Search Results (686)

As global food demand continues to rise, conventional agricultural practices face increasing difficulty in sustainably meeting production requirements. In response, deep learning-driven automated systems have emerged as promising solutions for enhancing precision farming. Nevertheless, accurate fruit segmentation remains a significant challenge in orchard environments due to factors such as occlusion, background clutter, and varying lighting conditions. This study proposes the Depthwise Asymmetric Bottleneck with Attention Mechanism Network (DABAMNet), an advanced convolutional neural network (CNN) architecture composed of multiple Depthwise Asymmetric Bottleneck Units (DABou), specifically designed to improve apple segmentation in RGB imagery. The model incorporates the Convolutional Block Attention Module (CBAM), a dual attention mechanism that enhances channel and spatial feature discrimination by adaptively emphasizing salient information while suppressing irrelevant content. Furthermore, the CBAM attention module employs multiple global pooling strategies to enrich feature representation across varying spatial resolutions. Through comprehensive ablation studies, the optimal configuration was identified as early CBAM placement after DABou unit 5, using a reduction ratio of 2 and combined global max-min pooling, which significantly improved segmentation accuracy. DABAMNet achieved an accuracy of 0.9813 and an Intersection over Union (IoU) of 0.7291, outperforming four state-of-the-art CNN benchmarks. These results demonstrate the model’s robustness in complex agricultural scenes and its potential for real-time deployment in fruit detection and harvesting systems. Overall, these findings underscore the value of attention-based architectures for agricultural image segmentation and pave the way for broader applications in sustainable crop monitoring systems. Full article

This paper investigates the error rate performance of simultaneous wireless information and power transfer (SWIPT) systems employing opportunistic amplify-and-forward (OAF) relaying under Rayleigh fading conditions. To support both data forwarding and energy harvesting at relays, a power splitting (PS) mechanism is applied. We derive exact and asymptotic symbol error rate (SER) expressions using moment-generating function (MGF) methods, providing analytical insights into how the power splitting ratio $\rho$ and the quality of source–relay (SR) and relay–destination (RD) links jointly affect system behavior. Additionally, we propose a novel approximation that interprets the SWIPT-OAF configuration as an equivalent non-SWIPT OAF model. This enables tractable performance analysis while preserving key diversity characteristics. The framework is extended to include scenarios with partial channel state information (CSI) and $N_{\mathrm{th}}$ best relay selection, addressing practical concerns such as limited relay availability and imperfect decision-making. Extensive simulations validate the theoretical analysis and demonstrate the robustness of the proposed approach under a wide range of signal-to-noise ratio (SNR) and channel conditions. These findings contribute to a flexible and scalable design strategy for SWIPT-OAF relay systems, making them suitable for deployment in emerging wireless sensor and internet of things (IoT) networks. Full article

The increasing frequency and intensity of wildfires pose serious threats to ecosystems, property, and human safety worldwide. Accurate semantic segmentation of wildfire images is essential for real-time fire monitoring, spread prediction, and disaster response. However, existing deep learning methods heavily rely on large volumes of pixel-level annotated data, which are difficult and costly to obtain in real-world wildfire scenarios due to complex environments and urgent time constraints. To address this challenge, we propose a semi-supervised wildfire image segmentation framework that enhances segmentation performance under limited annotation conditions by integrating multi-scale structural information fusion and pixel-level contrastive consistency learning. Specifically, a Lagrange Interpolation Module (LIM) is designed to construct structured interpolation representations between multi-scale feature maps during the decoding stage, enabling effective fusion of spatial details and semantic information, and improving the model’s ability to capture flame boundaries and complex textures. Meanwhile, a Pixel Contrast Consistency (PCC) mechanism is introduced to establish pixel-level semantic constraints between CutMix and Flip augmented views, guiding the model to learn consistent intra-class and discriminative inter-class feature representations, thereby reducing the reliance on large labeled datasets. Extensive experiments on two public wildfire image datasets, Flame and D-Fire, demonstrate that our method consistently outperforms other approaches under various annotation ratios. For example, with only half of the labeled data, our model achieves 5.0% and 6.4% mIoU improvements on the Flame and D-Fire datasets, respectively, compared to the baseline. This work provides technical support for efficient wildfire perception and response in practical applications. Full article

This paper proposes a low-complexity signal detection method for orthogonal time frequency space (OTFS) communication systems, based on a separable convolutional neural network (SeCNN), termed SeCNN-OTFS. A novel SeparableBlock architecture is introduced, which integrates residual connections and a channel attention mechanism to enhance feature discrimination and training stability under high Doppler conditions. By decomposing standard convolutions into depthwise and pointwise operations, the model achieves a substantial reduction in computational complexity. To validate its effectiveness, simulations are conducted under a standard OTFS configuration with 64-QAM modulation, comparing the proposed SeCNN-OTFS with conventional CNN-based models and classical linear estimators, such as least squares (LS) and minimum mean square error (MMSE). The results show that SeCNN-OTFS consistently outperforms LS and MMSE, and when the signal-to-noise ratio (SNR) exceeds 12.5 dB, its bit error rate (BER) performance becomes nearly identical to that of 2D-CNN. Notably, SeCNN-OTFS requires only 19% of the parameters compared to 2D-CNN, making it highly suitable for resource-constrained environments such as satellite and IoT communication systems. For scenarios where higher accuracy is required and computational resources are sufficient, the CNN-OTFS model—with conventional convolutional layers replacing the separable convolutional layers—can be adopted as a more precise alternative. Full article

Modern cloud-based Internet of Things (IoT) infrastructures face increasingly sophisticated and diverse cyber threats that challenge traditional detection systems in terms of scalability, adaptability, and explainability. In this paper, we present (H-DIR)², a hybrid entropy-based framework designed to detect and mitigate anomalies in large-scale heterogeneous networks. The framework combines Shannon entropy analysis with Associated Random Neural Networks (ARNNs) and integrates semantic reasoning through RDF/SPARQL, all embedded within a distributed Apache Spark 3.5.0 pipeline. We validate (H-DIR)² across three critical attack scenarios—SYN Flood (TCP), DAO-DIO (RPL), and NTP amplification (UDP)—using real-world datasets. The system achieves a mean detection latency of 247 ms and an AUC of 0.978 for SYN floods. For DAO-DIO manipulations, it increases the packet delivery ratio from 81.2% to 96.4% (p < 0.01), and for NTP amplification, it reduces the peak load by 88%. The framework achieves vertical scalability across millions of endpoints and horizontal scalability on datasets exceeding 10 TB. All code, datasets, and Docker images are provided to ensure full reproducibility. By coupling adaptive neural inference with semantic explainability, (H-DIR)² offers a transparent and scalable solution for cloud–IoT cybersecurity, establishing a robust baseline for future developments in edge-aware and zero-day threat detection. Full article

Automated surface defect detection in steel manufacturing is pivotal for ensuring product quality, yet it remains an open challenge owing to the extreme heterogeneity of defect morphologies—ranging from hairline cracks and microscopic pores to elongated scratches and shallow dents. Existing approaches, whether classical vision pipelines or recent deep-learning paradigms, struggle to simultaneously satisfy the stringent demands of industrial scenarios: high accuracy on sub-millimeter flaws, insensitivity to texture-rich backgrounds, and real-time throughput on resource-constrained hardware. Although contemporary detectors have narrowed the gap, they still exhibit pronounced sensitivity–robustness trade-offs, particularly in the presence of scale-varying defects and cluttered surfaces. To address these limitations, we introduce MBY (MBDNet-Attention-YOLO), a lightweight yet powerful framework that synergistically couples the MBDNet backbone with the YOLO detection head. Specifically, the backbone embeds three novel components: (1) HGStem, a hierarchical stem block that enriches low-level representations while suppressing redundant activations; (2) Dynamic Align Fusion (DAF), an adaptive cross-scale fusion mechanism that dynamically re-weights feature contributions according to defect saliency; and (3) C2f-DWR, a depth-wise residual variant that progressively expands receptive fields without incurring prohibitive computational costs. Building upon this enriched feature hierarchy, the neck employs our proposed MultiSEAM module—a cascaded squeeze-and-excitation attention mechanism operating at multiple granularities—to harmonize fine-grained and semantic cues, thereby amplifying weak defect signals against complex textures. Finally, we integrate the Inner-SIoU loss, which refines the geometric alignment between predicted and ground-truth boxes by jointly optimizing center distance, aspect ratio consistency, and IoU overlap, leading to faster convergence and tighter localization. Extensive experiments on two publicly available steel-defect benchmarks—NEU-DET and PVEL-AD—demonstrate the superiority of MBY. Without bells and whistles, our model achieves 85.8% mAP@0.5 on NEU-DET and 75.9% mAP@0.5 on PVEL-AD, surpassing the best-reported results by significant margins while maintaining real-time inference on an NVIDIA Jetson Xavier. Ablation studies corroborate the complementary roles of each component, underscoring MBY’s robustness across defect scales and surface conditions. These results suggest that MBY strikes an appealing balance between accuracy, efficiency, and deployability, offering a pragmatic solution for next-generation industrial quality-control systems. Full article

Wildfires have become increasingly frequent and destructive environmental hazards, especially in boreal ecosystems facing prolonged droughts and temperature extremes. This study presents an integrated spatio-temporal framework that combines Spatio-Temporal Density-Based Spatial Clustering of Applications with Noise (ST-DBSCAN), Fire Radiative Power (FRP), and the differenced Normalized Burn Ratio (ΔNBR) to characterize the dynamics and ecological impacts of large-scale wildfires, using the extreme 2023 Quebec fire season as a case study. The analysis of 80,228 VIIRS fire detections resulted in 19 distinct clusters across four fire zones. Validation against the National Burned Area Composite (NBAC) showed high spatial agreement in densely burned areas, with Intersection over Union (IoU) scores reaching 62.6%. Gaussian Process Regression (GPR) revealed significant non-linear relationships between FRP and key fire behavior metrics. Higher mean FRP was associated with both longer durations and greater burn severity. While FRP was also linked to faster spread rates, this relationship varied by zone. Notably, Fire Zone 2 exhibited the most severe ecological impact, with 83.8% of the area classified as high-severity burn. These findings demonstrate the value of integrating spatial clustering, radiative intensity, and post-fire vegetation damage into a unified analytical framework. Unlike traditional methods, this approach enables scalable, hypothesis-driven assessment of fire behavior, supporting improved fire management, ecosystem recovery planning, and climate resilience efforts in fire-prone regions. Full article

This study investigates indoor and outdoor air quality monitoring in a student dormitory located in northern Milan (Italy) using low-cost sensors. This research compares two monitoring periods in June and October 2024 to examine common PM_2.5 vertical patterns and differences at the building level, as well as their influence on the indoor spaces at the corresponding positions. In each period, around 30 sensors were installed at various heights and orientations across indoor and outdoor spots for 2 weeks to capture spatial variations around the building. Meanwhile, qualitative surveys on occupation presence, satisfaction, and well-being were distributed in selected rooms. The analysis of PM_2.5 data reveals that the building’s lower floors tended to have slightly higher outdoor PM_2.5 concentrations, while the upper floors generally had lower PM_2.5 indoor/outdoor (I/O) ratios, with the top-floor rooms often below 1. High outdoor humidity reduced PM infiltration, but when outdoor PM fell below 20 µg/m³ in these two periods, indoor sources became dominant, especially on the lower floors. Air pressure I/O differences had minimal impact on PM_2.5 I/O ratios, though slightly positive indoor pressure might help prevent indoor PM infiltration. Lower ventilation in Period-2 possibly contributed to more reported symptoms, especially in rooms with higher PM from shared kitchens. While outdoor air quality affects IAQ, occupant behavior—especially window opening and ventilation management—remains crucial in minimizing indoor pollutants. Users can also manage exposure by ventilating at night based on comfort and avoiding periods of high outdoor PM. Full article

This study presents a compact and high-efficiency microstrip antenna integrated with a square electromagnetic band-gap (EBG) structure for radio frequency energy harvesting to power battery-less Internet of Things (IoT) sensors and medical devices in the 5.8 GHz Industrial, Scientific, and Medical (ISM) band. The proposed antenna features a compact design with reduced physical dimensions of 36 × 40 mm² (0.69λ_o × 0.76λ_o) while providing high-performance parameters such as a reflection coefficient of −27.9 dB, a voltage standing wave ratio (VSWR) of 1.08, a gain of 7.91 dBi, directivity of 8.1 dBi, a bandwidth of 188 MHz, and radiation efficiency of 95.5%. Incorporating EBG cells suppresses surface waves, enhances gain, and optimizes impedance matching through 50 Ω inset feeding. The simulated and measured results of the designed antenna show a high correlation. This study demonstrates a robust and promising solution for high-performance wireless systems requiring a compact size and energy-efficient operation. Full article

Background/Objective: Accurately identifying women at high risk for preterm birth among those with threatened preterm labor (PTL) is crucial for effective interventions or tocolytic management to reduce preterm birth and its complications. This study aimed to determine the predictive value of cervical elastography for preterm delivery before 37 weeks of gestation in patients with threatened PTL and a cervical length greater than 15 mm. Methods: This prospective cohort study included pregnant women presenting with threatened PTL at between 24 and 34 weeks gestation. All participants underwent cervical elastography at diagnosis. We compared cervical elastography parameters between women who delivered spontaneously preterm (<37 weeks) and those who delivered at full term and assessed the ability of these parameters to predict spontaneous preterm delivery. Results: Among the 107 enrolled individuals with threatened PTL and a cervical length of ≥15 mm, 55 (42%) experienced preterm birth (<37 weeks). Internal os stiffness (IOS), internal-to-external os stiffness ratio (IOS/EOS ratio), and elasticity contrast index (ECI) were significantly associated with a risk of preterm birth compared to full-term birth. The IOS/EOS ratio was associated with 10-fold higher odds of preterm birth at <37 weeks (95% confidence interval [CI], 1.82–59.98), and ECI was associated with 1.5-fold higher odds (95% CI, 1.01–2.37). The IOS/EOS ratio demonstrated good predictive value (area under the curve (AUC) = 0.678) and the combination of CL ≤ 25 mm and the IOS/EOS ratio had good diagnostic performance for predicting preterm birth (AUC = 0.708). Conclusions: Cervical elastography using the E-Cervix™ system appears to improve the ability to predict preterm birth in pregnant women with threatened PTL and a cervical length greater than 15 mm. Full article

The quality of foundation pile is greatly influenced by human factors, and quality assessment is delayed. This paper introduces a new evaluation system based on Internet of Things (IoT) monitoring data of the foundation pile construction process. First, an IoT monitoring system of foundation pile construction process quality is established to monitor the key parameters for quality control in the foundation pile construction process, such as pile length, position, verticality, water–cement ratio, grouting volume, drilling/lifting speed, etc. Next, the absolute gray relational degree analysis method and the analytic hierarchy process (AHP) entropy-weighted combination weighting method are used to divide the monitoring data into different levels and determine the weight coefficients for quality indicators during foundation pile construction. Last, the IoT monitoring and evaluation system of the foundation piles construction process quality is applied to engineering. The results indicate that the monitoring system is convenient and efficient, and the quality evaluation method is reliable. The construction process quality of cement-mixing piles is rated as excellent. The construction process quality of bored piles Z0103 and Z0232 is excellent, and pile Z0012 is qualified. Full article

The extensive utilization of the Snappy compression algorithm in digital devices such as smartphones, IoT, and digital cameras has played a crucial role in alleviating demands on network bandwidth and storage space. This paper presents an improved Snappy compression algorithm optimized for implementation on field programmable gate arrays (FPGAs). The proposed algorithm enhances the compression ratio by refining the encoding format of Snappy and introduces an innovative approach utilizing fingerprints within the dictionary to minimize storage space requirements. Additionally, the algorithm incorporates a pipeline structure to optimize performance. Experimental results demonstrate that the proposed algorithm achieves a throughput of 1.6 GB/s for eight hardware kernels. The average compression ratio is 2.27, representing a 6.1% improvement over the state-of-the-art Snappy FPGA implementation. Notably, the proposed algorithm architecture consumes fewer on-chip storage resources compared to other advanced algorithms, striking a balance between logic and storage resource utilization. This optimization leads to higher FPGA resource utilization efficiency. Our design addresses the growing demand for efficient lossless data compression solutions in consumer electronics, ultimately contributing to advancements in modern digital ecosystems. Full article

According to the World Health Organization (WHO), indoor air quality (IAQ) is becoming a serious global concern due to its significant impact on human health. However, not all relevant health parameters are currently regulated. For example, particle number concentration (PNC) and its associated carbonaceous species, such as black carbon (BC), which are classified as carcinogenic by the International Agency for Research on Cancer (IARC), are not currently regulated. Compared with IAQ studies in other types of buildings, studies focusing on IAQ in hospitals or other healthcare facilities are scarce. Therefore, this study aims to evaluate the impact of these outdoor pollutants, among others, on the indoor environment of a hospital under different atmospheric conditions. To identify the seasonal influence, two different periods of two consecutive seasons (summer 2020 and winter 2021) were selected for the measurements. Regulated pollutants (NO, NO₂, O₃, PM₁₀, and PM_2.5) and nonregulated pollutants (PM₁, PNC, and equivalent BC (eBC)) in outdoor air were simultaneously measured indoor and outdoor. This study also investigated the impact of indoor activities on indoor air quality. In the absence of indoor activities, outdoor sources significantly contribute to indoor traffic-related pollutants. Indoor and outdoor (I-O) measurements showed similar behavior, but indoor concentrations were lower, with peak levels delayed by up to two hours. Seasonal variations in indoor/outdoor ($\mathrm{I}/\mathrm{O}$) ratios were lower for particles than for associated gaseous pollutants. Particle infiltration depended on particle size, with it being higher the smaller the particle size. Indoor activities also significantly affected indoor pollutants. PM_x (especially PM₁₀ and PM_2.5) concentrations were mainly modulated by walking-induced particle resuspension. Vertical eBC profiles indicated a relatively well-mixed environment. Ventilation through open windows rapidly altered indoor air quality. Outdoor-dominant pollutants (PNC, eBC, and NO_X) had $\mathrm{I}/\mathrm{O}$ ratios ≥ 1. Staying in the room with an open window had a synergistic effect, increasing the $\mathrm{I}/\mathrm{O}$ ratios for all pollutants. Higher I/O ratios were associated with turbulent outdoor conditions in both unoccupied and occupied conditions. Statistically significant differences were observed between stable (TKE ≤ 1 m² s⁻²) and unstable (TKE > 1 m² s⁻²) conditions, except for NO₂ in summer. This finding was particularly significant when the wind direction was westerly or easterly during unstable conditions. The results of this study highlight the importance of understanding the behavior of indoor particulate matter and related pollutants. These pollutants are highly variable, and knowledge about them is crucial for determining their health effects, particularly in public buildings such as hospitals, where information on IAQ is often limited. More measurement data is particularly important for further research into I-O transport mechanisms, which are essential for developing preventive measures and improving IAQ. Full article

Data over sound (DoS) is an established technique that has experienced a resurgence in recent years, finding applications in areas such as contactless payments, device pairing, authentication, presence detection, toys, and offline data transfer. This study introduces CardiaWhisper, a system that extends the DoS concept to the medical domain by using a medical data-over-sound (MDoS) framework. CardiaWhisper integrates wearable biomedical sensors with home care systems, edge or IoT gateways, and telemedical networks or cloud platforms. Using a transmitter device, vital signs such as ECG (electrocardiogram) signals, PPG (photoplethysmogram) signals, RR (respiratory rate), and ACC (acceleration/movement) are sensed, conditioned, encoded, and acoustically transmitted to a nearby receiver—typically a smartphone, tablet, or other gadget—and can be further relayed to edge and cloud infrastructures. As a case study, this paper presents the real-time transmission and processing of ECG signals. The transmitter integrates an ECG sensing module, an encoder (either a PLL-based FM modulator chip or a microcontroller), and a sound emitter in the form of a standard piezoelectric speaker. The receiver, in the form of a mobile phone, tablet, or desktop computer, captures the acoustic signal via its built-in microphone and executes software routines to decode the data. It then enables a range of control and visualization functions for both local and remote users. Emphasis is placed on describing the system architecture and its key components, as well as the software methodologies used for signal decoding on the receiver side, where several algorithms are implemented using open-source, platform-independent technologies, such as JavaScript, HTML, and CSS. While the main focus is on the transmission of analog data, digital data transmission is also illustrated. The CardiaWhisper system is evaluated across several performance parameters, including functionality, complexity, speed, noise immunity, power consumption, range, and cost-efficiency. Quantitative measurements of the signal-to-noise ratio (SNR) were performed in various realistic indoor scenarios, including different distances, obstacles, and noise environments. Preliminary results are presented, along with a discussion of design challenges, limitations, and feasible applications. Our experience demonstrates that CardiaWhisper provides a low-power, eco-friendly alternative to traditional RF or Bluetooth-based medical wearables in various applications. Full article

Terahertz (THz) communications and simultaneous wireless information and power transfer (SWIPT) hold the potential to energize battery-less Internet-of-Things (IoT) devices while enabling multi-gigabit data transmission. However, severe path loss, blockages, and rectifier nonlinearity significantly hinder both throughput and harvested energy. Additionally, high-power THz beams pose safety concerns by potentially exceeding specific absorption rate (SAR) limits. We propose a sensing-adaptive power-focusing (APF) framework in which a reconfigurable intelligent surface (RIS) embeds low-rate THz sensors. Real-time backscatter measurements construct a spatial map used for the joint optimisation of (i) RIS phase configurations, (ii) multi-tone SWIPT waveforms, and (iii) nonlinear power-splitting ratios. A weighted MMSE inner loop maximizes the data rate, while an outer alternating optimisation applies semidefinite relaxation to enforce passive-element constraints and SAR compliance. Full-stack simulations at 0.3 THz with 20 GHz bandwidth and up to 256 RIS elements show that APF (i) improves the rate–energy Pareto frontier by 30–75% over recent adaptive baselines; (ii) achieves a 150% gain in harvested energy and a 440 Mbps peak per-user rate; (iii) reduces energy-efficiency variance by half while maintaining a Jain fairness index of 0.999;; and (iv) caps SAR at 1.6 W/kg, which is 20% below the IEEE C95.1 safety threshold. The algorithm converges in seven iterations and executes within <3 ms on a Cortex-A78 processor, ensuring compliance with real-time 6G control budgets. The proposed architecture supports sustainable THz-powered networks for smart factories, digital-twin logistics, wire-free extended reality (XR), and low-maintenance structural health monitors, combining high-capacity communication, safe wireless power transfer, and carbon-aware operation for future 6G cyber–physical systems. Full article