Search Results (91)

In modern intensive mutton sheep farming, the high cost and limited supply of conventional feed resources necessitate the exploration of sustainable alternatives. Cotton stalks and sugar beet pulp, abundant agricultural by-products in China, have potential as ruminant feed after proper fermentation treatment, yet their systematic application in sheep production remains underinvestigated. This study evaluated the effects of replacing whole-plant corn microsilage with mixed fermented feed (cotton stalks and sugar beet pulp, 1:1 dry matter ratio) on Suffolk rams (n = 84, 4 months old). Animals were randomly assigned to four groups: control (CK, 0% replacement), MS30 (30% replacement), MS60 (60% replacement), and MS90 (90% replacement). After a 15-day adaptation, the 120-day feeding trial assessed growth performance, slaughter characteristics, meat quality, muscle metabolomics (LC-MS), and jejunal microbiota (16S rRNA sequencing). The MS60 group significantly outperformed the CK group in final body weight, carcass weight, and net weight gain (p < 0.01), slaughter rate (p < 0.05), and meat tenderness (p < 0.05). Fatty acid composition was optimized, with lower SFAs (p < 0.01) and higher MUFAs (p < 0.01). Metabolomic analysis revealed 206 differentially abundant metabolites, with significant enrichment in linoleic acid metabolism, unsaturated fatty acid biosynthesis, and primary bile acid synthesis pathways. The MS60 group exhibited significantly altered jejunal microbiota structure (p < 0.05), including increased Patescibacteria abundance (p < 0.05) and decreased Bifidobacterium (p < 0.001). Replacing 60% of whole-plant corn microsilage with cotton stalk–beet pulp mixed microsilage effectively improved production performance, meat quality, and fatty acid profiles in Suffolk rams, while modulating muscle metabolism and intestinal microbiota structure. These findings provide a practical strategy for sustainable sheep farming utilizing regional agricultural by-products. Full article

Classifying power quality disturbances (PQDs) under strong noise conditions remains challenging for existing deep learning models. These models typically separate denoising from feature extraction, often rely on attention mechanisms that operate along only a single dimension, and tend to achieve high accuracy at the cost of high complexity, which limits their performance under low signal-to-noise ratio conditions and hinders practical deployment. To address these limitations, this paper proposes AD-PDAF-Net, which organically integrates three key mechanisms through a co-design strategy. Unlike conventional methods that depend on preprocessing, an adaptive soft thresholding denoising layer is embedded into a lightweight residual network to progressively suppress noise during feature extraction, thereby unifying denoising with feature learning. A parallel dual attention module independently refines features along the channel and temporal dimensions, then adaptively fuses them using learnable weights to capture both frequency domain and temporal characteristics of disturbances. The lightweight network entry replaces aggressive downsampling with small convolutions to preserve transient details, and a bidirectional long short-term memory network (BiLSTM) efficiently captures temporal dependencies. Evaluated on a dataset of 25 disturbance categories defined in IEEE Std 1159-2019, the model achieves a classification accuracy of 97.26% and a Kappa coefficient of 97.02% under 20 dB white Gaussian noise, along with an accuracy of 98.78% under mixed noise conditions. The model has only 0.36 million parameters and a computational cost of just 1.50 GFLOPS. Through this co-design, AD-PDAF-Net achieves both high noise robustness and high classification accuracy with minimal computational overhead, offering an effective solution for time series signal recognition in resource constrained environments. Full article

Line spectrum purification is a fundamental task in underwater detection and identification tasks. A dual-input architecture based on Dense U-net is introduced to extract clean line spectra from strong interference. The U-net model features a symmetric encoder–decoder structure that accepts two-dimensional data as both input and output. The DenseBlock, a core component of DenseNets, offers greater parameter efficiency compared to conventional convolutional layers. In this paper, standard convolutional layers inside the original U-net are replaced by DenseBlocks. This model possesses two input channels, thus allowing the time–frequency feature of the interference and that of the interference–target mixture to be fed simultaneously. With supervised learning, the model is capable of eliminating the strong interference components and background noise from the superimposed spectrum, thereby producing a purified target line spectrum. Compared to traditional interference suppression methods, this approach offers higher feature accuracy and greater signal-to-interference-and-noise ratio (SINR) gain. Moreover, the model is trainable using simulation datasets and then deployed to real-world measurements, demonstrating strong generalization capabilities—a valuable property given the limited availability of labeled samples in underwater detection tasks. Being data-driven, this method operates without requiring prior assumptions about the array configuration, and consequently exhibits greater resilience to array imperfections relative to conventional model-based interference suppression techniques. Simulation and experimental results demonstrate that the proposed method achieves an output SINR improvement of more than 8 dB under low SINR conditions and exhibits significantly better robustness to array position errors than conventional methods, verifying its excellent line spectrum purification capability. Full article

Grain byproducts can serve as cost-effective alternatives to corn, but may lead to reduced production performance and increased greenhouse gas emissions. This study aimed to investigate the effects of replacing corn with the grain byproducts (wheat bran, sprayed corn bran) subjected to bacterial-enzymatic fermentation treatment or not in Hu sheep, mainly focusing on production performance, energy-nitrogen metabolism, rumen fermentation and greenhouse gas emissions. A total of fifty-four 6-month-old Hu sheep were divided into three groups, with 6 pens per group and 3 sheep per pen, and then randomly allocated to one of the three dietary groups for 60 days, i.e., a control group (CON), a group (RC) that corn was partially (~42%) replaced with grain byproducts, and a group (BF) that corn was partially replaced by fermented grain byproducts. Compared with the CON group, the RC group showed numerically lower rumen total volatile fatty acid (TVFA) concentration and its propionate proportion, nitrogen retention content (NR; −10.22%) and its retention ratio (NR/NI decreased by 4.27 percentage points, absolute reduction from 24.30% to 20.04%), corresponding to a relative decrease of 17.6%.) as well as a numerically reduced net profit (−2.18%) with a decreased feed price (−¥0.16/kg TMR). Meanwhile, the RC group showed a significant increase in the relative abundance of Methanobrevibacter (p < 0.05), accompanied by numerically higher daily methane emissions (+6.14%) and emission intensity (+4.08%), although these methane-related differences did not reach statistical significance (p > 0.05). Compared to the RC group, the BF group resulted in a numerical increase in feed price (+¥0.03/kg TMR), net profit (+27.93%), TVFA concentration, propionate proportion, NR (+28.17%), NR/NI (an increase of 5.38 percentage points), the relative abundance of Prevotella, Shuttleworthia and Succinivibrio as well as the decrease of fecal nitrogen (FN; −12.29%), daily methane emissions (−8.75%), emission intensity (−5.83%) and the relative abundance of Methanobrevibacter. In summary, replacing dietary corn by 42% with wheat bran and sprayed corn bran numerically reduced formula cost and nitrogen utilization, while increasing methane emissions and methanogens abundance, without significantly affecting growth performance. This combination led to no improvement in economic returns for fattening Hu sheep. Bacterial-enzymatic fermentation treatment of these byproducts could mitigate these drawbacks, being superior energy-nitrogen metabolism and lower greenhouse gas emissions intensity, presenting a potential strategy for cost reduction and efficiency enhancement. Further research with larger sample sizes is warranted to confirm these findings and support broader application. Full article

Although concentrate feeds supply most carbohydrates and proteins, the protein component represents the most expensive fraction. A substantial portion of concentrate protein can be replaced with more economical protein sources from forages, particularly from legumes such as pea (Pisum sativum ssp. arvense L.), combined with cereals like oat (Avena sativa L.). Mixtures of these annual legumes and cereals generate a synergistic effect, where oats contribute yield stability and energy, while peas enhance protein concentration and improve forage preservation quality. Assessing protein quality requires understanding the distribution of individual protein fractions classified according to the Cornell Net Carbohydrate and Protein System (CNCPS), which categorizes proteins from PA (non-protein nitrogen) to PC (undegradable proteins bound to lignin, tannins, or Maillard products). This study investigated the influence of pea–oat seed ratios—SR (80:20, 60:40, 40:60, and 20:80) and developmental stages—S (early flowering and pod filling) on the dynamics of protein fractions in green biomass. Results showed that soluble protein fractions (PA, PB₁) decreased during maturation due to nitrogen translocation to developing grains, while structural and undegradable fractions (PB₂, PB₃, PC) increased, particularly in mixtures with higher oat proportions. The 60:40 pea:oat ratio produced the most balanced protein profile, maximizing the proportion of moderately degradable proteins (PB₂), which are crucial for efficient microbial protein synthesis in the rumen. This ratio also optimized the synergy between legume nitrogen fixation and cereal energy supply, enhancing sustainable ruminant nutrition. Statistical analysis confirmed significant differences between growth stages and mixture compositions. Overall, pea–oat mixtures represent a key component of economically viable and ecologically sustainable forage production for ruminant livestock systems. Full article

Accurate and efficient ship detection in remote sensing imagery is a key enabler of intelligent maritime surveillance operations, supporting real-time decision-making in search and rescue, traffic management, and maritime law enforcement. However, remote ship images pose unique challenges for detection. These include densely distributed targets, complex sea-land backgrounds, large aspect ratios, diverse ship geometries, and high color similarity between ships and their surroundings. To address these issues under the computational constraints of unmanned aerial platforms, we propose GTO-YOLO11n, an enhanced YOLOv11n-based detection model tailored for efficient maritime ship sensing. First, we introduce the GatedFDConvBlock, which employs gated convolutional filtering to strengthen feature extraction for small and elongated ships while suppressing background clutter, thereby reducing missed and false detections in dense scenes. Second, we improve the C2PSA module with a dynamic multi-scale attention design, TSSABlock_DMS, to adaptively model cross-scale feature interactions and enhance robustness to complex maritime environments. Third, we replace the original detection head with OBB_ED, a parameter-sharing head that incorporates depthwise separable convolution (DSConv) and an angle prediction branch to lower model complexity while preserving high-quality localization and classification. To verify the performance of the algorithm, we were conducted on the public datasets HRSC2016, HRSC2016-MS, and ShipRSImageNet. The mAP@50 results were 95.2%, 88.3%, and 76.7%, showing improvements of 3.2%, 2.2%, and 2.6% compared to the original YOLOv11n. Full article

Brain magnetic resonance imaging (MRI) is highly susceptible to motion artifacts that degrade fine structural details and undermine quantitative analysis. Conventional U-Net-based deep learning approaches for motion artifact reduction typically operate only in the image domain and are often trained on data with simplified motion patterns, thereby limiting physical plausibility and generalization. We propose Sim-DDNet, a simulation-data-based dual-domain network that combines k-space-based motion simulation with a joint image-k-space reconstruction architecture. Motion-corrupted data were generated from T2-weighted Alzheimer’s Disease Neuroimaging Initiative brain MR scans using a k-space replacement scheme with three to five random rotational and translational events per volume, yielding 69,283 paired samples (49,852/6969/12,462 for training/validation/testing). Sim-DDNet integrates a real-valued U-Net-like image branch and a complex-valued k-space branch using cross attention, FiLM-based feature modulation, soft data consistency, and composite loss comprising L1, structural similarity index measure (SSIM), perceptual, and k-space-weighted terms. On the independent test set, Sim-DDNet achieved a peak signal-to-noise ratio of 31.05 dB, SSIM of 0.85, and gradient magnitude similarity deviation of 0.077, consistently outperforming U-Net and U-Net++ across all three metrics while producing less blurring, fewer residual ghost/streak artifacts, and reduced hallucination of non-existent structures. These results indicate that dual-domain, data-consistency-aware learning, which explicitly exploits k-space information, is a promising approach for physically plausible motion artifact correction in brain MRI. Full article

Biochar, a carbon-rich material produced through the pyrolysis of wood residues and agricultural byproducts, has carbon storage capacity and potential as a low-carbon construction material. This study predicts the compressive strength of cementitious composites in which cement is partially replaced with biochar using machine learning models. A total of 716 data samples were analyzed, including 480 experimental measurements and 236 literature-derived values. Input variables included the water-to-cement ratio (W/C), biochar content, cement, sand, aggregate, silica fume, blast furnace slag, superplasticizer, and curing conditions. Predictive performance was evaluated using Multiple Linear Regression (MLR), Elastic Net Regression (ENR), Support Vector Regression (SVR), and Gradient Boosting Machine (GBM), with GBM showing the highest accuracy. Further optimization was conducted using XGBoost, Light Gradient-Boosting Machine (LightGBM), CatBoost, and NGBoost with GridSearchCV and Optuna. LightGBM achieved the best predictive performance (mean absolute error (MAE) = 3.3258, root mean squared error (RMSE) = 4.6673, mean absolute percentage error (MAPE) = 11.19%, and R² = 0.8271). SHAP analysis identified the W/C and cement content as dominant predictors, with fresh water curing and blast furnace slag also exerting strong influence. These results support the potential of biochar as a partial cement replacement in low-carbon construction material. Full article

This study investigates the carbonation degree of reclaimed water (RW) and its potential use as mixing water for cementitious materials under controlled laboratory conditions using a simplified CO₂ injection method. To reproduce the chemical environment of actual RW, a synthetic reclaimed water (SRW) system with a cement-to-sand ratio of 8:2 was prepared and used throughout the evaluation. Thermogravimetric analysis revealed that the cementitious solids suspended in SRW exhibit high reactivity with CO₂, achieving a net CO₂ uptake of 16.8%, equivalent to 8.31 g of CO₂ sequestered per kilogram of RW. The use of untreated RW as mixing water slightly reduced flowability and increased superplasticizer demand compared with distilled water, whereas carbonation treatment of RW improved workability and mitigated the rapid initial setting typically observed with untreated RW. Notably, replacing 3% of the cement with carbonated RW solids did not cause any reduction in compressive strength, indicating that the carbonated solids can be incorporated without compromising mechanical performance. These results confirm that the CaCO₃ formed during RW carbonation remains stably retained within mortar and concrete, demonstrating the feasibility of using carbonated RW as a dual-function material—serving both as mixing water and as a medium for CO₂ sequestration. Full article

Infrared small target detection (IRSTD) is hindered by low signal-to-noise ratios, minute object scales, and strong target–background similarity. Although long-range skip fusion is exploited in SCTransNet, the global context is insufficiently captured by its convolutional encoder, and the fusion block remains vulnerable to structured clutter. To address these issues, a Mamba-enhanced framework, MixMambaNet, is proposed with three mutually reinforcing components. First, ResBlocks are replaced by a perception-aware hybrid encoder, in which local perceptual attention is coupled with mixed pixel–channel attention along multi-branch paths to emphasize weak target cues while modeling image-wide context. Second, at the bottleneck, dense pre-enhancement is integrated with a selective-scan 2D (SS2D) state-space (Mamba) core and a lightweight hybrid-attention tail, enabling linear-complexity long-range reasoning that is better suited to faint signals than quadratic self-attention. Third, the baseline fusion is substituted with a non-local Mamba aggregation module, where DASI-inspired multi-scale integration, SS2D-driven scanning, and adaptive non-local enhancement are employed to align cross-scale semantics and suppress structured noise. The resulting U-shaped network with deep supervision achieves higher accuracy and fewer false alarms at a competitive cost. Extensive evaluations on NUDT-SIRST, NUAA-SIRST, and IRSTD-1k demonstrate consistent improvements over prevailing IRSTD approaches, including SCTransNet. Full article

Large-scale transitions to eco-friendly vehicle fleets present complex capital budgeting challenges, requiring the integration of extensive operational data with financial modeling while balancing economic profitability and environmental sustainability. Traditional approaches often struggle to manage this complexity and quantify the inherent trade-offs. This study develops and applies an innovative integrated accounting-and-finance framework to evaluate the economic and environmental implications of green fleet transition projects, explicitly quantifying the trade-off between profitability and sustainability. Focusing on waste vehicle replacement of Iren Spa, a leading European multi-utility company, we employ the recently developed Split-Screen Approach, a unified accounting-and-finance framework grounded in the laws of motion and conservation. It automatically reconciles pro forma financial statements and generates internally consistent valuation metrics, eliminating the manual adjustments and inconsistencies of traditional models. Its built-in diagnostic checks and scalability for highly complex datasets overcome the manual adjustments and inconsistencies inherent in traditional financial models. We process 2303 inputs across multiple “green” scenarios. This methodology integrates an Engineering Model, describing fleet evolution, operating costs, and CO₂ reduction, with a HookUp Model, which serves to transform scenarios into well-defined projects. The latter model is then integrated with a Financial Model that generates pro forma financial statements, incorporates financing and payout policies, and assesses economic profitability through Net Present Value (NPV) and consistent accounting rates of return. Together, these elements form a robust framework for managing complex data integration and analysis. Our research reveals a fundamental trade-off: enhanced environmental sustainability (measured by Net Green Value, NGV), which quantifies CO₂ reduction, is achieved at the expense of economic profitability, measured by NPV. This financial sacrifice is captured by the Net Value Curve, a Pareto frontier, while the NPV-to-NGV ratio provides “shadow prices” for CO₂ reduction, revealing the financial cost per unit of sustainability gained. Based on 21 project scenarios and additional sensitivity analyses on financial inputs and energy prices, the results confirm a decreasing relationship between NGV and NPV. This study makes three main contributions: (1) it demonstrates the practical application of the Split- Screen Approach for capital budgeting under complexity, (2) it introduces the Net Value Curve framework as a useful tool for visualizing and quantifying the trade-off between profitability and sustainability, (3) it provides managers and policymakers actionable insights, supporting more informed decisions in green fleet transition planning where economic and environmental objectives may conflict. The findings provide managers and policymakers with a rigorous and transparent accounting-and-finance framework that enhances the reliability of capital budgeting decisions compared with traditional financial modeling, while offering a Paretian frontier for evaluating environmental trade-offs. Full article

To address the challenge of low modulation recognition accuracy for Low Probability of Intercept (LPI) radar waveforms under low Signal-to-Noise Ratio (SNR) conditions—a critical limitation in current radar signal processing research—this study proposes a novel recognition framework anchored in an improved EfficientNet model. First, to generate time–frequency images, the radar signals are initially subjected to time–frequency analysis using the Choi–Williams Distribution (CWD). Second, the Mobile Inverted Bottle-neck Convolution (MBConv) structure incorporates the Simple Attention Module (SimAM) to improve the network’s capacity to extract features from time–frequency images. Specifically, the original serial mechanism within the MBConv structure is replaced with a parallel convolution and attention approach, further optimizing feature extraction efficiency. Third, the network’s loss function is upgraded to Focal Loss. This modification aims to mitigate the issue of low recognition rates for specific radar signal types during training: by dynamically adjusting the loss weights of hard-to-recognize samples, it effectively improves the classification accuracy of challenging categories. Simulation experiments were conducted on 13 distinct types of LPI radar signals. The results demonstrate that the improved model validates the effectiveness of the proposed approach for LPI waveform modulation recognition, achieving an overall recognition accuracy of 96.48% on the test set. Full article

The study aims to identify optimum retrofitting strategies that mitigate climate change and support Australia’s net-zero emissions target by 2050. Current heating and cooling demands, as well as the energy performance of three stand-alone houses built before 2003, were evaluated to determine optimal retrofitting measures. Based on a comprehensive literature review and physical building surveys and energy simulations using FirstRate5 of three selected case studies of stand-alone houses in Australia’s climate zone 5, the study identifies and proposes effective retrofitting opportunities in Western Australia. Additionally, the outcomes from FirstRate5 illustrate that improving ceiling and exterior wall insulation in living and dining areas, sealing air leaks, reducing overshading, and replacing single-glazed windows with double-glazed units while enlarging north-facing windows, following the recommended wall–window ratio significantly improve the energy rating of the selected houses. The average energy rating performance of the three selected stand-alone houses increases from an average below 3.5 stars (211.5 MJ/m²) to above 7.5 stars (46.7 MJ/m²), representing around 76.6% improvement in energy efficiency. Just to contextualise the scale up, such retrofitting of all old stand-alone houses built before 2003 would potentially reduce emissions by 12.73 Mt CO₂-e/year, representing a 3.16% contribution toward Australia’s national emission reduction target by 2035. Additionally, installing solar energy systems could reduce an extra 4.5 Mt CO₂-e/year. The study’s findings demand robust retrofitting strategies for Australia to achieve its 2050 net-zero emissions targets. Full article

Oral cancer remains a critical global health challenge, with delayed diagnosis driving high morbidity and mortality. Despite progress in artificial intelligence, computer vision, and medical imaging, early detection tools that are accessible, explainable, and designed for patient engagement remain limited. This study presents a novel system that combines a patient-centred chatbot with a deep learning framework to support early diagnosis, symptom triage, and health education. The system integrates convolutional neural networks, class activation mapping, and natural language processing within a conversational interface. Five deep learning models were evaluated (CNN, DenseNet121, DenseNet169, DenseNet201, and InceptionV3) using two balanced public datasets. Model performance was assessed using accuracy, sensitivity, specificity, diagnostic odds ratio (DOR), and Cohen’s Kappa. InceptionV3 consistently outperformed the other models across these metrics, achieving the highest diagnostic accuracy (77.6%) and DOR (20.67), and was selected as the core engine of the chatbot’s diagnostic module. The deployed chatbot provides real-time image assessments and personalised conversational support via multilingual web and mobile platforms. By combining automated image interpretation with interactive guidance, the system promotes timely consultation and informed decision-making. It offers a prototype for a chatbot, which is scalable and serves as a low-cost solution for underserved populations and demonstrates strong potential for integration into digital health pathways. Importantly, the system is not intended to function as a formal screening tool or replace clinical diagnosis; rather, it provides preliminary guidance to encourage early medical consultation and informed health decisions. Full article

Access to clean cooking technologies is crucial for achieving SDG7 for remote households in small Pacific Islands like Fiji and for developed countries alike. Many small households in Fiji still rely on traditional biomass for cooking. This study explores the environmental sustainability and clean cooking potential of the Home Biogas 2.0 flexible balloon digester installed at Kamil Muslim College in Ba, Fiji. Comparative bench experiments were also performed. The bench-scale experiments produced higher biogas yields than the digester trials, with optimal outputs recorded from fresh cow dung (541 mL of cumulative biogas) and vegetable waste excluding rice (125 mL). When scaled, annual energy production from fresh cow dung reached 4644.64 MJ, equivalent to replacing 7.82 standard LPG cylinders, while vegetable waste produced 3763.76 MJ, offsetting 6.34 cylinders. Notably, biogas from cow dung exceeded the estimated annual household cooking demand of 3840 MJ for a family of four persons. The biogas produced from fresh cow dung provided an average cooking duration of 1 h 29 min, while biogas from vegetable waste lasted for 1 h 21 min. The economic analysis indicated that combining liquid digestate, used as biofertiliser, and biogas from cow dung resulted in the highest financial return, with a 67% Internal Rate of Return, a Net Present Value of $12,364.30, a Benefit Cost Ratio of 5.12, and a Discounted Payback Period of 1.28 years. This indicates the potential of Home Biogas 2.0 as a climate-smart technology that integrates renewable energy production, waste reduction, and sustainable agriculture, making it particularly suitable for rural and remote communities. Full article