Search Results (113)

Limited depth of field in modern optical imaging systems often results in partially focused images. Multi-focus image fusion (MFF) addresses this by synthesizing an all-in-focus image from multiple source images captured at different focal planes. While deep learning-based MFF methods have shown promising results, existing approaches face significant challenges. Convolutional Neural Networks (CNNs) often struggle to capture long-range dependencies effectively, while Transformer and Mamba-based architectures, despite their strengths, suffer from high computational costs and rigid input size constraints, frequently necessitating patch-wise fusion during inference—a compromise that undermines the realization of a true global receptive field. To overcome these limitations, we propose MLP-MFF, a novel lightweight, end-to-end MFF network built upon the Pyramid Fusion Multi-Layer Perceptron (PFMLP) architecture. MLP-MFF is specifically designed to handle flexible input scales, efficiently learn multi-scale feature representations, and capture critical long-range dependencies. Furthermore, we introduce a Dual-Path Adaptive Multi-scale Feature-Fusion Module based on Hybrid Attention (DAMFFM-HA), which adaptively integrates hybrid attention mechanisms and allocates weights to optimally fuse multi-scale features, thereby significantly enhancing fusion performance. Extensive experiments on public multi-focus image datasets demonstrate that our proposed MLP-MFF achieves competitive, and often superior, fusion quality compared to current state-of-the-art MFF methods, all while maintaining a lightweight and efficient architecture. Full article

Hyperspectral classification of wetland environments remains challenging due to high spectral similarity, class imbalance, and blurred boundaries. To address these issues, we propose a novel Multi-Branch Channel-Gated Swin Transformer network (MBCG-SwinNet). In contrast to previous CNN-based designs, our model introduces a Swin Transformer spectral branch to enhance global contextual modeling, enabling improved spectral discrimination. To effectively fuse spatial and spectral features, we design a residual feature interaction chain comprising a Residual Spatial Fusion (RSF) module, a channel-wise gating mechanism, and a multi-scale feature fusion (MFF) module, which together enhance spatial adaptivity and feature integration. Additionally, a DenseCRF-based post-processing step is employed to refine classification boundaries and suppress salt-and-pepper noise. Experimental results on three UAV-based hyperspectral wetland datasets from the Yellow River Delta (Shandong, China)—NC12, NC13, and NC16—demonstrate that MBCG-SwinNet achieves superior classification performance, with overall accuracies of 97.62%, 82.37%, and 97.32%, respectively—surpassing state-of-the-art methods. The proposed architecture offers a robust and scalable solution for hyperspectral image classification in complex ecological settings. Full article

Monitoring the growth status of crop leaves is an integral part of agricultural management and involves important tasks such as leaf shape analysis and area calculation. To achieve this goal, accurate leaf segmentation is a critical step. However, this task presents a challenge, as crop leaf images often feature substantial overlap, obstructing the precise differentiation of individual leaf edges. Moreover, existing segmentation methods fail to preserve fine edge details, a deficiency that compromises precise morphological analysis. To overcome these challenges, we introduce MSFUnet, an innovative network for semantic segmentation. MSFUnet integrates a multi-path feature fusion (MFF) mechanism and an edge-detail focus (EDF) module. The MFF module integrates multi-scale features to improve the model’s capacity for distinguishing overlapping leaf areas, while the EDF module employs extended convolution to accurately capture fine edge details. Collectively, these modules enable MSFUnet to achieve high-precision individual leaf segmentation. In addition, standard image augmentations (e.g., contrast/brightness adjustments) were applied to mitigate the impact of variable lighting conditions on leaf appearance in the input images, thereby improving model robustness. Experimental results indicate that MSFUnet attains an MIoU of 93.35%, outperforming conventional segmentation methods and highlighting its effectiveness in crop leaf growth monitoring. Full article

Robust and efficient contactless human–machine interaction is critical for integrated sensor systems in clinical settings, demanding low-power solutions adaptable to edge computing platforms. This paper presents a real-time hand gesture recognition system using a low-power Frequency-Modulated Continuous Wave (FMCW) radar sensor, featuring a novel Multiple Feature Fusion (MFF) framework optimized for deployment on edge devices. The proposed system integrates velocity profiles, angular variations, and spatial-temporal features through a dual-stage processing architecture: an adaptive energy thresholding detector segments gestures, followed by an attention-enhanced neural classifier. Innovations include dynamic clutter suppression and multi-path cancellation optimized for complex clinical environments. Experimental validation demonstrates high performance, achieving 98% detection recall and 93.87% classification accuracy under LOSO cross-validation. On embedded hardware, the system processes at 28 FPS, showing higher robustness against environmental noise and lower computational overhead compared with existing methods. This low-power, edge-based solution is highly suitable for applications like sterile medical control and patient monitoring, advancing contactless interaction in healthcare by addressing efficiency and robustness challenges in radar sensing for edge computing. Full article

In recent years, deep learning-based multi-focus image fusion (MFF) methods have demonstrated remarkable performance. However, their reliance on complex network architectures often demands substantial computational resources, limiting practical applications. To address this, we propose LightMFF, an ultra-lightweight fusion network that achieves superior performance with minimal computational overhead. Our core insight is to reformulate the multi-focus fusion problem from a classification perspective to a refinement perspective, where coarse initial decision maps and explicit edge information are leveraged to guide the final decision map generation. This novel formulation enables a significantly simplified architecture, requiring only 0.02 M parameters while maintaining state-of-the-art fusion quality. Extensive experiments demonstrate that LightMFF achieves real-time performance at 0.02 s per image pair with merely 0.06 G FLOPs, representing a 98.05% reduction in computational cost compared to prior approaches. Crucially, LightMFF consistently surpasses existing methods across standard fusion quality metrics. Full article

This study investigates the mechanical properties and optimization of hybrid composites composed of flax, vetiver, and mahogany fruit fillers (MFFs) using epoxy resin as the matrix material. Nine distinct composite configurations were fabricated using different MFF concentrations (0, 5, and 10 wt.%) to evaluate their influence on tensile strength, flexural strength, and impact resistance. The MFF was subjected to alkali treatment and characterized using FTIR, XRD, and particle size analysis to enhance its compatibility with the polymer matrix. Vetiver and flax fibers also underwent alkali treatment to improve interfacial bonding. The composite fabrication process followed the Taguchi L9 orthogonal array to optimize the design. Mechanical testing revealed that the incorporation of MFF significantly improved the overall performance, with FVM9 (10 wt.% MFF) exhibiting the highest tensile strength (56.32 MPa), flexural strength (89.65 MPa), and impact resistance (10.46 kJ/m²). The CRITIC–EDAS method was employed to rank the composite configurations, and FVM9 was identified as the optimal configuration. Comparisons with alternative MCDM methods (WASPAS, COPRAS, TOPSIS, and VIKOR) validated the reliability of the rankings, and FVM9 consistently performed the best. The sensitivity analysis demonstrated the robustness of the CRITIC–EDAS approach, as the rankings remained stable despite variations in the criterion weights. The synergistic effect of flax, vetiver, and MFF, along with improved interfacial bonding, contributed to the superior mechanical properties of the hybrid composites. These findings highlight the potential of FVM composites as sustainable, high-performance materials for various industrial applications in the automotive, construction, and aerospace sectors. Full article

With the widespread adoption of encryption technologies, encrypted traffic classification has become essential for maintaining network security awareness and optimizing service quality. However, existing deep learning-based methods often rely on fixed-length truncation during preprocessing, which can lead to the loss of critical information and degraded classification performance. To address this issue, we propose a Multi-Feature Fusion (MFF) model that learns robust representations of encrypted traffic through a dual-path feature extraction architecture. The temporal modeling branch incorporates a Squeeze-and-Excitation (SE) attention mechanism into ResNet18 to dynamically emphasize salient temporal patterns. Meanwhile, the global statistical feature branch uses an autoencoder for the nonlinear dimensionality reduction and semantic reconstruction of 52-dimensional statistical features, effectively preserving high-level semantic information of traffic interactions. MFF integrates both feature types to achieve feature enhancement and construct a more robust representation, thereby improving classification accuracy and generalization. In addition, SHAP-based interpretability analysis further validates the model’s decision-making process and reliability. Experimental results show that MFF achieves classification accuracies of 99.61% and 99.99% on the ISCX VPN-nonVPN and USTC-TFC datasets, respectively, outperforming mainstream baselines. Full article

Myocardial infarction (MI) is a critical cardiovascular disease characterized by extensive myocardial necrosis occurring within a short timeframe. Traditional MI detection and localization techniques predominantly utilize single-domain features as input. However, relying solely on single-domain features of the electrocardiogram (ECG) proves challenging for accurate MI detection and localization due to the inability of these features to fully capture the complexity and variability in cardiac electrical activity. To address this, we propose a multi-domain feature fusion convolutional neural network (MFF–CNN) that integrates the time domain, frequency domain, and time-frequency domain features of ECG for automatic MI detection and localization. Initially, we generate 2D frequency domain and time-frequency domain images to combine with single-dimensional time domain features, forming multi-domain input features to overcome the limitations inherent in single-domain approaches. Subsequently, we introduce a novel MFF–CNN comprising a 1D CNN and two 2D CNNs for multi-domain feature learning and MI detection and localization. The experimental results demonstrate that in rigorous inter-patient validation, our method achieves 99.98% detection accuracy and 84.86% localization accuracy. This represents a 3.43% absolute improvement in detection and a 16.97% enhancement in localization over state-of-the-art methods. We believe that our approach will greatly benefit future research on cardiovascular disease. Full article

Background/Objectives: In 4-octylphenol (4-OP), a toxic environmental pollutant with endocrine disruptive effect, the use of 4-OP causes pollution in the freshwater environment and poses risks to aquatic organisms. Common carps (Cyprinus carpio L.) live in freshwater and are experimental animals for studying the toxic effects of environmental pollutants on fish. Its heart is susceptible to toxicants. However, whether 4-OP has a toxic effect on common carp heart remains unknown. Methods: Here, we conducted a common carp 4-OP exposure experiment (carp treated with 17 μg/L 4-OP for 45 days), aiming to investigate whether 4-OP has a toxic effect on common carp hearts. We observed the microstructure and ultrastructure of carp heart and detected autophagy genes, mitochondrial fission genes, mitochondrial fusion genes, glycolytic enzymes, AMPK, ATPase, and oxidative stress factors, to investigate the molecular mechanism of 4-OP induced damage in common carp hearts. Results: Our results showed that 4-OP exposure caused mitochondrial damage, autophagy, and damage in common carp hearts. 4-OP exposure increased the levels of miR-144, and eight autophagy factors (Beclin1, RB1CC1, ULK1, LC3-I, LC3-II, ATG5, ATG12, and ATG13), and decreased the levels of four autophagy factors (PI3K, AKT, mTOR, and SQSTM1). Furthermore, 4-OP exposure induced the imbalance between mitochondrial fission and fusion and mitochondrial dynamics imbalance, as demonstrated by the increase in three mitochondrial fission factors (Mff, Drp1, and Fis1) and the decrease in three mitochondrial fusion factors (Mfn1, Mfn2, and Opa1). Moreover, excess 4-OP treatment caused energy metabolism disorder, as demonstrated by the reduction in four ATPase (Na⁺K⁺-ATPase, Ca²⁺Mg²⁺-ATPase, Ca²⁺-ATPase, and Mg²⁺-ATPase), elevation in four glycolysis genes (HK1, HK2, LDHA, and PGK1), reduction in glycolysis gen (PGAM2), and the elevation in energy-sensing AMPK. Finally, 4-OP treatment induced the imbalance between antioxidant and oxidant and oxidative stress, as demonstrated by the increase in oxidant H₂O₂, and the decreases in five antioxidant factors (CAT, SOD, T-AOC, Nrf2, and HO-1). Conclusions: miR-144 mediated autophagy by targeting PI3K, mTOR, and SQSTM1, and the miR-144/PI3K-AKT-mTOR/ULK1 pathway was involved in 4-OP-induced autophagy. Mff-Drp1 axis took part in 4-OP-caused mitochondrial dynamics imbalance, and mitochondrial dynamics imbalance mediated autophagy via Mfn2-SQSTM1, Mfn2/Beclin1, and Mff-LC3-II axes. Energy metabolism disorder mediated mitochondrial dynamics imbalance through the AMPK-Mff-Drp1 pathway. Oxidative stress mediated energy metabolism disorder via the H₂O₂-AMPK axis. Taken together, oxidative stress triggered energy metabolism disorder, induced mitochondrial dynamics imbalance, and caused autophagy via the H₂O₂-AMPK-Mff-LC3-II pathway. Our study provided references for the toxic effects of endocrine disruptor on common carp hearts, and provided a basis for assessing environmental pollutant-induced damage in common carp heart. We only studied the toxic effects of 4-OP on common carp, and the toxic effects of 4-OP on other fish species need to be further studied. Full article

Investigating the functional interactions between rumen microbial fermentation and epithelial mitochondrial dynamics/energy metabolism in Tibetan sheep at different altitudes, this study examined ultrastructural changes in rumen epithelial tissues, expression levels of mitochondrial dynamics-related genes (fusion: Mfn1, Mfn2, OPA1, Mic60; fission: Drp1, Fis1, MFF), and ketogenesis pathway genes (HMGS2, HMGCL) in Tibetan sheep raised at three altitudes (TS 2500m, TS 3500m, TS 4500m). Correlation analysis was performed between rumen microbiota/metabolites and mitochondrial energy metabolism. Results: Ultrastructural variations were observed across altitudes. With increasing altitude, keratinized layer became more compact; desmosome connections between granular layer cells increased; mitochondrial quantity and distribution in spinous and basal layers increased. Mitochondrial dynamics regulation: Fission genes (FIS1, DRP1, MFF) showed significantly higher expression at TS 4500m (p < 0.01); fusion genes (Mfn1, OPA1) exhibited altitude-dependent upregulation. Energy metabolism markers: Pyruvate (PA) decreased significantly at TS 3500m/TS 4500m (p < 0.01); citrate (CA) increased with altitude; NAD⁺ peaked at TS 3500m but decreased significantly at TS 4500m (p < 0.01); Complex II (SDH) and Complex IV (CO) activities decreased at TS 4500m (p < 0.01). Ketogenesis pathway: β-hydroxybutyrate increased significantly with altitude (p < 0.01); acetoacetate peaked at TS 2500 m/TS 4500 m; HMGCS2 expression exceeded HMGCL, showing altitude-dependent upregulation at TS 4500m (p < 0.01). Microbiome–metabolism correlations: Butyrivibrio_2 and Fibrobacter negatively correlated with Mic60 (p < 0.01); Ruminococcaceae_NK4A214_Group positively correlated with Mfn1/OPA1 (p < 0.05); WGCNA identified 17 metabolite modules, with MEturquoise module positively correlated with DRP1/Mfn2/MFF (p < 0.05). Conclusion: Altitude-induced ultrastructural adaptations in rumen epithelium correlate with mitochondrial dynamics stability and ketogenesis upregulation. Mitochondrial fission predominates at extreme altitudes, while microbiota–metabolite interactions suggest compensatory energy regulation mechanisms. Full article

The accurate segmentation of small target buildings in high-resolution remote sensing images remains challenging due to two critical issues: (1) small target buildings often occupy few pixels in complex backgrounds, leading to frequent background confusion, and (2) significant intra-class variance complicates feature representation compared to conventional semantic segmentation tasks. To address these challenges, we propose a novel Multi-Scale Feature Fusion and Foreground Perception Enhancement Network (MFFP-Net). This framework introduces three key innovations: (1) a Multi-Scale Feature Fusion (MFF) module that hierarchically aggregates shallow features through cross-level connections to enhance fine-grained detail preservation, (2) a Foreground Perception Enhancement (FPE) module that establishes pixel-wise affinity relationships within foreground regions to mitigate intra-class variance effects, and (3) a Dual-Path Attention (DPA) mechanism combining parallel global and local attention pathways to jointly capture structural details and long-range contextual dependencies. Experimental results demonstrate that the IoU of the proposed method achieves improvements of 0.44%, 0.98% and 0.61% compared to mainstream state-of-the-art methods on the WHU Building, Massachusetts Building, and Inria Aerial Image Labeling datasets, respectively, validating its effectiveness in handling small targets and intra-class variance while maintaining robustness in complex scenarios. Full article

The ability to survive starvation is a critical evolutionary adaptation, yet the molecular mechanisms underlying this capability remain incompletely understood. Pore-forming proteins (PFPs) are typically associated with immune defense, where they disturb the membranes of target cells. However, the role of PFPs in non-immune functions, particularly in metabolic and structural adaptations to starvation, is less explored. Here, we investigate the aerolysin-like PFP LIN-24 in Caenorhabditis elegans and uncover its novel function in enhancing starvation resistance. We found that LIN-24 expression is upregulated during starvation, leading to increased expression of the lipase-encoding gene lipl-3. This upregulation accelerates the mobilization and degradation of lipid stores, thereby sustaining energy levels. Additionally, LIN-24 overexpression significantly preserves muscle integrity, as evidenced by the maintenance of muscle structure compared to wild-type worms. Furthermore, we demonstrate that LIN-24 induces the formation of donut-shaped mitochondria, a structural change likely aimed at reducing ATP production to conserve energy during prolonged nutrient deprivation. This mitochondrial remodeling depends on genes involved in mitochondrial dynamics, including mff-1, mff-2, drp-1, and clk-1. Collectively, these findings expand our understanding of PFPs, demonstrating their multifaceted role in stress resistance beyond immune defense. LIN-24’s involvement in regulating metabolism, preserving muscle structure, and remodeling mitochondria highlights its crucial role in the adaptive response to starvation, offering novel insights into the evolution of stress resistance mechanisms and potential therapeutic targets for conditions related to muscle preservation and metabolic regulation. Full article

Neurotrophic keratopathy (NK) is a degenerative corneal disease characterized by impaired corneal sensitivity and epithelial repair that is often linked to sensory nerve dysfunction. To establish a clinically relevant model and explore the mechanisms underlying NK pathogenesis, we developed a novel mouse model through partial transection of the ciliary nerve. This approach mimics the progressive nature of NK, reproducing key clinical features such as corneal epithelial defects, reduced sensitivity, diminished tear secretion, and delayed wound healing. Using this model, we investigated how disruptions in mitochondrial dynamics contribute to corneal epithelial dysfunction and impaired repair in NK. Our findings revealed substantial disruptions in mitochondrial dynamics, including reduced expression of fusion proteins (OPA1), downregulation of fission regulators (FIS1 and MFF), and impaired mitochondrial transport, as evidenced by decreased expression of Rhot1 and Kif5b. Additionally, the downregulation of mitophagy-related genes (Pink1 and Prkn) contributed to the accumulation of dysfunctional mitochondria, leading to DNA damage and impaired corneal epithelial repair. These mitochondrial abnormalities were accompanied by increased γH2AX staining, indicative of DNA double-strand breaks and cellular stress. This study highlights the pivotal role of mitochondrial dynamics in corneal epithelial health and repair, suggesting that therapeutic strategies aimed at restoring mitochondrial function, enhancing mitophagy, and mitigating oxidative stress may offer promising avenues for treating NK. Full article

Addressing the common issues of low brightness, poor contrast, and blurred details in images captured under conditions such as night, backlight, and adverse weather, we propose a zero-reference dual-path network based on multi-scale depth curve estimation for low-light image enhancement. Utilizing a no-reference loss function, the enhancement of low-light images is converted into depth curve estimation, with three curves fitted to enhance the dark details of the image: a brightness adjustment curve (LE-curve), a contrast enhancement curve (CE-curve), and a multi-scale feature fusion curve (MF-curve). Initially, we introduce the TCE-L and TCE-C modules to improve image brightness and enhance image contrast, respectively. Subsequently, we design a multi-scale feature fusion (MFF) module that integrates the original and enhanced images at multiple scales in the HSV color space based on the brightness distribution characteristics of low-light images, yielding an optimally enhanced image that avoids overexposure and color distortion. We compare our proposed method against ten other advanced algorithms based on multiple datasets, including LOL, DICM, MEF, NPE, and ExDark, that encompass complex illumination variations. Experimental results demonstrate that the proposed algorithm adapts better to the characteristics of images captured in low-light environments, producing enhanced images with sharp contrast, rich details, and preserved color authenticity, while effectively mitigating the issue of overexposure. Full article

Fires play a significant role in shaping the Miombo woodlands. Understanding how fire affects the Miombo region’s resilience is crucial for ensuring its sustainability. This study evaluated plant composition and structure across different fire frequencies in the Miombo woodlands of the LevasFlor Forest Concession (LFC), central Mozambique. Fire frequency clusters-high (HFF), moderate (MFF), and low (LFF)-were identified using a 21-year remote-sensing dataset. In each cluster, 90 random sampling plots were established (30 per cluster). In each plot, the diameter at breast height (DBH) and total height of the saplings and trees were measured. Subplots were used to count and identify seedlings, herbs, climbers, and grasses. Plant species richness, evenness,—diversity, the importance value index (IVI), and similarity were computed to assess plant composition. For the structure, stem density, biomass, basal area, diameter, and height were assessed. A total of 124 plant species-including trees, saplings, seedlings, herbs, climbers, and grasses-were identified across the three clusters. The Bray-Curtis Dissimilarity Index, tested with an ANOSIM similarity test, revealed significant differences in plant species composition among clusters (p < 0.0003), with an overall average dissimilarity of 71.98%. In the HFF cluster, fire-tolerant species were among the five species with the highest IVI, while fire-sensitive species predominated in the LFF. Additionally, the Kruskal-Wallis test indicated significant differences in seedling stem density (p < 0.005) between the LFF and other clusters. However, overall, the composition and structure attributes suggested that current fire regime does not significantly compromise the plant species resilience of the Miombo woodlands in the LFC. Still, it is essential to concentrate management and conservation efforts on seedlings of some key Miombo species, such as Brachystegia spiciformis, whose ecology is particularly affected by fire. Full article