Search Results (36)

Overlapping cells in plant suspension-culture microscopy pose a particular challenge, for instance, segmentation because a single pixel may belong to more than one cell. Most standard instance-segmentation methods are not designed for this setting and tend to treat overlapping objects as mutually exclusive regions. We instead represent each cell as an independent full-cell instance and introduce MC-SlotNet, an architecture that separates competitive object-slot feature assignment from mask decoding. This allows multiple predicted masks to occupy the same image region. We further introduce a mask-level multiplicity-consistency loss that encourages the predicted number of masks covering a pixel to agree with the underlying cell occupancy. We evaluate MC-SlotNet on a newly annotated dataset of 53 Siraitia grosvenorii suspension-culture micrographs containing 4131 full-cell instances acquired at 4×–40× magnification. Using grouped five-fold cross-validation and an overlap-preserving evaluation protocol, we compare the method with Mask R-CNN, SOLOv2, and Mask2Former. MC-SlotNet achieves the best performance on AP50 (0.800), mAP$50:95$ (0.565), F150 (0.842), all-ground-truth Dice (0.761), AJI+ (0.754), overlap-region Dice (0.705), and overlap-instance recall (0.828). Its AP75 (0.649) is comparable to Mask2Former’s (0.651). MC-SlotNet also has the lowest inference time among the evaluated methods, at 1.113 s/image. These results indicate that decoding full-cell masks independently, rather than enforcing an exclusive partition of image pixels, is well-suited to instance segmentation in plant suspension-culture microscopy images with substantial cell overlap. Full article

In multi-hop UAV ad hoc networks employing the Statistical Priority-based Multiple Access (SPMA) protocol, the HELLO broadcast interval cannot be arbitrarily shortened due to the inherent upper bound on the per-slot transmission probability of each node in saturated networks, which fundamentally limits the estimation accuracy of Channel Occupancy Statistics (COS). To address this problem, this paper proposes a spatial correlation-aided multi-source asynchronous Kalman filtering method, abbreviated as SMA-KF. On the basis of conventional COS broadcasting, SMA-KF introduces two complementary observation sources: COS measurements piggybacked on data packets, and spatially correlated observations from common neighbors compensated by historical biases. These three types of observations are integrated into a unified Kalman filtering framework, and a state-space model suitable for asynchronous intermittent observations is constructed. Theoretical analysis verifies the convergence of the algorithm. Simulation results demonstrate that the proposed algorithm significantly outperforms the EWMA and TW algorithms across all test scenarios, and achieves overall lower error than BiLSTM. Under the extremely sparse observation condition with a HELLO broadcast interval of 600 slots, the Normalized Root Mean Square Error (NRMSE) of SMA-KF is 28.28%, which is 32.8% and 26.2% lower than those of EWMA (42.11%) and TW (38.33%), respectively. In the heavy-traffic scenario with an average data packet arrival interval of 20 slots, the NRMSE of SMA-KF is as low as 4.80%, whereas those of EWMA and TW are 17.49% and 15.40%, respectively, corresponding to reductions of 72.6% and 68.8%. In comparison with BiLSTM, SMA-KF achieves lower NRMSE in five out of seven traffic configurations, while BiLSTM exhibits only marginal and statistically insignificant advantages in the remaining two configurations. Link interruption experiments show that SMA-KF maintains NRMSE between 5.68% and 10.40% across the entire meaningful interruption coverage range of 0% to 53%, consistently outperforming all benchmark algorithms. Moreover, SMA-KF consistently achieves the lowest estimation error under varying node mobility speeds. Parameter sensitivity analysis confirms that SMA-KF maintains stable performance across a wide range of parameter values. These results validate the effectiveness of multi-source observation fusion and spatial cooperative estimation in improving both the accuracy and robustness of COS estimation. Full article

Efficient parking space management in urban settings represents a growing challenge owing to the sustained increase in the vehicle fleet. This study presents a comparative evaluation of five object detection architectures —YOLOv8s, YOLOv11s, YOLOv12s, RT-DETR-L and Faster R-CNN—applied to real-time intelligent vehicle occupancy monitoring under variable lighting conditions. The models were trained via transfer learning on a custom dataset of 1463 source images (21,944 annotated instances; expanded to 3511 files and 52,664 instances through offline augmentation of the training subset; three classes: free, occupied and unavailable) captured on a university campus located in Juliaca (Puno region), Peru, at 3824 m a.s.l. under daytime and nighttime clear-sky conditions from a single fixed-camera viewpoint. Each architecture was evaluated in ten independent experiments. Six dataset partitioning schemes of increasing strictness—a random control (R0) plus five leakage-controlled partitions—were evaluated. Under the strictest scheme (D3), simultaneously disjoint in acquisition date and camera viewpoint and therefore the most rigorous generalization estimate obtained in this study, accuracy ranges from mAP@0.5:0.95 of 0.9325 for Faster R-CNN to 0.8763 for YOLOv11s. Under the random partitioning conventionally applied to fixed-camera datasets, the same five architectures fell within 0.0055 of one another, all above 0.985, and their ranking was essentially inverted (Spearman $\rho =-0.80$). The differences in computational efficiency across architectures were statistically significant ($H=47.06$, $p<0.001$). YOLOv8s was the fastest of the four non-dominated architectures under the disjoint partition and was selected in 73.3% of weightings, although it ranked fourth in accuracy; its recommendation therefore rests on computational efficiency under a real-time constraint, whereas deployments that prioritize accuracy are better served by Faster R-CNN. The integrated system YOLOv8s + ByteTrack + FastAPI + Next.js 14 achieved per-slot accuracies of 87.5% and 91.8% under daytime and nighttime clear-sky conditions, respectively, using 1395 observations collected in a single university parking lot. For YOLOv8s, the transition from random to disjoint partitioning costs 0.1085 in mAP@0.5:0.95 (0.9913 to 0.8828), indicating that the near-saturated performance obtained under random partitioning substantially reflects the memorization of a fixed spatial configuration rather than generalization. The results support the feasibility of single-stage CNN architectures for intelligent parking monitoring in high-altitude Andean university environments under the evaluated acquisition conditions. Full article

This paper focuses on the binary task of parking-slot occupancy detection (vacant vs. occupied) from UAV aerial imagery. Accurate parking-slot detection from UAV imagery is challenged by small target sizes, highly regular rectangular shapes, large-scale variations, complex backgrounds, and perspective distortions. Compared with fixed surveillance cameras, UAV-based detection offers flexible deployment, wide-area coverage, and no requirement for pre-installed infrastructure, making it especially suitable for large open-air parking lots and temporary parking scenarios. To address this, this study proposes a task-specific framework for UAV-based parking-slot detection, with improvements in backbone design, attention modeling, and bounding-box regression. A lightweight MnasNet-inspired backbone is used to improve multi-scale feature extraction at low computational cost. An enhanced EMA module with adaptive grouping, FFT-based frequency enhancement, and gated fusion is introduced to better model the structured patterns of parking lot scenes. In addition, a UIoU+ loss tailored to rectangular geometry is proposed to improve localization quality. Sensitivity analysis and repeated experiments show that the method is stable and statistically reliable. All main metrics are evaluated on an independent held-out test set to ensure generalization. Extensive experiments demonstrate that each component brings consistent performance gains. The proposed model achieves 99.44 ± 0.12% mAP@0.5, 90.31 ± 0.27% mAP@0.5:0.95, 99.27 ± 0.15% precision, and 99.00 ± 0.18% recall on the self-built UAV Parking Lot dataset. Its mAP@0.5:0.95 is 26.01 percentage points higher than the YOLOv11n baseline. Consistent performance improvements are also validated on two additional public benchmarks (CARPK and PKLot), confirming the generalization of the proposed method beyond the self-built dataset. Most importantly, the method supports real-time inference on embedded UAV platforms and achieves state-of-the-art performance among lightweight detectors, making it an ideal solution for practical intelligent parking management. Ablation studies further confirm the complementary synergy between the proposed backbone, attention module, and loss function. Full implementation code, pre-trained weights, and detailed reproduction guidelines are publicly available to ensure research reproducibility. Full article

This paper proposes a time-based no-cost category of energy conservation measures (ECMs) enabled by audit-driven building energy modeling. The study presents an audit-to-simulation framework applied to an academic building (Electrical Engineering Department, Assiut University, Egypt) following the audit levels and requirements of ASHRAE Standard 100-2024. The building operation is characterized via audit findings, high-resolution electrical monitoring, and occupancy profiling, then translated into a calibrated building energy model (BEM) developed using SketchUp, OpenStudio, and EnergyPlus. The validated BEM serves as a decision-support testbed to evaluate the proposed ECMs prior to implementation, enabling quantification of their impacts on annual and daily energy use, peak reduction, and load-profile shape. The proposed ECMs are classified into two subcategories: working-day ECMs and time-slot-modification ECMs. The first category involves adjusting the number of working days per week. The second category includes several scheduling-based strategies, namely seasonal time shifts, modification of lecture and tutorial session durations, rearrangement of lectures and tutorial sessions, and shifting peak-demand time slots. The simulation results show that modifying lecture and tutorial durations (ECM3) is the most effective measure, achieving 6.2% annual energy savings, followed by seasonal time shifts (ECM2) with 5.8%. For peak demand, reducing operation during peak periods (ECM5) lowers the daily peak load by 25.9%. The combined implementation of the proposed ECMs reduces annual energy consumption by up to 16% and daily peak demand by 29.4%. The findings highlight the substantial potential of structured audit-informed operational strategies in university buildings, emphasizing their role as low-risk, high-impact interventions for peak management and energy performance enhancement. Full article

This paper presents an intelligent UAV-based parking occupancy monitoring system using a lightweight DJI Mini 3 UAV platform and the YOLOv11 object-detection model. A proprietary aerial dataset was collected from a university parking lot and augmented to address data scarcity, defining two task-oriented classes: vehicle and parking. The proposed framework integrates UAV data acquisition, annotation, data augmentation, training, real-time inference, and occupancy computation into a deployable end-to-end pipeline. Experimental results demonstrate strong detection performance and stable real-time inference, achieving competitive precision, recall, and mAP (mean Average Precision) metrics while maintaining high frame rates suitable for real-time deployment. Comparative evaluation against YOLOv8 and YOLOv9 highlights deployment-oriented advantages rather than architectural novelty. The study confirms that UAV-based vision systems can provide a scalable, low-infrastructure solution for real-time parking monitoring and urban mobility applications, contributing an applied, system-level framework focused on integration and deployment feasibility. Full article

This paper presents a Genetic Algorithm-based Home Energy Management System designed to exploit the energy flexibility of user-dependent loads by identifying and recommending optimal operating schedules that minimize electricity costs. To determine the most advantageous 15 min activation slot for the following day for each load, the algorithm uses as input the forecasted consumption profile of non-optimizable loads and photovoltaic generation, both obtained through an LSTM-based model, along with the contracted power, applicable tariffs, and the load profiles of the selected appliances. Unlike previous approaches, the proposed framework allows users to select which loads to optimize and define specific operational constraints. Additionally, a user-friendly interface was developed to facilitate seamless interaction between the user and the system. To validate the proposed framework, a case study was conducted on a residential household with four occupants located in Portugal, considering user-dependent flexible loads such as a washing machine, tumble dryer, and dishwasher. The results demonstrated that the developed system operated effectively, reducing electricity costs by approximately 9% compared to a scenario without the proposed solution. Full article

This study proposes an edge-centric outdoor parking management system that performs on-site inference on a low-power embedded device and outputs slot-level occupancy decisions in real time. A dataset comprising 13,691 images was constructed using two cameras capturing frames every 3–5 s under diverse weather and illumination conditions, and a YOLOv8-based detector was trained for vehicle recognition. Beyond raw detections, a temporal occupancy decision module is introduced to map detections to predefined slot regions of interest (ROIs) while applying temporal smoothing and occlusion-robust rules, thereby improving stability under rainy and nighttime conditions. When deployed on an AI-BOX edge platform, the proposed system achieves end-to-end latency p50/p95 of 195 ms and 400 ms, respectively, while sustaining 10 FPS at 3.35 W (2.99 FPS/W) during continuous 24-hour operation. Compared with conventional sensor-based architectures, the proposed design significantly reduces upfront deployment costs and recurring maintenance requirements. Furthermore, when integrated with dynamic pricing mechanisms, it enables accurate and automated fee calculation based on real-time occupancy data. Overall, the results demonstrate that the proposed approach provides a flexible, scalable, and cost-efficient foundation for next-generation smart parking infrastructure. Full article

By increasing the number of vehicles, an intelligent parking system can help drivers in finding parking slots by providing real-time information. To address this issue, this study developed an Arduino-based automated parking system integrating sensors to assist drivers in quickly discovering available parking slots with real-time space detection and dynamic access control. This system consists of ultrasonic sensors, NodeMCU, an LCD screen, a servo motor, and an Arduino Uno. Each ultrasonic sensor is assigned a specific number corresponding to its slot number, which helps to identify the locations. These sensors were connected to the NodeMCU to collect, process, and transfer data to the Arduino board. If the ultrasonic sensor cannot detect the vehicle in the parking space, the LCD screen will show the number of specific slots. The Arduino will use the servo motor to open the entrance gate if a vehicle is detected by another ultrasonic sensor next to it. Otherwise, the system prevents any vehicle from entering the parking area when all of the available spaces are occupied. The system prototype is constructed and empirically evaluated to verify its performance and efficiency. The results indicate that the system successfully monitors parking spot occupancy and validates its capacity for real-time information updates. Full article

The efficient management of urban parking systems has emerged as a pivotal issue in today’s smart cities, where increasing vehicle populations strain limited parking infrastructure and challenge sustainable urban mobility. Aligned with the United Nations 2030 Agenda for Sustainable Development and the strategic goals of smart city planning, this study presents a sustainability-driven, multiagent simulation-based framework to model, analyze, and optimize smart parking dynamics in congested urban settings. The system architecture integrates ground-level IoT sensors installed in parking spaces, enabling real-time occupancy detection and communication with a centralized system using low-power wide-area communication protocols (LPWAN). This study introduces an intelligent parking guidance mechanism that dynamically directs drivers to the nearest available slots based on location, historical traffic flow, and predicted availability. To manage real-time data flow, the framework incorporates message queuing telemetry transport (MQTT) protocols and edge processing units for low-latency updates. A predictive algorithm, combining spatial data, usage patterns, and time-series forecasting, supports decision-making for future slot allocation and dynamic pricing policies. Field simulations, calibrated with sensor data in a representative high-density urban district, assess system performance under peak and off-peak conditions. A comparative evaluation against traditional first-come-first-served and static parking systems highlights significant gains: average parking search time is reduced by 42%, vehicular congestion near parking zones declines by 35%, and emissions from circling vehicles drop by 27%. The system also improves user satisfaction by enabling mobile app-based reservation and payment options. These findings contribute to broader sustainability goals by supporting efficient land use, reducing environmental impacts, and enhancing urban livability—key dimensions emphasized in sustainable smart city strategies. The proposed framework offers a scalable, interdisciplinary solution for urban planners and policymakers striving to design inclusive, resilient, and environmentally responsible urban mobility systems. Full article

Improving the fuel economy of electric buses requires traction motors that provide not only high-power density but also high efficiency under diverse driving conditions. While high slot fill factor (HSFF) coils such as the maximum slot occupation (MSO) coil improve motor torque and power density, they inevitably increase AC copper losses due to elevated AC resistance, especially at high speeds. Unlike conventional motor optimization studies that mainly focus on efficiency at specific operating points, this paper proposes a drive-cycle-aware design optimization method that minimizes AC copper loss to enhance real-world fuel economy. By combining 2D finite element analysis (FEA) with vehicle-level simulations under three representative driving cycles (Manhattan, HWFET, HDUDDS), an optimal motor design was derived. The optimized motor achieved improvements in fuel economy by 0.20%, 0.86%, and 0.36%, respectively, compared to the initial design. Experimental validation through prototype fabrication confirmed the effectiveness of the proposed method. These results demonstrate that the proposed design approach can contribute to improving energy efficiency and reducing operational costs in electric bus applications. Full article

To address the inadequacies of traditional Advanced Orbiting Systems (AOS) multiplexing algorithms in accommodating the networked and diverse transmission demands of space data, this paper proposes an efficient network AOS integrated multiplexing algorithm based on elastic time slots. The AOS network traffic is categorized into three types based on its characteristics, and a strongly scalable AOS integrated multiplexing model is established, which consists of a packet multiplexing layer, a virtual channel multiplexing layer, and a decision-making layer. For synchronous services, an isochronous frame generation algorithm and a periodic polling virtual channel scheduling algorithm are employed to meet the periodic transmission requirements. For asynchronous non-real-time services, a high-efficiency frame generation algorithm and a uniform queue length virtual channel scheduling algorithm are utilized to satisfy the high-efficiency transmission requirements. For asynchronous real-time services, an adaptive frame generation algorithm based on traffic prediction and a virtual channel scheduling algorithm based on comprehensive channel state are proposed. These algorithms optimize frame generation efficiency and dynamically calculate optimal scheduling results based on virtual channel scheduling status, transmission frame scheduling status, virtual channel priority status, and traffic prediction status, thereby meeting the high dynamics, low latency, and high efficiency transmission requirements. Additionally, a slot preemption-based elastic time slot scheduling strategy is proposed at the decision layer, which dynamically adjusts and optimizes the time slot allocation for the three types of traffic based on the current service request status and time slot occupancy status. Simulation results show that the proposed algorithm not only achieves lower average delay, fewer frame residuals, and higher transmission efficiency, but also maintains high stability under different working conditions, effectively meeting the transmission requirements of various types of space network traffic. Full article

The maritime industry is undergoing a paradigm shift driven by rapid advancements in wireless communication and an increase in maritime traffic data. However, the existing automatic identification system (AIS) struggles to accommodate the increasing maritime traffic data, leading to the introduction of the very-high-frequency (VHF) data exchange system (VDES). While the VDES increases bandwidth and data rates, ensuring the stable transmission of maritime IoT (MIoT) application data in congested coastal areas remains a challenge due to frequent collisions of AIS messages. This paper presents a slot occupancy-based collision avoidance algorithm (SOCA) for a VDES network in the MIoT. SOCA is designed to mitigate the impact of interference caused by transmissions of AIS messages on transmissions of VDE-Terrestrial (VDE-TER) data in coastal areas. To this end, SOCA provides four steps: (1) construction of the neighbor information table (NIT) and VDES frame maps, (2) construction of the candidate slot list, (3) TDMA channel selection, and (4) slot selection for collision avoidance. SOCA operates by constructing the NIT based on AIS messages to estimate the transmission intervals of AIS messages and updating VDES frame maps upon receiving VDES messages to monitor slot usage dynamically. After that, it generates a candidate slot list for VDE-TER channels, classifying the slots into interference and non-interference categories. SOCA then selects a TDMA channel that minimizes AIS interference and allocates slots with low expected occupancy probabilities to avoid collisions. To evaluate the performance of SOCA, we conducted experimental simulations under static and dynamic ship scenarios. In the static ship scenario, SOCA outperforms the existing VDES, achieving improvements of 13.58% in aggregate throughput, 11.50% in average latency, 33.60% in collision ratio, and 22.64% in packet delivery ratio. Similarly, in the dynamic ship scenario, SOCA demonstrates improvements of 7.30%, 11.99%, 39.27%, and 11.82% in the same metrics, respectively. Full article

Finding a parking space is a major concern in megacities, causing drivers to lose time and money while contributing to atmospheric pollution and global warming. This study proposes a method to predict parking slot availability by comparing Long Short-Term Memory (LSTM) and Multi-Head Attention (MHA) methods using the CityPulse Smart City Datasets. The initial experiments assessed the impact of pollution and time features on prediction accuracy. In a subsequent experiment, the dataset was expanded by incorporating weather-related features and a broader time range while excluding pollution and time features, as informed by the initial results. Various experiments were conducted with different parameters, such as model depth and activation functions. The results demonstrated that MHA outperformed LSTM in predicting occupancy rates, achieving a Mean Absolute Error (MAE) score of 0.0589 on the extended dataset. This study marks a pioneering effort in using MHA for real-time parking occupancy prediction, showcasing significant success with fewer parameters and a smaller model size. Full article

In the 21st century, mega-constellations and interconnected satellite constellations deployed at various orbital altitudes, such as LEO, MEO, and GEO, with low Earth orbits (LEOs) being the most commonly used, have emerged as a trend, aiming to enhance the productivity and reduce the costs in space service delivery. The UNOOSA has noted the uncertainty in the exact number of satellites but conducted simulations based on a substantial sample, projecting a significant increase from the 2075 satellites recorded in orbit in 2018. This surge in the launch of mega-constellations poses profound challenges to existing international space laws, originally formulated with limited consideration for private space actors, who are increasingly engaging in space activities, particularly with the cost-effective utilization of mega-constellations. This study critically analyzes the compatibility of mega-constellations with the current international space laws by examining the applicability of mega-constellations concerning equitable access and the non-appropriation principle, addressing their potential occupation of substantial orbital spaces during activities, and analyzing whether the acquisition of orbital slot licenses violates these two principles. Following an in-depth analysis, this study proposes recommendations to amend the existing laws, aiming to resolve ambiguities and address emerging challenges. Recognizing the time-consuming process of amending international space laws, this study suggests practical recommendations for supplementary rules of the road, prompting reflection on the potential obsolescence of the current international space laws in the face of evolving space activities. Full article