Search Results (166)

Remote sensing images are often degraded by atmospheric haze, which not only reduces image quality but also complicates information extraction, particularly in high-level visual analysis tasks such as object detection and scene classification. State-space models (SSMs) have recently emerged as a powerful paradigm for vision tasks, showing great promise due to their computational efficiency and robust capacity to model global dependencies. However, most existing learning-based dehazing methods lack physical interpretability, leading to weak generalization. Furthermore, they typically rely on spatial features while neglecting crucial frequency domain information, resulting in incomplete feature representation. To address these challenges, we propose ScaleViM-PDD, a novel network that enhances an SSM backbone with two key innovations: a Multi-scale EfficientViM with Physical Decoupling (ScaleViM-P) module and a Dual-Domain Fusion (DD Fusion) module. The ScaleViM-P module synergistically integrates a Physical Decoupling block within a Multi-scale EfficientViM architecture. This design enables the network to mitigate haze interference in a physically grounded manner at each representational scale while simultaneously capturing global contextual information to adaptively handle complex haze distributions. To further address detail loss, the DD Fusion module replaces conventional skip connections by incorporating a novel Frequency Domain Module (FDM) alongside channel and position attention. This allows for a more effective fusion of spatial and frequency features, significantly improving the recovery of fine-grained details, including color and texture information. Extensive experiments on nine publicly available remote sensing datasets demonstrate that ScaleViM-PDD consistently surpasses state-of-the-art baselines in both qualitative and quantitative evaluations, highlighting its strong generalization ability. Full article

Free-space optical (FSO) communication is a promising high-capacity solution for future wireless networks, particularly for backhaul and fronthaul links in 5G and emerging 6G systems. However, it remains highly vulnerable to environmental impairment, especially in arid regions prone to dust storms. While prior studies have addressed atmospheric effects such as fog and turbulence, the specific impact of dust on signal performance remains insufficiently explored. This work presents a probabilistic modeling framework for evaluating the performance of an intensity modulation/direct detection (IM/DD) FSO system under dust storm conditions. Using a controlled laboratory environment, we conducted measurements of the optical signal under dust-induced channel conditions using real-world dust samples collected from an actual dust storm. We identified the Beta distribution as the most accurate model for the measured signal fluctuations. Closed-form expressions were derived for average bit error rate (BER), outage probability, and channel capacity. The close agreement between the analytical, approximate, and simulated results validates the proposed model as a reliable tool for evaluating FSO system performance. The results show that the forward error correction (FEC) BER threshold of ${10}^{-3}$ is achieved at approximately 10.5 dB, and the outage probability drops below ${10}^{-3}$ at 10 dB average SNR. Full article

Accurate detection of dangerous driving behaviors is crucial for improving the safety of intelligent transportation systems. However, existing methods often struggle with limited feature extraction capabilities and insufficient attention to multiscale and contextual information. To overcome these limitations, we propose YOLO-AFR (YOLO with Adaptive Feature Refinement) for dangerous driving behavior detection. YOLO-AFR builds upon the YOLOv12 architecture and introduces three key innovations: (1) the redesign of the original A2C2f module by introducing a Feature-Refinement Feedback Network (FRFN), resulting in a new A2C2f-FRFN structure that adaptively refines multiscale features, (2) the integration of self-calibrated convolution (SC-Conv) modules in the backbone to enhance multiscale contextual modeling, and (3) the employment of a SEAM-based detection head to improve global contextual awareness and prediction accuracy. These three modules combine to form a Calibration-Refinement Loop, which progressively reduces redundancy and enhances discriminative features layer by layer. We evaluate YOLO-AFR on two public driver behavior datasets, YawDD-E and SfdDD. Experimental results show that YOLO-AFR significantly outperforms the baseline YOLOv12 model, achieving improvements of 1.3% and 1.8% in mAP@0.5, and 2.6% and 12.3% in mAP@0.5:0.95 on the YawDD-E and SfdDD datasets, respectively, demonstrating its superior performance in complex driving scenarios while maintaining high inference speed. Full article

The serendipitous discovery of antiparasitic drugs, such as quinine and artemisinin, of plant origin reveals that searching new chemical pharmacophores from medicinal plants is valuable. The present study sought to explore the antiplasmodial, antileishmanial, and antitrypanosomal activities of Lippia adoensis extracts. Crude extracts of L. adoensis leaves and twigs, which were obtained by extraction using 70% ethanol in water, were assayed for antiplasmodial activity against P. falciparum 3D7 and Dd2 through the SYBR green I-based fluorescence assay; and for antileishmanial, antitrypanosomal, and cytotoxic effects on Leishmania donovani, Trypanosoma brucei brucei, and Vero cells, respectively, using resazurin colorimetric assays. In vitro phytochemical analysis of L. adoensis extracts was performed using standard methods. Moreover, liquid chromatography–mass spectrometry (LC-MS) feature-based detection and molecular networking flow on Global Natural Product Social (GNPS) were also used for the phytochemical screening of L. adoensis extracts. Crude extracts from L. adoensis inhibited the growth of P. falciparum (3D7 and Dd2) (IC₅₀s; (3D7): 10.00 and 97.46 μg/mL; (Dd2): 29.48 and 26.96 μg/mL), L. donovani (IC₅₀s: 22.87–10.52 μg/mL), and T. brucei brucei (IC₅₀s: 2.30–55.06 μg/mL). The extracts were found to be non-cytotoxic to Vero cells, thus yielding median cytotoxic concentrations (CC₅₀s) above 100 μg/mL. In vitro phytochemical analysis of the crude extracts revealed the presence of alkaloids, terpenoids, phenolic compounds, and carbohydrates. The LC-MS tandem molecular networking flow predicted that the extracts contained valsafungin A and bacillamidin in the first cluster, and fatty acids, ketone, and aldehyde derivatives in the second cluster. Overall, the present study demonstrated the antiparasitic effects of L. adoensis extracts, thus justifying the use of this plant in the traditional treatment of fever and malaria conditions. Nevertheless, detailed metabolomic studies and antiparasitic mechanisms of action of the extracts are expected to unveil the potential antiparasitic hit compounds. Full article

China is the world’s largest producer of chili peppers, which occupy particularly important economic and social values in various fields such as medicine, food, and industry. However, during its production process, chili peppers are affected by pests and diseases, resulting in significant yield reduction due to the temperature and environment. In this study, a lightweight pepper disease identification method, DD-YOLO, based on the YOLOv8n model, is proposed. First, the deformable convolutional module DCNv2 (Deformable ConvNetsv2) and the inverted residual mobile block iRMB (Inverted Residual Mobile Block) are introduced into the C2F module to improve the accuracy of the sampling range and reduce the computational amount. Secondly, the DySample sampling operator (Dynamic Sample) is integrated into the head network to reduce the amount of data and the complexity of computation. Finally, we use Large Separable Kernel Attention (LSKA) to improve the SPPF module (Spatial Pyramid Pooling Fast) to enhance the performance of multi-scale feature fusion. The experimental results show that the accuracy, recall, and average precision of the DD-YOLO model are 91.6%, 88.9%, and 94.4%, respectively. Compared with the base network YOLOv8n, it improves 6.2, 2.3, and 2.8 percentage points, respectively. The model weight is reduced by 22.6%, and the number of floating-point operations per second is improved by 11.1%. This method provides a technical basis for intensive cultivation and management of chili peppers, as well as efficiently and cost-effectively accomplishing the task of identifying chili pepper pests and diseases. Full article

Mastitis is an important disease limiting milk production in dairy cows. Somatic cell score is commonly used as one of the main ways to gauge the level of mastitis in dairy cows, with higher somatic cell scores usually indicating possible mastitis. However, the main molecular markers affecting somatic cell scores remain unknown. The aim of this study was to investigate the association between single nucleotide polymorphisms in the SLC11A1 gene and somatic cell score in Chinese Holstein cows. In this study, 210 Chinese Holstein cows were genotyped and potential SNPs were detected by DNA sequencing, PCR-SSCP and PCR-RFLP analysis. Our results revealed two SNPs were identified in the CDS region of SLC11A1: c.723C>T and c.1144C>G. For the c.723C>T polymorphic site, two genotypes (AA, AB) were found and the genotype frequencies were 0.790 and 0.210, respectively. The results of the association analysis showed that the mean somatic cell score of the AA genotypes were significantly lower than those of the AB genotypes, suggesting that the A allele is a potential marker for improving mastitis resistance in Chinese Holstein cows. For the c.1144C>G polymorphic site, three genotypes (CC, CD, and DD) were found and the genotype frequencies were 0.629, 0.352 and 0.019, respectively. The association analysis revealed that the mean somatic cell score of CC genotypes was lower than that of CD and DD genotypes, however, no significant differences were observed among the various genotype groups when subjected to pair-wise comparisons. The bioinformatic analysis showed that these mutations affected the secondary and tertiary structure of SLC11A1 mRNA, suggesting that they may affect gene expression or protein translation and function. Finally, we predicted the SLC11A1 protein interaction network and found that SPI1, NOD2, TLR2 and S100A12 interacted with SLC11A1 and were reported as candidate genes associated with mastitis resistance. The results indicated that the SNP (c.723C>T) could be potential molecular marker for improving mastitis resistance traits in Chinese Holstein cows. We recommend further validation of this SNP in larger populations and its potential integration into breeding programs to enhance mastitis resistance in dairy cows. Full article

Skeletal muscle satellite cells (SMSCs) are quiescent stem cells located in skeletal muscle tissue and function as the primary reservoir of myogenic progenitors for muscle growth and regeneration. However, the molecular and metabolic mechanisms governing their differentiation in geese remain largely unexplored. This study comprehensively examined the morphological, transcriptional, and metabolic dynamics of goose SMSCs across three critical differentiation stages: the quiescent stage (DD0), the differentiation stage (DD4), and the late differentiation stage (DD6). By integrating transcriptomic and metabolomic analyses, stage-specific molecular signatures and regulatory networks involved in SMSC differentiation were identified. Principal component analysis revealed distinct clustering patterns in gene expression and metabolite profiles across these stages, highlighting dynamic shifts in lipid metabolism and myogenesis. The PPAR signaling pathway emerged as a key regulator, with crucial genes such as PPARG, IGF1, ACSL5, FABP5, and PLIN1 exhibiting differentiation-dependent expression patterns. Notably, PPARG and IGF1 displayed negative correlations with adenosine and L-carnitine levels, suggesting their role in metabolic reprogramming during myotube formation. Additionally, MYOM2 and MYBPC1 exhibited stage-specific regulation and positively correlated with 2,3-dimethoxyphenylamine. This study provides a foundational framework for understanding muscle development and regeneration, offering valuable insights for both agricultural and biomedical research. Full article

Reproduction and breeding are crucial to maintaining abalone aquaculture. Understanding the molecular underpinnings of sexual maturation is essential for advancing knowledge in reproductive biology. However, the molecular mechanisms of gonadal development in abalones remain poorly understood, particularly in microRNA (miRNA)-mediated regulation. Thus, this study conducted a comprehensive transcriptomic analysis of abalone Haliotis discus hannai (H. discus hannai) to identify genes and miRNAs associated with ovarian and testicular discovery. This study identified 685 differentially expressed (DE) genes between the H. discus hannai ovary (DD_ovary) and testis (DD_testis) groups, comprising 479 upregulated and 206 downregulated genes in the DD_ovary. Moreover, 137 miRNAs, including 83 novel and 54 known miRNAs, were detected, with 30 upregulated and 27 downregulated in the DD_ovary compared to the DD_testis. Bioinformatics analysis revealed that these miRNAs regulate key processes such as carbohydrate metabolic processes, kinase and hydrolase activity, and starch and sucrose metabolism, all potentially associated with reproductive traits. Further, key mRNA candidates, including Vitelline envelope sperm lysin receptor (Verl) and Testis-specific serine/threonine-protein kinase (Tssk) 1, and miRNAs such as novel_90 and novel_120, were identified as components of a functional miRNA-mRNA network associated with sexual maturity and sex determination. These key genes were verified using qRT-PCR and fluorescence in situ hybridization (FISH). These transcriptomic and miRNA datasets provide valuable resources for understanding abalone reproductive biology and may support molecular breeding strategies. Full article

Advanced communication systems, particularly in the context of autonomous driving and integrated sensing and communication (ISAC), require high precision and refresh rates for environmental perception, alongside reliable data transmission. This paper presents a novel approach to enhance the ISAC performance in existing 4G and 5G systems by utilizing a two-dimensional offset in the Delay–Doppler (DD) domain, effectively leveraging the sounding reference signal (SRS) resources. This method aims to improve spectrum efficiency and sensing accuracy in vehicular networks. However, a key challenge arises from interference between multiple users after the wireless propagation of signals. To address this, we propose a deep learning-based interference mitigation solution using an UNet architecture, which operates on the Range–Doppler maps. The UNet model, with its encoder–decoder structure, efficiently filters out unwanted signals, therefore enhancing the system performance. Simulation results show that the proposed method significantly improves the accuracy of environmental sensing and resource utilization while mitigating interference, even in dense network scenarios. Our findings suggest that this DD-domain-based approach offers a promising solution to optimizing ISAC capabilities in current and future communication systems. Full article

Chitosan was hydro-thermally extracted from grey shrimp carapaces and characterized using various techniques (degree of deacetylation (DD), viscosity, thermogravimetric analysis (TGA), scanning electron microscopy (SEM), and surface area analysis (BET)). It was then used for Cu(II) removal in a batch system, achieving a maximum capacity of 89 mg/g under standard conditions. Both pseudo-first-order and pseudo-second-order nonlinear kinetic models described the adsorption of Cu(II) ions on chitosan well, with a better fit of the pseudo-first-order model at low concentrations, while the equilibrium data suggested that the Langmuir model was suitable for describing the adsorption system, with a maximum adsorption capacity of 123 mg/g. A response surface methodology and central composite design were used to optimise and evaluate the effects of six independent parameters: initial Cu(II) concentration, pH, chitosan concentration (S/L), temperature (T), contact time (t), and NaCl concentration on the adsorption efficiency of Cu(II) by the synthesised chitosan. The proposed model was confirmed to accurately describe the phenomenon within the experimental range, achieving an R² value of 1. ANOVA indicated that the initial concentrations of Cu(II) and chitosan concentration (S/L) were the most significant factors, while the other variables had no significant effect on the process. The adsorption capacity of Cu(II) onto the prepared chitosan was also optimised and modelled using artificial neural networks (ANNs). The maximum amount, qmax = 468 mg·g⁻¹, shows that chitosan is a highly effective adsorbent, chelating and complexing for copper ions. Full article

In domains such as fraud detection, healthcare, and industrial equipment maintenance, streaming data often exhibit characteristics such as continuous generation, high real-time processing requirements, and complex distributions, making it susceptible to concept drift. Traditional shallow models, with their limited representational capacity, struggle to fully capture the latent conceptual knowledge inherent in the dynamic and evolving nature of streaming data. To address this challenge, we propose a concept drift detection method based on deep neural networks combined with autoencoders (Concept Drift Detection Based on Deep Neural Network Combined with Autoencoder, DNN+AE-DD). In the DNN+AE-DD, a deep neural network is first employed as the base model for pretraining, and the hidden layer parameters of the model are transferred to a network with an identical structure for stream data processing, where certain hidden layers are frozen. Subsequently, the hidden layer outputs from both the pretraining and stream data processing phases are collected and used as training and testing data to initialize and predict using an autoencoder model. Concept drift is then detected by combining the reconstruction error of the autoencoder with the 3$\sigma$ principle. Experimental results on both real-world and synthetic datasets demonstrate that, compared to traditional shallow concept drift detection methods, this approach effectively and accurately detects anomalies in streaming data, confirming the proposed model’s high sensitivity to concept drift. Full article

Conventional inspections in car damage assessments depend on visual judgments by human inspectors, which are labor-intensive and prone to fraudulent practices through manipulating damages. Recent advancements in artificial intelligence have given rise to a state-of-the-art object detection algorithm, the You Only Look Once algorithm (YOLO), that sets a new standard in smart and automated damage assessment. This study proposes an enhanced YOLOv9 network tailored to detect six types of car damage. The enhancements include the convolutional block attention module (CBAM), applied to the backbone layer to enhance the model’s ability to focus on key damaged regions, and the SCYLLA-IoU (SIoU) loss function, introduced for bounding box regression. To be able to assess the damage severity comprehensively, we propose a novel formula named damage severity index (DSI) for quantifying damage severity directly from images, integrating multiple factors such as the number of detected damages, the ratio of damage to the image size, object detection confidence, and the type of damage. Experimental results on the CarDD dataset show that the proposed model outperforms state-of-the-art YOLO algorithms by 1.75% and that the proposed DSI demonstrates intuitive assessment of damage severity with numbers, aiding repair decisions. Full article

Fracture toughness is a key property of epoxy resins with a high glass transition temperature (T_g), used in carbon fiber/epoxy composites for aerospace applications. Conventional toughening methods rely on adding toughening agents, often compromising the processibility and thermal stability. This study introduces a simple self-toughening approach that enhances the fracture toughness without sacrificing other properties by controlling the cured epoxy network structure. Tetraglycidyl 4,4′-diaminodiphenylmethane (TGDDM) epoxy resin was cured using mixtures of structural isomeric curing agents, 3,3′- and 4,4′-diaminodiphenyl sulfone (3,3′- and 4,4′-DDS), at ratios of 7:3, 5:5, and 3:7. The optimal 7:3 ratio produced a resin with 30% higher fracture toughness compared to TGDDM/3,3′-DDS and 100% higher than the TGDDM/4,4′-DDS system. The T_g of the self-toughened resin ranged from 241 to 266 °C, which was intermediate between the T_g values of the TGDDM/3,3′-DDS and TGDDM/4,4′-DDS systems. This improvement is attributed to the higher crosslink density and reduced free volume of the epoxy network. These findings demonstrate that simply mixing isomeric curing agents enables self-toughening, providing a practical and efficient strategy to enhance the performance of high-T_g epoxy resins in advanced composite applications. Full article

In this paper, delay Doppler (DD) domain is utilized for enabling an efficient handover-triggering mechanism in highly dynamic unmanned aerial vehicles (UAVs) or drones to ground networks. In the proposed scheme, the estimated DD channel gains using DD multi-carrier modulation (DDMC), e.g., orthogonal time frequency space (OTFS) modulation, are utilized for triggering the handover decisions. This is motivated by the fact that the estimated DD channel gain is time-invariant throughout the whole OTFS symbol despite the entity speed. This results in more stable handover decisions over that based on the time-varying received-signal strength (RSS) or frequency time (FT) channel gains using orthogonal frequency division multiplexing (OFDM) modulation employed in fifth-generation–new radio (5G-NR) and its predecessors. To mathematically bind the performance of the proposed scheme, we studied its performance under channel estimation errors of the most dominant DD channel estimators, i.e., least square (LS) and minimum mean square error (MMSE), and we prove that they have marginal effects on its performance. Numerical analyses demonstrated the superiority of the proposed DD-based handover-triggering scheme over candidate benchmarks in terms of the handover overhead, the achievable throughput, and ping-pong ratio under different simulation conditions. Full article

Conventional epoxy thermosets, with irreversible crosslinking networks, cannot be reprocessed and recycled. Furthermore, the utilization of petroleum-based materials accelerates the depletion of non-renewable resources. The introduction of dynamic covalent bonds and the use of bio-based materials for thermosets can effectively address the above issues. Herein, a series of bio-based epoxy vitrimers with dynamic covalent imine bonds were synthesized via a simple solvent-free, one-pot method using vanillin-derived aldehyde monomers, 4,4-diaminodiphenylsulfone (DDS) and bisphenol F diglycidyl ether (BFDGE) as raw materials. The effect of crosslinking density, crosslinking structure and imine bond content on the resulting bio-based vitrimers was studied, demonstrating their excellent thermal properties, UV shielding and solvent resistance, as well as outstanding mechanical properties compared to those of the previously reported vitrimers. In particular, the cured neat resin of vitrimer had a maximum tensile strength of 109 MPa and Young’s modulus of 6257 MPa, which are higher than those of previously reported imine-based vitrimers. The dynamic imine bonds endow these vitrimers with good reprocessability upon heating (over 70% recovery) and degradation under acidic conditions, enabling recycling by physical routes and gentle degradation by chemical routes. This study demonstrates a simple and effective process to prepare high-performance bio-based and recycled epoxy thermosets. Full article