Search Results (30)

Recently, the convergence of advanced sensor technologies and innovations in artificial intelligence and robotics has highlighted facial emotion recognition (FER) as an essential component of human–computer interaction (HCI). Traditional FER studies based on handcrafted features and shallow machine learning have shown a limited performance, while convolutional neural networks (CNNs) have improved nonlinear emotion pattern analysis but have been constrained by local feature extraction. Vision transformers (ViTs) have addressed this by leveraging global correlations, yet both CNN- and ViT-based single networks often suffer from overfitting, single-network dependency, and information loss in ensemble operations. To overcome these limitations, we propose ArecaNet, an assembled residual enhanced cross-attention network that integrates multiple feature streams without information loss. The framework comprises (i) channel and spatial feature extraction via SCSESResNet, (ii) landmark feature extraction from specialized sub-networks, (iii) iterative fusion through residual enhanced cross-attention, (iv) final emotion classification from the fused representation. Our research introduces a novel approach by integrating pre-trained sub-networks specialized in facial recognition with an attention mechanism and our uniquely designed main network, which is optimized for size reduction and efficient feature extraction. The extracted features are fused through an iterative residual enhanced cross-attention mechanism, which minimizes information loss and preserves complementary representations across networks. This strategy overcomes the limitations of conventional ensemble methods, enabling seamless feature integration and robust recognition. The experimental results show that the proposed ArecaNet achieved accuracies of 97.0% and 97.8% using the public databases, FER-2013 and RAF-DB, which were 4.5% better than the existing state-of-the-art method, PAtt-Lite, for FER-2013 and 2.75% for RAF-DB, and achieved a new state-of-the-art accuracy for each database. Full article

The enhanced Loran (eLoran) system, a critical terrestrial backup for the Global Satellite Navigation System (GNSS), traditionally utilizes a Reed-Solomon (RS) code for its data communication, which presents limitations in error performance, particularly due to its decoding method. This paper introduces a significant advancement by proposing the replacement of the conventional RS code with a 128-ary polar code, which is designed to maintain compatibility with the established 128-ary Pulse Position Modulation (PPM) scheme integral to eLoran’s positioning function. A Soft–Soft (SS) demodulation method, based on a correlation receiver, is developed to provide the requisite soft information for the effective Successive Cancellation List (SCL) decoding of the 128-ary polar code. Comprehensive simulations demonstrate that the proposed 128-ary polar code with SS demodulation achieves a substantial error performance improvement, yielding an approximate 9.3 dB gain at the 0.01 FER level over the RS code in eLoran data communication with EPD-MD demodulation. Additionally, the proposed scheme improves data transmission efficiency—either reducing transmission duration by 2/3 or increasing message bit number by 250% for comparable error performance—without impacting the system’s primary positioning capabilities. Full article

The Alternating Direction Method of Multipliers (ADMM) has proven to be an efficient approach for implementing linear programming (LP) decoding of low-density parity-check (LDPC) codes. By introducing penalty terms into the LP decoding model’s objective function, ADMM-based variable node penalized decoding effectively mitigates non-integral solutions, thereby improving frame error rate (FER) performance, especially in the low signal-to-noise ratio (SNR) region. In this paper, we leverage the ADMM framework to derive explicit iterative steps for solving the LP decoding problem for LDPC codes with penalty functions. To further enhance decoding efficiency and accuracy, We propose an LDPC code decoding algorithm based on the symmetric ADMM (S-ADMM). We also establish some contraction properties satisfied by the iterative sequence of the algorithm. Through simulation experiments, we evaluate the proposed S-ADMM decoder using three standard LDPC codes and three representative fifth-generation (5G) codes. The results show that the S-ADMM decoder consistently outperforms conventional ADMM penalized decoders, offering significant improvements in decoding performance. Full article

Facial expression recognition (FER) plays a pivotal role in affective computing and human–computer interaction by enabling machines to interpret human emotions. However, conventional FER models often overlook individual differences in emotional intelligence (EI), which may significantly influence how emotions are perceived and expressed. This study investigates the effect of EI on facial expression recognition accuracy within the valence–arousal space. Participants were divided into high and low EI groups based on a composite score derived from the Tromsø Social Intelligence Scale and performance-based emotion tasks. Five deep learning models (EfficientNetV2-L/S, MaxViT-B/T, and VGG16) were trained on the AffectNet dataset and evaluated using facial expression data collected from participants. Emotional states were predicted as continuous valence and arousal values, which were then mapped onto discrete emotion categories for interpretability. The results indicated that individuals with higher EI achieved significantly greater recognition accuracy, particularly for emotions requiring contextual understanding (e.g., anger, sadness, and happiness), while fear was better recognized by individuals with lower EI. These findings highlight the role of emotional intelligence in modulating FER performance and suggest that integrating EI-related features into valence–arousal-based models could enhance the adaptiveness of affective computing systems. Full article

Facial expression recognition (FER) utilizes artificial intelligence for the detection and analysis of human faces, with significant applications across various scenarios. Our objective is to deploy the facial emotion recognition network on mobile devices and extend its application to diverse areas, including classroom effect monitoring, human–computer interaction, specialized training for athletes (such as in figure skating and rhythmic gymnastics), and actor emotion training. Recent studies have employed advanced deep learning models to address this task, though these models often encounter challenges like subpar performance and an excessive number of parameters that do not align with the requirements of FER for embedded devices. To tackle this issue, we have devised a lightweight network structure named RS-Xception, which is straightforward yet highly effective. Drawing on the strengths of ResNet and SENet, this network integrates elements from the Xception architecture. Our models have been trained on FER2013 datasets and demonstrate superior efficiency compared to conventional network models. Furthermore, we have assessed the model’s performance on the CK+, FER2013, and Bigfer2013 datasets, achieving accuracy rates of 97.13%, 69.02%, and 72.06%, respectively. Evaluation on the complex RAF-DB dataset yielded an accuracy rate of 82.98%. The incorporation of transfer learning notably enhanced the model’s accuracy, with a performance of 75.38% on the Bigfer2013 dataset, underscoring its significance in our research. In conclusion, our proposed model proves to be a viable solution for precise sentiment detection and estimation. In the future, our lightweight model may be deployed on embedded devices for research purposes. Full article

Ferulic acid (Fer) and geraniol (Ger) are natural compounds whose antioxidant and anti-inflammatory activity confer beneficial properties, such as antibacterial, anticancer, and neuroprotective effects. However, the short half-lives of these compounds impair their therapeutic activities after conventional administration. We propose, therefore, a new prodrug (Fer-Ger) obtained by a bio-catalyzed ester conjugation of Fer and Ger to enhance the loading of solid lipid microparticles (SLMs) designed as Fer-Ger delivery and targeting systems. SLMs were obtained by hot emulsion techniques without organic solvents. HPLC-UV analysis evidenced that Fer-Ger is hydrolyzed in human or rat whole blood and rat liver homogenates, with half-lives of 193.64 ± 20.93, 20.15 ± 0.75, and 3.94 ± 0.33 min, respectively, but not in rat brain homogenates. Studies on neuronal-differentiated mouse neuroblastoma N2a cells incubated with the reactive oxygen species (ROS) inductor H₂O₂ evidenced the Fer-Ger ability to prevent oxidative injury, despite the fact that it appears ROS-promoting. The amounts of Fer-Ger encapsulated in tristearin SLMs, obtained in the absence or presence of glucose, were 1.5 ± 0.1%, allowing the control of the prodrug release (glucose absence) or to sensibly enhance its water dissolution rate (glucose presence). These new “green” carriers can potentially prolong the beneficial effects of Fer and Ger or induce neuroprotection as nasal formulations. Full article

Facial emotion recognition (FER) is crucial across psychology, neuroscience, computer vision, and machine learning due to the diversified and subjective nature of emotions, varying considerably across individuals, cultures, and contexts. This study explored FER through convolutional neural networks (CNNs) and Histogram Equalization techniques. It investigated the impact of histogram equalization, data augmentation, and various model optimization strategies on FER accuracy across different datasets like KDEF, CK+, and FER2013. Using pre-trained VGG architectures, such as VGG19 and VGG16, this study also examined the effectiveness of fine-tuning hyperparameters and implementing different learning rate schedulers. The evaluation encompassed diverse metrics including accuracy, Area Under the Receiver Operating Characteristic Curve (AUC-ROC), Area Under the Precision–Recall Curve (AUC-PRC), and Weighted F1 score. Notably, the fine-tuned VGG architecture demonstrated a state-of-the-art performance compared to conventional transfer learning models and achieved 100%, 95.92%, and 69.65% on the CK+, KDEF, and FER2013 datasets, respectively. Full article

Facial expression recognition (FER) plays a crucial role in understanding human emotions and is becoming increasingly relevant in educational contexts, where personalized and empathetic interactions are essential. The problems with existing approaches are typically solved using a single deep learning method, which is not robust with complex datasets, such as FER data, which have a characteristic imbalance and multi-class labels. In this research paper, an innovative approach to FER using a homogeneous ensemble convolutional neural network, called HoE-CNN, is presented for future online learning education. This paper aims to transfer the knowledge of models and FER classification using ensembled homogeneous conventional neural network architectures. FER is challenging to research because there are many real-world applications to consider, such as adaptive user interfaces, games, education, and robot integration. HoE-CNN is used to improve the classification performance on an FER dataset, encompassing seven main multi-classes (Angry, Disgust, Fear, Happy, Sad, Surprise, Neutral). The experiment shows that the proposed framework, which uses an ensemble of deep learning models, performs better than a single deep learning model. In summary, the proposed model will increase the efficiency of FER classification results and solve FER2013 at a accuracy of 75.51%, addressing both imbalanced datasets and multi-class classification to transfer the application of the model to online learning applications. Full article

Cereals are a valuable source of biologically active compounds. Phenolic compounds, of which the phenolic acids (PA) found in cereal grains constitute a significant proportion, are characterized by health-promoting properties largely due to their antioxidant capacity. PA, located mainly in the outer parts of the grain, play an important role in preventing environmental stresses. Triticale is a cereal species of increasing economic value, and also value for human consumption. The aim of this study was to demonstrate the effect of conventional (CONV) and integrated (INTEG) cropping systems on antioxidant activity and content of selected PA in triticale cultivars (Meloman, Panteon, Belcanto) grain. The experiment was conducted in seasons from 2019/2020 to 2021/2022. Among the PA tested, ferulic acid (FER) had the highest contribution to total PA content (TPAs), with 519, 99, and 1115 μg g⁻¹ in whole grain, flour, and bran, respectively. The unfavorable hydrothermal conditions occurring in the seasons (rainfall deficit) increased TPA, mainly in whole grain. Grain cv. Meloman had the highest PA content in whole grain, flour, and bran and cv. Belcanto had the lowest, with differences of 22.7, 18.2, and 15.7% respectively. Cultivation of triticale under the CONV vs. INTEG cropping system resulted in reduced amounts of TPAs in flour and bran and PA: p-hydroxybenzoic acid (p-HB) in flour, syringic acid (SYR) in whole grain and bran, and ferulic acid (FER) and sinapic acid (SIN) in bran. The CONV cropping system also caused a decrease in antioxidant activity (AOA) in flour and bran. In most of the cases analyzed, the highest antioxidant activity and content of PA were found in bran, and the lowest were found in flour. The high presence of PA in triticale grain indicates that this cereal, especially when grown under the INTEG cropping system, can be destined for consumption and provide a source of valuable antioxidants for various food and nutraceutical purposes. Full article

High-performance Ferritic (HiperFer) steels are a novel class of heat-resistant, fully ferritic, Laves phase precipitation hardened materials. In comparison to conventional creep strength-enhanced 9–12 wt.% Cr ferritic–martensitic steels, HiperFer features increased mechanical strength, based on a thermodynamically stable distribution of small (Fe,Cr,Si)₂(Nb,W) Laves phase precipitates, and—owing to its increased chromium content of 17 wt.%—improved resistance to steam oxidation, resulting in superior temperature capability up to 650 °C. Previous publications focused on alloying, thermomechanical processing, and basic mechanical property evaluation. The current paper concentrates on the effect of heat treatment on microstructural features, especially Laves phase population, and the resulting creep performance. At 650 °C and a creep stress of 100 MPa, an increase in rupture time of about 100% was achieved in comparison to the solely thermomechanically processed state. Full article

The Société Nationale des Chemins de fer Français (SNCF) is facing a significant challenge to meet the growth in rail traffic while maintaining continuous service, particularly in densely populated areas such as Paris. To tackle this challenge, the SNCF has implemented several electrification projects. These projects aim to reduce line losses and decrease voltage drops on the railway network. Amongst the possible technological choices, high temperature superconductor (HTS) cables have been evaluated, since they offer greater energy density at lower electrical losses than conventional cables. This feature is advantageous in order to transmit more electrical energy at a lesser footprint than conventional cable, therefore avoiding costly modifications of the existing infrastructures. In the present work, the electromagnetic response of two HTS cables topologies, unipolar and bipolar, was analyzed, and their impact on a direct current (DC) railway network under load was assessed. A commercial finite element (FE) software, COMSOL Multiphysics, was used to carry out a detailed FE model that accounts for the non-linearity of the electrical resistivity ρ (J, B, θ) of the superconducting cable. This FE model was coupled with a lumped-parameter circuit model of the railway network, which is particularly suited for transient simulations considering train motion. Based on a case study representing a portion of the Parisian railway network, it was found that the insertion of a superconducting cable can result in a reduction of electrical losses by 60% compared to conventional cable as well as an 8.6% reduction in the total electrical consumption of the traction network. Full article

Facial Expression Recognition (FER) is a growing area of research due to its numerous applications in market research, video gaming, healthcare, security, e-learning, and robotics. One of the most common frameworks for recognizing facial expressions is by extracting facial features from an image and classifying them as one of several prototypic expressions. Despite the recent advances, it is still a challenging task to develop robust facial expression descriptors. This study aimed to analyze the performances of various local descriptors and classifiers in the FER problem. Several experiments were conducted under different settings, such as varied extraction parameters, different numbers of expressions, and two datasets, to discover the best combinations of local descriptors and classifiers. Of all the considered descriptors, HOG (Histogram of Oriented Gradients) and ALDP (Angled Local Directional Patterns) were some of the most promising, while SVM (Support Vector Machines) and MLP (Multi-Layer Perceptron) were the best among the considered classifiers. The results obtained signify that conventional FER approaches are still comparable to state-of-the-art methods based on deep learning. Full article

Background: The trunk strength conventional ratio (CR) has been evaluated. However, the functional ratio and the ratio of strength to body weight (BW) or muscle mass (MM) have been poorly explored. Relative strength is a measure of muscle quality. Objectives: To analyze the trunk strength ratio normalized by BW and MM and compare the trunk’s conventional and functional ratios collected in isokinetic and isometric conditions. Methods: Twenty-seven healthy males (21.48 ± 2.08 years, 70.22 ± 7.65 kg) were evaluated for trunk isometric and isokinetic strength using a functional electromechanical dynamometer. Results: The extensor’s strength was greater than the flexors, with a CR of 0.41 ± 0.10 to 0.44 ± 0.10. Muscle quality was higher in eccentric contraction and high velocity for flexors and extensors. The functional flexor ratio (FFR) ranged between 0.41 ± 0.09 and 0.92 ± 0.27. The functional extensor ratio (FER) ranged between 2.53 ± 0.65 and 4.92 ± 1.26. The FFR and FER showed significant differences between velocities when considering the peak strength (p = 0.001) and mean strength (p = 0.001). Conclusions: Trunk extensors were stronger than the flexors; thus, the CR was less than one. Muscle quality was higher at a high velocity. Unlike CR, FFR and FER behaved differently at distinct velocities. This finding highlights the need to explore the behavior of the functional ratio in different populations. Full article

Ferulic acid (Fer) is known for its antioxidant and anti-inflammatory activities, which are possibly useful against neurodegenerative diseases. Despite the ability of Fer to permeate the brain, its fast elimination from the body does not allow its therapeutic use to be optimized. The present study proposes the preparation and characterization of tristearin- or stearic acid-based solid lipid microparticles (SLMs) as sustained delivery and targeting systems for Fer. The microparticles were produced by conventional hot emulsion techniques. The synthesis of the methyl ester of Fer (Fer-Me) allowed its encapsulation in the SLMs to increase. Fer-Me was hydrolyzed to Fer in rat whole blood and liver homogenate, evidencing its prodrug behavior. Furthermore, Fer-Me displayed antioxidant and anti-inflammatory properties. The amount of encapsulated Fer-Me was 0.719 ± 0.005% or 1.507 ± 0.014% in tristearin or stearic acid SLMs, respectively. The tristearin SLMs were able to control the prodrug release, while the stearic acid SLMs induced a significant increase of its dissolution rate in water. Jointly, the present results suggest that the tristearin SLMs loaded with Fer-Me could be a potential formulation against peripheral neuropathic pain; conversely, the stearic acid SLMs could be useful for Fer-Me uptake in the brain after nasal administration of the formulation. Full article

In the present study, the first tailored steel based on HiperFer (high-performance ferrite) was developed specifically for the additive manufacturing process. This steel demonstrates its full performance potential when produced via additive manufacturing, e.g., through a high cooling rate, an in-build heat treatment, a tailored microstructure and counteracts potential process-induced defects (e.g. pores and cavities) via “active” crack-inhibiting mechanisms, such as thermomechanically induced precipitation of intermetallic (Fe,Cr,Si)₂(W,Nb) Laves phase particles. Two governing mechanisms can be used to accomplish this: (I) “in-build heat treatment” by utilizing the “temper bead effect” during additive manufacturing and (II) “dynamic strengthening” under cyclic, plastic deformation at high temperature. To achieve this, the first HiperFer^AM (additive manufacturing) model alloy with high precipitation kinetics was developed. Initial mechanical tests indicated great potential in terms of the tensile strength, elongation at rupture and minimum creep rate. During the thermomechanical loading, global sub-grain formation occurred in the HiperFer^AM, which refined the grain structure and allowed for higher plastic deformation, and consequently, increased the elongation at rupture. The additive manufacturing process also enabled the reduction of grain size to a region, which has not been accessible by conventional processing routes (casting, rolling, heat treatment) so far. Full article