseline acoustic localization (LBL) is a fundamental technique for underwater positioning, but it faces significant challenges in adversarial environments. These challenges include abrupt target maneuvers and intentional signal interference, both of which degrade the performance of traditional localization algorithms. Although particle filter-based Track-Before-Detect (PFTBD) algorithms are effective under normal submarine conditions, they struggle to maintain accuracy in adversarial environments due to their dependence on conventional likelihood calculations. To address this, we propose the BiLSTM-Attention-PFTBD algorithm, which enhances the traditional PFTBD framework by integrating bidirectional Long Short-Term Memory (BiLSTM) networks with multi-head attention mechanisms. This combination enables better feature extraction and adaptation for localizing AUVs in adversarial underwater environments. Simulation results demonstrate that the proposed method outperforms traditional PFTBD algorithms, significantly reducing localization errors and maintaining robust tracking accuracy in adversarial settings. Full article

Small-Structure Camouflaged Object Detection (SSCOD) is a highly promising yet challenging task, as small-structure targets often exhibit weaker features and occupy a significantly smaller proportion of the image compared to normal-sized targets. Such data are not only prevalent in existing benchmark camouflaged object detection datasets but also frequently encountered in real-world scenarios. Although existing camouflaged object detection (COD) methods have significantly improved detection accuracy, research specifically focused on SSCOD remains limited. To further advance the SSCOD task, we propose a detail-preserving multi-scale adaptive network architecture that incorporates the following key components: (1) An adaptive scaling strategy designed to mimic human visual perception when observing blurry targets. (2) An Attentive Atrous Spatial Pyramid Pooling (A2SPP) module, enabling each position in the feature map to autonomously learn the optimal feature scale. (3) A scale integration mechanism, leveraging Haar Wavelet-based Downsampling (HWD) and bilinear upsampling to preserve both contextual and fine-grained details across multiple scales. (4) A Feature Enhancement Module (FEM), specifically tailored to refine feature representations in small-structure detection scenarios. Extensive comparative experiments and ablation studies conducted on three camouflaged object detection datasets, as well as our proposed small-structure test datasets, demonstrated that our framework outperformed existing state-of-the-art (SOTA) methods. Notably, our approach achieved superior performance in detecting small-structured targets, highlighting its effectiveness and robustness in addressing the challenges of SSCOD tasks. Additionally, we conducted polyp segmentation experiments on four datasets, and the results showed that our framework is also well-suited for polyp segmentation, consistently outperforming other recent methods. Full article

Named Entity Recognition (NER) is a fundamental task in Natural Language Processing (NLP) that supports applications such as information retrieval, sentiment analysis, and text summarization. While substantial progress has been made in NER for widely studied languages like English, Arabic presents unique challenges due to its morphological richness, orthographic ambiguity, and the frequent occurrence of nested and overlapping entities. This paper introduces a novel Arabic NER framework that addresses these complexities through architectural innovations. The proposed model incorporates a Hybrid Feature Fusion Layer, which integrates external lexical features using a cross-attention mechanism and a Gated Lexical Unit (GLU) to filter noise, while a Compound Span Representation Layer employs Rotary Positional Encoding (RoPE) and Bidirectional GRUs to enhance the detection of complex entity structures. Additionally, an Enhanced Multi-Label Classification Layer improves the disambiguation of overlapping spans and assigns multiple entity types where applicable. The model is evaluated on three benchmark datasets—ANERcorp, ACE 2005, and a custom biomedical dataset—achieving an F1-score of 93.0% on ANERcorp and 89.6% on ACE 2005, significantly outperforming state-of-the-art methods. A case study further highlights the model’s real-world applicability in handling compound and nested entities with high confidence. By establishing a new benchmark for Arabic NER, this work provides a robust foundation for advancing NLP research in morphologically rich languages. Full article

Background: Detecting intracellular diffusion dynamics with high spatiotemporal resolution is critical for understanding the complex molecular mechanisms that govern viral infection, drug delivery, and sustained receptor signaling within cellular compartments. Although considerable progress has been made, accurately distinguishing between different types of diffusion in three dimensions remains a significant challenge. Methods: This study extends a previously established two-dimensional, machine learning-based diffusional fingerprinting approach into a three-dimensional framework to overcome this limitation. It presents an algorithm that predicts intracellular motion types based on a comprehensive feature set, including custom statistical descriptors and standard Imaris-derived trajectory features, which capture subtle variations in individual trajectories. The approach employs an extended gradient-boosted decision trees classifier trained on an array of synthetic trajectories designed to simulate diffusion behaviors typical of intracellular environments. Results: The machine learning classifier demonstrated a classification accuracy of over 90% on synthetic datasets, effectively capturing and distinguishing complex diffusion patterns. Subsequent validation using an experimental dataset confirmed the robustness of the approach. The incorporation of the Imaris track features streamlined diffusion classification and enhanced adaptability across diverse volumetric imaging modalities. Conclusions: This work advances our ability to classify intracellular diffusion dynamics in three dimensions and provides a method that is well-suited for high-resolution analysis of intracellular receptor trafficking, intracellular transport of pathogenic agents, and drug delivery mechanisms. Full article

The urgency to retrofit buildings for energy efficiency highlights the need for effective financing mechanisms. Energy Performance Contracts (EPCs) present a viable solution by financing building retrofits based on anticipated energy savings. Reliable baseline models are essential to quantifying these savings accurately. EPCs facilitate retrofits by allowing Energy Service Companies (ESCOs) to cover the upfront costs of energy-saving measures, with repayment derived from the cost savings generated by the reduced energy consumption. This performance-based approach demands accurate and reliable baseline models to predict the expected savings. This study introduces a white-box calibration methodology that accurately estimates energy consumption even with limited monitoring data, making it valuable for cases with scarce or incomplete historical data. In addition to addressing data limitations, the research examines scenarios with restricted control parameters, demonstrating that indoor temperature and energy demand are essential to obtaining a robust baseline model. The present work focuses on performing the calibration process through a single-stage approach that operates on EnergyPlus’ Ideal Loads component and the building-envelope parameters simultaneously. The paper demonstrates that it is possible to accurately assess the building’s energy performance and capture its indoor climate while reducing the time and resources required to train the model. This method achieved a Coefficient of Variation of Mean Square Error (CV(RMSE)) of 26.40% and a Normalized Mean Bias Error (NMBE) of −8.49% during training, with stability maintained during the checking period. The resulting calibrated white-box model serves as a powerful tool for EPCs, enabling reliable prediction of energy savings and offering a predictive framework for building management. By incorporating both energy and temperature, the model supports more informed decision-making and proactive energy management, enhancing the overall sustainability and efficiency of building operations. The methodology is limited to air-based HVAC systems and depends on high-resolution data and monitoring infrastructure. Additionally, the methodology was tested on a single demonstration site, and further research is needed to assess its adaptability to diverse building types and HVAC configurations. Full article

An emulsion gel was developed to replace animal fats in meat products while preserving desirable sensory and structural attributes. The gel was prepared by emulsifying pine nut oil and sunflower oil with whey protein concentrate (WPC) and polysaccharides (inulin and carrageenan). Process parameters, including the inulin-to-water ratio, homogenization speed, and temperature, were optimized to achieve stable gels exhibiting high water- and fat-binding capacities. Scanning electron micrographs revealed a cohesive network containing uniformly dispersed lipid droplets, with carrageenan promoting a denser matrix. Chemical assessments demonstrated a notably lower saturated fatty acid content (10.85%) and only 0.179% trans-isomers, alongside an elevated proportion (71.17%) of polyunsaturated fatty acids. This fatty acid profile suggests potential cardiovascular health benefits compared with conventional animal fats. Texture analyses showed that carrageenan increased gel strength and hardness; Experiment 4 recorded values of 15.87 N and 279.62 N, respectively. Incorporation of WPC at moderate levels (3–4%) further enhanced the yield stress, reflecting a robust protein–polysaccharide network. These findings indicate that the developed emulsion gel offers a viable alternative to animal fats in meat products, combining superior nutritional attributes with acceptable textural properties. The substantial polyunsaturated fatty acid content and minimal trans-isomers, coupled with the gel’s mechanical stability, support the feasibility of creating reduced-fat, functional formulations that align with consumer demands for healthier alternatives. Full article

Large language models (LLMs) are increasingly being utilized to develop tools and services in various domains, including education. However, due to the nature of the training data, these models are susceptible to inherent social or cognitive biases, which can influence their outputs. Furthermore, their handling of critical topics, such as privacy and sensitive questions, is essential for responsible deployment. This study proposes a framework for the automatic detection of biases and violations of responsible use using a synthetic question-based dataset mimicking student–chatbot interactions. We employ the LLM-as-a-judge method to evaluate multiple LLMs for biased responses. Our findings show that some models exhibit more bias than others, highlighting the need for careful consideration when selecting models for deployment in educational and other high-stakes applications. These results emphasize the importance of addressing bias in LLMs and implementing robust mechanisms to uphold responsible AI use in real-world services. Full article

Quality assessment of the resistance spot welding process (RSW) is vital during manufacturing. Evaluating the quality without altering the joint material’s physical and mechanical properties has gained interest. This study uses a trained computer vision model to propose a cheap, non-destructive quality-evaluation methodology. The methodology connects the welding input and during-process parameters with the output visual quality information. A manual resistance spot welding machine was used to monitor and record the process input and output parameters to generate the dataset for training. The welding current, welding time, and electrode pressure data were correlated with the welding spot nugget’s quality, mechanical characteristics, and thermal and visible images. Six machine learning models were trained on visible and thermographic images to classify the weld’s quality and connect the quality characteristics (pull force and welding diameter) and the manufacturing process parameters with the visible and thermographic images of the weld. Finally, a cross-validation method validated the robustness of these models. The results indicate that the welding time and the angle between electrodes are highly influential parameters on the mechanical strength of the joint. Additionally, models using visible images of the welding spot exhibited superior performance compared to thermal images. Full article

Graph-to-text generation (G2T) involves converting structured graph data into natural language text, a task made challenging by the need for encoders to capture the entities and their relationships within the graph effectively. While transformer-based encoders have advanced natural language processing, their reliance on linearized data often obscures the complex interrelationships in graph structures, leading to structural loss. Conversely, graph attention networks excel at capturing graph structures but lack the pre-training advantages of transformers. To leverage the strengths of both modalities and bridge this gap, we propose a novel bidirectional dual cross-attention and concatenation (BDCC) mechanism that integrates outputs from a transformer-based encoder and a graph attention encoder. The bidirectional dual cross-attention computes attention scores bidirectionally, allowing graph features to attend to transformer features and vice versa, effectively capturing inter-modal relationships. The concatenation is applied to fuse the attended outputs, enabling robust feature fusion across modalities. We empirically validate BDCC on PathQuestions and WebNLG benchmark datasets, achieving BLEU scores of 67.41% and 66.58% and METEOR scores of 49.63% and 47.44%, respectively. The results outperform the baseline models and demonstrate that BDCC significantly improves G2T tasks by leveraging the synergistic benefits of graph attention and transformer encoders, addressing the limitations of existing approaches and showcasing the potential for future research in this area. Full article

Forest communities around the world have great difficulties in utilizing the economic potential of their forests, especially timber, under current technical requirements and legal frameworks. The present study examines the feasibility of motor–manual timber management among indigenous Chiquitano communities in Bolivia’s Eastern Lowlands. It evaluates local practices, tests technical optimization options, and assesses their technical, financial, and environmental impacts. Findings reveal that traditional motor–manual timber production is scarcely profitable, exacerbated by burdensome legal frameworks and limited market access. However, motor–manual forest management remains an essential source of income for communities, and it constitutes an important option for rural development. Field tests demonstrate that, with the use of better equipment such as quality chainsaws, and improved maintenance and workflows, productivity and profitability of local logging can be enhanced. Despite a low environmental impact, optimized motor–manual timber management continues to be constrained by governance challenges, logistical limitations, and limited markets for locally produced timber. The study recommends optimizing these aspects, including targeted technical support, market development, simplified legal frameworks, and the setting up of robust local governance structures to replace ineffective centralized command and control approaches. These improvements would enable communities to sustainably use timber from their forests while addressing their socio-economic needs. The findings underscore the potential of logging by local communities as an alternative to large-scale mechanized logging, for Bolivia and in other tropical forest countries. Full article

The increasingly severe issue of pesticide packaging waste (PPW) pollution poses a significant threat to human health and sustainable agricultural development. Encouraging farmers to recycle PPW is critical to addressing the “tragedy of the commons” problem in rural areas. Using data from the 2020 China Rural Revitalization Survey (CRRS), this paper examines the impact of digital literacy on farmers’ PPW recycling behavior. The results indicate that (1) a one-unit increase in digital literacy raises the likelihood of farmers recycling PPW by 20.1%. (2) Mechanism analysis shows that subjective cognition, information transmission, and social network are the key channels through which digital literacy affects farmers’ PPW recycling behavior. (3) After conducting multiple robustness tests—including Propensity Score Matching (PSM), instrumental variable methods, alternative weighting approaches for digital literacy, and different model specifications and samples—the findings remain robust. Based on these results, we propose the following policy recommendations: improve digital infrastructure in rural areas; enhance farmers’ digital literacy; establish incentive mechanisms; encourage village self-governance; and reinforce social oversight. Full article

This paper introduces PK-Judge, a novel neural network watermarking framework designed to enhance the intellectual property (IP) protection by incorporating an asymmetric cryptograp hic approach in the verification process. Inspired by the paradigm shift from HTTP to HTTPS in enhancing web security, this work integrates public key infrastructure (PKI) principles to establish a secure and verifiable watermarking system. Unlike symmetric approaches, PK-Judge employs a public key infrastructure (PKI) to decouple ownership validation from the extraction process, significantly increasing its resilience against adversarial attacks. Additionally, it incorporates a robust challenge-response mechanism to mitigate replay attacks and leverages error correction codes (ECC) to achieve an Effective Bit Error Rate (EBER) of zero, ensuring watermark integrity even under conditions such as fine-tuning, pruning, and overwriting. Furthermore, PK-Judge introduces a new requirement based on the principle of separation of privilege, setting a foundation for secure and scalable watermarking mechanisms in machine learning. By addressing these critical challenges, PK-Judge advances the state-of-the-art in neural network IP protection and integrity, paving the way for trust-based AI technologies that prioritize security and verifiability. Full article

Skin cancer is a major global health concern and one of the deadliest forms of cancer. Early and accurate detection significantly increases the chances of survival. However, traditional visual inspection methods are time-consuming and prone to errors due to artifacts and noise in dermoscopic images. To address these challenges, this paper proposes an innovative deep learning-based framework that integrates an ensemble of two pre-trained convolutional neural networks (CNNs), SqueezeNet and InceptionResNet-V2, combined with an improved Whale Optimization Algorithm (WOA) for feature selection. The deep features extracted from both models are fused to create a comprehensive feature set, which is then optimized using the proposed enhanced WOA that employs a quadratic decay function for dynamic parameter tuning and an advanced mutation mechanism to prevent premature convergence. The optimized features are fed into machine learning classifiers to achieve robust classification performance. The effectiveness of the framework is evaluated on two benchmark datasets, PH2 and Med-Node, achieving state-of-the-art classification accuracies of 95.48% and 98.59%, respectively. Comparative analysis with existing optimization algorithms and skin cancer classification approaches demonstrates the superiority of the proposed method in terms of accuracy, robustness, and computational efficiency. Our method outperforms the genetic algorithm (GA), Particle Swarm Optimization (PSO), and the slime mould algorithm (SMA), as well as deep learning-based skin cancer classification models, which have reported accuracies of 87% to 94% in previous studies. A more effective feature selection methodology improves accuracy and reduces computational overhead while maintaining robust performance. Our enhanced deep learning ensemble and feature selection technique can improve early-stage skin cancer diagnosis, as shown by these data. Full article