Search Results (41,105)

This study adapts Fourier ptychography (FP) for high-resolution imaging in machine vision settings. We replace multi-angle illumination hardware with a single fixed light source and controlled object translation to enable a sequence of slightly shifted low-resolution frames to produce the requisite frequency-domain diversity for FP. The concept is validated in simulation using an embedded pupil function recovery algorithm to reconstruct a high-resolution complex field, recovering both amplitude and phase. For conveyor-belt transport, we introduce a lightweight preprocessing pipeline—background estimation, difference-based foreground detection, and morphological refinement—that yields robust masks and cropped inputs suitable for FP updates. The reconstructed images exhibit sharper fine structures and enhanced contrast relative to native lens imagery, indicating effective pupil synthesis without multi-LED arrays. The approach preserves compatibility with standard industrial optics and conveyor-style acquisition while reducing hardware complexity. We also discuss practical operating considerations, including blur-free capture and synchronization strategies. Full article

Weeding in melon and watermelon fields requires selective and pinpoint operation because the crop plants are sensitive to herbicides and tend to grow on the ground in all directions. Hyperspectral images have high spectral and spatial resolution, enabling an object’s classification according to its spectral properties. Spectral band selection is a common practice with hyperspectral images, as it reduces the number of bands in use with only a minor effect on the results. This study’s innovative contribution is the development and validation of a practical methodology to simplify complex hyperspectral data for real-world robotic weed management. This includes the introduction of the ‘normalized crop sample index’ (NCSI) to guide band selection and the use of machine learning methods, which revealed a set of four spectral bands—480 nm, 550 nm, 686 nm and 750 nm—that hold sufficient discriminating information between weeds and watermelon crop. This minimal set of bands enables the simulation and future development of a low-cost, high-speed multispectral camera system. An XGBoost model showed the lowest misclassification error level of 2–14%. The selected spectral bands were used to extract single-band images from the hyperspectral cube. In these images, vegetation pixels were separated using a normalized difference vegetation index filter, and each pixel was classified into a crop or weed class. The classified pixels were grouped into segmented objects, and weeding points were selected, suitable for robotic pinpoint operation. Full article

The objective of this work was to assess the association of ACC (1-aminocyclopropane-1-carboxylic acid), S-ABA (abscisic acid), and ethephon on color development and anthocyanin accumulation in berries, as well as on other quality attributes of ‘Benitaka’ and ‘Rubi’ table grapes grown in a subtropical region, in addition to postharvest conservation of clusters and vine regrowth. As a statistical model, a randomized block design consisting of nine treatments and four replications was used. The treatments included different associations of ACC, S-ABA, and ethephon, by using the commercial formulations Accede^®, ProTone^®, and Ethrel^® containing 400 g kg⁻¹ of ACC, 100 g L⁻¹ of S-ABA, and 720 g L⁻¹ of ethephon, respectively. The total anthocyanins, berry color index (CIRG), physicochemical characteristics, and cluster color coverage were assessed weekly, while berry firmness was assessed at harvest. After being harvested, the clusters were placed under cold storage at 1.0 ± 1.0 °C, and after 45 days, their postharvest attributes were assessed, as well as the vine regrowth in the following season. The exogenous and combined application of compounds at véraison was demonstrated to be a strategy to trigger the development of color in ‘Benitaka’ and ‘Rubi’ table grapes. For the ‘Benitaka’ table grape, the clusters treated with the different combinations of ACC and S-ABA, ethephon and S-ABA, or ethephon alone resulted in the highest concentration of total anthocyanins and the highest CIRG means (4.90; 4.86; 4.82; 4.81, 4.73, and 4.70 mg g⁻¹ for anthocyanins, and 6.12, 6.08, 5.97, 5.92, 5.85, and 5.74 for CIRG, respectively). For the ‘Rubi’ table grape, the combinations of ACC and S-ABA at 7 days after véraison (DAV), or ethephon and S-ABA at 7 and 14 days, resulted in higher means of anthocyanins and CIRG (3.86, 3.51, and 3.40 mg g⁻¹ for anthocyanins and 5.05, 4.68, 4.82, and 4.79 for CIRG, respectively). Furthermore, the firmness of the berries of both cultivars remained unchanged, and after 45 days of cold storage, no reduction in the quality of the evaluated postharvest attributes was found. It was concluded that a single application of ACC 0.20 g L⁻¹ + S-ABA 0.250 g L⁻¹ at 7 DAV was sufficient to promote the accumulation of anthocyanins and resulted in an intense and uniform color in the berries for both varieties assessed, with no adverse impacts on the postharvest conservation of the clusters or on the regrowth of the vines. The significance of this research was to demonstrate that table grapes with insufficient skin color can be improved through a combination of S-ABA and ACC at lower concentrations of active ingredients. Full article

Despite the superior performance of existing anomaly detection methods, they are often limited to single-class detection tasks, requiring separate models for each class. This constraint hinders their detection performance and deployment efficiency when applied to real-world multi-class data. In this paper, we propose a dynamic visual adaptation framework for multi-class anomaly detection, enabling the dynamic and adaptive capture of features based on multi-class data, thereby enhancing detection performance. Specifically, our method introduces a network plug-in, the Hyper AD Plug-in, which dynamically adjusts model parameters according to the input data to extract dynamic features. By leveraging the collaboration between the Mamba block, the CNN block, and the proposed Hyper AD Plug-in, we extract global, local, and dynamic features simultaneously. Furthermore, we incorporate the Mixture-of-Experts (MoE) module, which achieves a dynamic balance across different features through its dynamic routing mechanism and multi-expert collaboration. As a result, the proposed method achieves leading accuracy on the MVTec AD and VisA datasets, with image-level mAU-ROC scores of 98.8% and 95.1%, respectively. Full article

This paper presents an evaluation of the use of deep learning architectures for forecasting electrical energy consumption in residential environments. The main contribution of this study lies in the development and assessment of a hybrid forecasting framework that integrates multiscale temporal analysis and weather data, enabling evaluation of predictive performance across different temporal granularities, forecast horizons, and aggregation levels. Single and hybrid models were compared, trained with high-resolution data from a single residence, both considering only endogenous variables and including exogenous variables (weather data). The results showed that, among all models tested in this study, the hybrid LSTM + GRU model achieved the highest predictive performance, with R² values of 94.62% using energy data and 95.25% when weather variables were included. Intermediary granularities, particularly the 6 steps, offered the best balance between temporal detail and predictive robustness for the tests performed. Furthermore, short-time windows aggregation (1 to 5 min) showed better accuracy, while the inclusion of weather data in scenarios with larger aggregation windows and longer horizons provided additional gains. The results reinforce the potential of hybrid deep learning models as effective tools for forecasting residential electricity consumption, with possible practical applications in energy management, automation, and integration of distributed energy resources. Full article

The separable heat pipe, with its highly efficient heat transfer and flexible layout features, has become an innovative solution to the heat dissipation problem of batteries, especially suitable for the directional heat dissipation requirements of high-energy-density battery packs. However, most of the number–value models currently studied examine the flow of refrigerant working medium within the pump as an isentropic or isothermal process and are unable to effectively analyze the heat transfer characteristics of different internal regions. Based on the laws of energy conservation, momentum conservation, and mass conservation, this study establishes a steady-state mathematical model of the pump-driven microchannel-separated heat pipe. The influence of factors—such as the phase state change in the working medium inside the heat exchanger, the heat transfer flow mechanism, the liquid filling rate, the temperature difference, as well as the structural parameters of the microchannel heat exchanger on the steady-state heat transfer and flow performance of the pump-driven microchannel-separated heat pipe—were analyzed. It was found that the influence of liquid filling ratio on heat transfer quantity is reflected in the ratio of change in the sensible heat transfer and latent heat transfer. The sensible heat transfer ratio is higher when the liquid filling is too low or too high, and the two-phase heat transfer is higher when the liquid filling ratio is in the optimal range; the maximum heat transfer quantity can reach 3.79 KW. The decrease in heat transfer coefficient with tube length in the single-phase region is due to temperature and inlet effect, and the decrease in heat transfer coefficient in the two-phase region is due to the change in flow pattern and heat transfer mechanism. This technology has the advantages of long-distance heat transfer, which can adapt to the distributed heat dissipation needs of large-energy-storage power plants and help reduce the overall lifecycle cost. Full article

The current study investigates plasma-assisted chemical recycling as an innovative approach to recover valuable carbon fibers from composite waste while minimizing environmental impact. Nitrogen and argon plasma-in-bubbles are employed in a concentrated nitric acid solution, thus enhancing conventional nitric acid solvolysis with plasma chemistry. A systematic process framework is presented, revealing key operational stages, including composite pretreatment, composite solvolysis, carbon fiber recovery/characterization, NO_x recovery, nitric acid circulation, and byproduct management, demonstrating their role in the overall process efficiency and environmental impact. Moreover, the research examined different processing conditions, including plasma power, acid concentration, and reactor design, while comparing open-air systems to systems equipped with single-stage or two-stage wet scrubbers for NO_x recovery. Remarkably, recycled fibers from plasma-assisted solvolysis demonstrated preserved or even slightly enhanced mechanical properties compared to those of the virgin fibers. Recycled carbon fibers originating from the operation at 1200 W and 12 M HNO₃ demonstrated the best mechanical properties with 3138.92 MPa tensile strength and 307.02 GPa Young’s modulus. However, the parametric analysis revealed that operating the plasma reactor at 1200 W and 14 M, equipped with a two-stage scrubber, achieved optimal environmental performance. Full article

In-band full-duplex (IBFD) communication systems offer a promising means of improving spectral efficiency by enabling simultaneous transmission and reception on the same frequency channel. Despite this advantage, self-interference (SI) remains a major challenge to their practical deployment. Among the different SI cancellation (SIC) techniques, this paper focuses on digital SIC methodologies tailored for multiple-input multiple-output (MIMO) wireless transceivers operating under digital beamforming architectures. Two distinct digital SIC approaches are evaluated, employing a generalized memory polynomial (GMP) model augmented with Itô–Hermite polynomial basis functions and a phase-normalized neural network (PNN) to effectively model the nonlinearities and memory effects introduced by transmitter and receiver hardware impairments. The robustness of the SIC is further evaluated under both single off-line training and closed-loop real-time adaptation, employing estimation techniques such as least squares (LS), least mean squares (LMS), and fast Kalman (FK) for model coefficient estimation. The performance of the proposed digital SIC techniques is evaluated through detailed simulations that incorporate realistic power amplifier (PA) characteristics, channel conditions, and high-order modulation schemes. Metrics such as error vector magnitude (EVM) and total bit error rate (BER) are used to assess the quality of the received signal after SIC under different signal-to-interference ratio (SIR) and signal-to-noise ratio (SNR) conditions. The results show that, for time-variant scenarios, a low-complexity adaptive SIC can be realized using a GMP model with FK parameter estimation. However, in time-invariant scenarios, an open-loop SIC approach based on PNN offers superior performance and maintains robustness across various modulation schemes. Full article

Integrins, cell-surface adhesion receptors involved in tumor progression, invasion, and metastasis, serve as crucial biomarkers for cancer diagnosis and therapeutic targeting. Multiplexed detection of integrins and cancer cell classification at the single-cell level allows for comprehensive profiling, facilitating precise identification and categorization of tumor cells that are heterogeneous in integrin expression and cell subtype. In this study, we developed a five-plex detection platform and demonstrated integrin profile for cancer cell classification leveraging surface-enhanced Raman scattering (SERS) with gap-enhanced gold nanorods (GENRs) in conjunction with advanced computational analysis. Specifically, we synthesized GENRs bearing five distinct Raman nanotags, each producing a unique spectral fingerprint upon targeting a specific integrin subtype expressed on cancer cell surfaces. SERS signals from single cancer cells—after labeling simultaneously with the five-color SERS nanotags—were collected on single cells and subsequently analyzed with classical least squares regression to reliably deconvolute and quantify expression level of five different integrin monomers. Utilizing a random forest classifier trained on integrin profiles from individual cancer cell lines, we achieved simultaneous detections of three different breast cancer cell lines, with exceptional classification accuracy of 99.9%. The feasibility of this method for multiplexed detection of circulating tumor cells was tested using peripheral blood mononuclear cells (PBMCs) spiked with mixed breast cancer cells from three cell lines. By integrating GENRs, multiplexed SERS nanotag technology, and machine learning, our platform significantly advances cancer diagnostics through accurate integrin-based cell profiling and classification. These findings highlight the potential of multiplexed integrin detection using SERS technology as a powerful diagnostic approach, ultimately supporting improved cancer subtype characterization, personalized diagnostics, and more targeted therapeutic strategies. Full article

Monitoring respiratory activity is a key indicator of physiological health and an essential component in smart textile systems for unobtrusive vital sign assessment. In this work, we present a one-dimensional convolutional recurrent neural network (1D-CRNN) for automatic classification of breathing activity from inertial data acquired by a smart e-textile of 59 subjects. The proposed method integrates convolutional layers for local feature extraction with recurrent layers for temporal context modeling, enabling robust segmentation of breathing and noise segments. The model was trained and evaluated using a stratified five-fold cross-validation scheme to account for inter-subject variability and class imbalance. Across different window sizes, the classifier achieved a mean accuracy of 0.88 and an F1-score of 0.92 at a window size of 2000 samples. The best-performing configuration for a single fold, reached an accuracy of 0.995 and an F1-score of 0.99. Furthermore, near-real-time feasibility was demonstrated, with a total processing time—including data loading, classification, segmentation, and visualization—of only 1.76 s for a 250 s measurement, corresponding to more than 100× faster than the recording time. These results indicate that the proposed approach is highly suitable for embedded, on-device inference within wearable systems. Full article

As climate change and urbanization accelerate, urban flooding poses an increasingly severe threat to urban residents and their properties, creating an urgent need for effective solutions to achieve sustainable urban disaster management. While physically based hydrodynamic models can accurately simulate urban floods, they are data- and computational-resource-demanding. Meanwhile, artificial intelligence models driven by data often lack generalizability across different urban areas. To address these challenges, integrating spiking neural networks, graph convolutional networks (GCNs), and particle swarm optimization (PSO), a novel PSO-enhanced spiking graph convolutional neural network (P-SGCN) is proposed. The model is trained on a self-constructed dataset based on social media data, incorporating six representative Chinese cities: Beijing, Shanghai, Shenzhen, Wuhan, Hangzhou, and Shijiazhuang. These cities were selected for their diverse urban and flood characteristics to enhance model generalizability. The P-SGCN significantly outperforms baseline models such as GCN and long short-term memory, achieving an accuracy, precision, recall, and F1 score of 0.846, 0.847, 0.846, and 0.846, respectively. These results indicate our model’s capability to effectively handle data from six cities while maintaining high accuracy. Meanwhile, the model improves single-city performance through transfer learning and offers extremely fast inference with minimal energy consumption, making it suitable for real-time applications. This study provides a scalable and generalizable solution for urban flood risk management, with potential applications in disaster preparedness and urban planning across varied geographic and socioeconomic contexts. Full article

Energy efficiency presents significant potential, especially for Western Balkan (WB) countries, if effectively addressed through energy efficiency measures. The building sector, which includes residential, commercial, and public buildings, is the most energy-intensive sector globally. Public buildings in the Western Balkan countries are characterized by poor energy efficiency performance. The average energy consumption in public buildings is anticipated to exceed double the European Union (EU) requirement, given that more than 60-70% of these structures were built over 60 years ago with no regard for energy efficiency. This study assesses the Public Building–Energy Efficiency Readiness Index (PB-EERI) to evaluate how legislative specificity, institutional capacity, financing mechanisms, renovation guidelines, energy market conditions, and societal awareness collectively influence the readiness of Western Balkan economies to enhance energy efficiency in public buildings. The index serves as an operational diagnostic to identify the presence of enabling conditions, determine the most significant gaps, and prioritize policy efforts accordingly. This study presents a novel approach by integrating, within a single transparent index, (i) the existence of energy laws, (ii) market feasibility, (iii) renovation needs of public buildings, and (iv) societal awareness. The awareness pillar is both central and novel. By utilizing harmonized Regional Cooperation Council (RCC) data, this article quantifies societal awareness, thereby ensuring that the index accurately reflects the importance of stakeholder comprehension in the success of renovating initiatives for public buildings. The theoretical framework derives from the application of composite indicators in numerous studies and reports to illustrate the status of energy or energy efficiency. The methodology for developing this indicator is derived from the Organization for Economic Co-operation and Development (OECD) Handbook on Constructing Composite Indicators. For the aggregation method, the summation of weighted and normalized sub-indicators was used. The PB-EERI reveals considerable regional variations, with total scores ranging from around 39 to 72% and concentrating around the mid-0.5s. The findings reveal systematic differences in most indicators’ performance. The legal framework indicator significantly influences variation between countries, together with market conditions and societal awareness. Energy efficiency in public buildings, praised as the “first fuel”, should be prioritized beyond mere compliance with EU regulations. The PB-EERI emphasizes that success relies more on the capacity to transform formal strategies into concrete renovation programs, quantifiable objectives, and higher awareness of society to ensure uptake of the renovation measures. Full article

Energy efficiency in industrial environments is subject to regulatory and economic constraints. Automated intralogistics systems, such as High Rack Storage Systems (HRSS), exhibit complex and dynamic energy patterns. This paper proposes an unsupervised learning approach that uses Self-Organizing Maps (SOMs) to characterize operational energy states from HRSS measurements (power, voltage, and position). After preprocessing, we train an SOM and apply Watershed segmentation to obtain a topological map of states, and we analyze state transitions with a Markov model to study persistence and switching behavior. The approach yields an interpretable taxonomy of energy use and highlights operating conditions associated with different efficiency levels, as well as central states that influence system behavior. While the study focuses on a single demonstrator, the results suggest that SOM can support explainable monitoring and analysis of industrial energy behavior and may help guide proactive energy-management decisions in Industry 4.0 settings. Full article

Traditional multi-stage separated heat pipes (SHPs) face limitations in independently setting operation parameters for each stage. To address this issue, this paper presents a novel independent multi-stage microchannel Separated Gravity Heat Pipe (SGHP) for air compressor room cooling. The innovative structure and working principle of this novel multi-stage SGHP were introduced. Furthermore, numerical investigations on a single stage of the SGHP were then conducted to study the gas–liquid two-phase flow characteristics and phase-change heat transfer performance. Experimental research on a three-stage SGHP was carried out to further explore the impact of the filling ratio combinations and the temperature difference between the hot and cold ends on the heat transfer performance of the SGHP. The results show that the temperature difference between the hot and cold ends affects the flow pattern of the working fluid, which has a vital effect on the heat transfer performance of the SGHP. The optimum filling ratio combination of the three-stage SGHP depends on the temperature difference between the hot and cold ends. The optimum filling ratio combination is 37%/37%/30% at low temperature difference conditions and 43%/37%/37% at high temperature difference conditions, respectively. The highest heat transfer capacity of the three-stage SGHP reaches 15.3 kW, and the peak heat recovery efficiency is 74.0%. The findings provide a crucial foundation for developing novel independent multi-stage SGHP in compressor room cooling and similar industrial settings, promising high potential to reduce energy consumption and operational costs. Full article

Background: Some hypersensitivity pneumonitis (HP) patients exhibit autoimmune features (HPAF). This study compared outcomes of HPAF and HP without autoimmune features, focusing on progressive pulmonary fibrosis (PPF) and response to immunosuppression. Methods: A retrospective cohort study included HP patients from a single center. HPAF was defined as HP overlapping with autoimmune disease or presenting autoimmune markers/symptoms not fulfilling connective tissue disease criteria. A control HP group without autoimmune features was randomly selected. Demographics, autoimmune profiles, and outcomes over two years were analyzed. Results: 103 patients were included (52 HPAF; 51 HP). In HPAF, the most common autoimmune diseases were rheumatoid arthritis (9.6%), while 57.7% had isolated autoimmune serology. Groups showed no baseline differences in demographics, exposures, smoking, or lung function. Fibrotic disease on high-resolution CT at diagnosis was less frequent in HPAF (71.2% vs. 88.2%; p = 0.031). At two-year follow-up, survival, transplantation, and PPF prevalence were similar. HPAF patients received immunosuppression less often (69.2% vs. 86.3%; p = 0.038). Among patients under immunosuppression, PPF was significantly lower in HPAF group (8.6% vs. 29.5%; p = 0.021). Conclusion: Within two years post-diagnosis, HPAF and HP had comparable overall outcomes. However, under immunosuppression, HPAF patients had significantly lower odds of developing PPF (adjusted OR 0.08; 95% CI 0.008–0.816; p = 0.033) compared to HP patients, suggesting a more favorable treatment response. Full article