Search Results (696)

Following an earthquake, abnormal travel demand causes traffic congestion and poses significant problems for relief efforts. Research on post-earthquake travel demand is essential for disaster management. An effective disaster management strategy ensures achieving sustainable development goals. This study focused on this critical period and analyzed post-earthquake trip decisions. The city of Elazığ, a region not at risk of tsunami, was used as a case study. A 6.8 magnitude earthquake hit Elazığ in January 2020. After the earthquake, data from 2739 individuals were collected by a household survey conducted face-to-face. The data were segregated into two categories, depending on the earthquake’s intensity. The study used a binary logit model to examine the potential factors of trip decisions after an earthquake. The results showed that 75% of participants made at least one trip within 24 h after the earthquake. It was observed that household, building-and disaster-related attributes influence earthquake survivors’ trip decisions. The initial location at the time of the earthquake was the most significant factor affecting trip decisions. It was also found that individuals who experienced the earthquake outside their homes in both datasets were more likely to make a trip. Additionally, the dataset with higher earthquake intensity had more significant variables affecting the trip decision. Full article

The constant increase in multimedia Internet traffic in the form of video streaming requires new solutions for efficient video coding to save bandwidth and network resources. HTTP adaptive streaming (HAS), the most widely used solution for video streaming, allows the client to adaptively select the bitrate according to the transmission conditions. For this purpose, multiple presentations of the same video content are generated on the video server, which contains video sequences encoded at different bitrates with resolution adjustment to achieve the best Quality of Experience (QoE). This set of bitrate–resolution pairs is called a bitrate ladder. In addition to the traditional one-size-fits-all scheme for the bitrate ladder, context-aware solutions have recently been proposed that enable optimum bitrate–resolution pairs for video sequences of different complexity. However, these solutions use only spatial resolution for optimization, while the selection of the optimal combination of spatial and temporal resolution for a given bitrate has not been sufficiently investigated. This paper proposes bit-ladder optimization considering spatiotemporal features of video sequences and usage of optimal spatial and temporal resolution related to video content complexity. Optimization along two dimensions of resolution significantly increases the complexity of the problem and the approach of intensive encoding for all spatial and temporal resolutions in a wide range of bitrates, for each video sequence, is not feasible in real time. In order to reduce the level of complexity, we propose a data augmentation using a neural network (NN)-based model. To train the NN model, we used seven video sequences of different content complexity, encoded with the HEVC encoder at five different spatial resolutions (SR) up to 4K. Also, all video sequences were encoded using four frame rates up to 120 fps, presenting different temporal resolutions (TR). The Structural Similarity Index Measure (SSIM) is used as an objective video quality metric. After data augmentation, we propose NN models that estimate optimal TR and bitrate values as switching points to a higher SR. These results can be further used as input parameters for the bitrate ladder construction for video sequences of a certain complexity. Full article

Traffic-related roads are an underestimated source of synthetic particles in the environment. This study investigated the impact of traffic volume on microplastic (MP) and tire wear particle (TWP) pollution in road snow. An examination was conducted in a medium-sized city situated in northeastern Poland, known for being one of the cleanest regions in the country. MPs and TWPs were found at all 54 sites, regardless of the intensity of traffic. The average concentration for all samples was 354.72 pcs/L. Statistically significant differences were found between the average values of the particle concentration on low, medium, and heavy traffic roads, amounting to 62.32 pcs/L, 335.97 pcs/L, and 792.76 pcs/L, respectively. Within all three studied groups of roads, MPs and TWPs with the smallest size, ranging from 50 to 200 μm, were prevalent. In all of the studied groups of roads, four analyzed shapes of particles were found, with irregular fragments being the most abundant form (89.23%). The most frequently recorded color among the collected samples was black (99.85%), and the least frequently recorded color was blue, constituting only 0.01%. This study suggests that snow cover on the roads may act like a temporary storage of pollutants during winter particularly in the temperate climate zone and, after thawing can significantly increase the concentration of MPs and TWPs in surface waters. Possible measures to decrease the release of MPs and TWPs into the environment in the city may include reducing the traffic volume and speed, implementing street sweeping, utilizing filtration chambers, and installing stormwater bioretention systems or settling ponds. Full article

Accurately classifying road signs is crucial for autonomous driving due to the high stakes involved in ensuring safety and compliance. As Convolutional Neural Networks (CNNs) have largely replaced traditional Machine Learning models in this domain, the demand for substantial training data has increased. This study aims to compare the performance of classical Machine Learning (ML) models and Deep Learning (DL) models under varying amounts of training data, particularly focusing on altered signs to mimic real-world conditions. We evaluated three classical models: Support Vector Machine (SVM), Random Forest, and Linear Discriminant Analysis (LDA), and one Deep Learning model: Convolutional Neural Network (CNN). Using the German Traffic Sign Recognition Benchmark (GTSRB) dataset, which includes approximately 40,000 German traffic signs, we introduced digital alterations to simulate conditions such as environmental wear or vandalism. Additionally, the Histogram of Oriented Gradients (HOG) descriptor was used to assist classical models. Bayesian optimization and k-fold cross-validation were employed for model fine-tuning and performance assessment. Our findings reveal a threshold in training data beyond which accuracy plateaus. Classical models showed a linear performance decrease under increasing alteration, while CNNs, despite being more robust to alterations, did not significantly outperform classical models in overall accuracy. Ultimately, classical Machine Learning models demonstrated performance comparable to CNNs under certain conditions, suggesting that effective road sign classification can be achieved with less computationally intensive approaches. Full article

In the production of high-definition maps, it is necessary to achieve the three-dimensional instantiation of road furniture that is difficult to depict on traditional maps. The development of mobile laser measurement technology provides a new means for acquiring road furniture data. To address the issue of traffic marking extraction accuracy in practical production, which is affected by degradation, occlusion, and non-standard variations, this paper proposes a 3D reconstruction method based on energy functions and template matching, using zebra crossings in vehicle-mounted LiDAR point clouds as an example. First, regions of interest (RoIs) containing zebra crossings are obtained through manual selection. Candidate point sets are then obtained at fixed distances, and their neighborhood intensity features are calculated to determine the number of zebra stripes using non-maximum suppression. Next, the slice intensity feature of each zebra stripe is calculated, followed by outlier filtering to determine the optimized length. Finally, a matching template is selected, and an energy function composed of the average intensity of the point cloud within the template, the intensity information entropy, and the intensity gradient at the template boundary is constructed. The 3D reconstruction result is obtained by solving the energy function, performing mode statistics, and normalization. This method enables the complete 3D reconstruction of zebra stripes within the RoI, maintaining an average planar corner accuracy within 0.05 m and an elevation accuracy within 0.02 m. The matching and reconstruction time does not exceed 1 s, and it has been applied in practical production. Full article

Objectives: Facial fractures and associated traumatic brain injuries represent a worldwide public health concern. Therefore, we aimed to determine the pattern of brain injury accompanying facial fractures by comparing adult patients with and without facial fractures in terms of demographic, clinical, and imaging features. Methods: This single-center, retrospective study included 492 polytrauma patients presenting at our emergency department from January 2019 to July 2023, which were divided in two groups: with facial fractures (FF) and without facial fractures (non-FF). The following data were collected: age, sex, mechanism of trauma (road traffic accident, fall, and other causes), Glasgow Coma Scale (GCS), the evolution of the patient (admitted to a medical ward or intensive care unit, neurosurgery performed, death), and imaging features of the injury. Data were analyzed using descriptive tests, Chi-square tests, and regression analyses. A p-value less than 0.05 was considered statistically significant. Results: In the FF group, there were 79% (n = 102) men and 21% (n = 27) women, with a mean age of 45 ± 17 years, while in the non-FF group, there were 70% (n = 253) men and 30% (n = 110) women, with a mean age 46 ± 17 years. There was a significant association between brain injuries and facial fractures (p < 0.001, AOR 1.7). The most frequent facial fracture affected the zygoma bone in 28.1% (n = 67) cases. The most frequent brain injury associated with FF was subdural hematoma 23.4% (n = 44), and in the non-FF group, the most common head injury was intraparenchymal hematoma 29% (n = 73); Conclusions: Both groups shared similarities regarding gender, age, cause of traumatic event, and outcome but had significant differences in association with brain injuries, ICU admission, and clinical status. Full article

Aiming to reduce noise pollution and optimize the acoustic quality in highway tunnels, the noise characteristics on different types of pavements were analyzed and compared in this research, based on the on-site noise measurement in two tunnels with the free fields as a control group. Specifically, the noise characteristics include two aspects: various noise and noise time attenuation performance. Various noise includes on-board sound intensity (OBSI) noise and cabin noise. The noise time attenuation performance uses the indicator of reverberation time. Three types of pavements were measured, including dense-graded asphalt concrete (DAC) and single-layered and double-layered porous asphalt (PA) pavement. The results showed that, for the same type of pavement, compared with the free fields, the difference in OBSI noise in tunnels was within a range of less than 1 dBA; the cabin noise increased by 3.4 dBA~6.6 dBA. The noise level in tunnels was greater than that outside tunnels, and the longer tunnel exhibited higher traffic noise and worse noise time attenuation performances. For the same tunnel, PA pavement could reduce the cabin noise by 3.8 dBA~6.7 dBA. PA pavement also exhibited shorter reverberation time. The application of PA pavement could effectively improve the acoustic quality in the tunnel. This research contributes to noise pollution abatement and the improvement of the comfort and safety of drivers in tunnels. Full article

Bus lanes are commonly implemented to ensure absolute priority for buses at signalized intersections. However, while prioritizing buses, existing bus lane management strategies often exacerbate traffic demand imbalances among lanes. To address this issue, this paper proposes a dynamic Multi-Function Lane (MFL) management strategy. The proposed strategy transforms traditional bus lanes into Multi-Function Lanes (MFLs) that permit access to Connected and Automated Vehicles (CAVs). By fully utilizing the idle right-of-way of the MFL, the proposed strategy can achieve traffic efficiency improvement. To evaluate the proposed strategy, some experiments are conducted under various demand levels and CAV penetration rates. The results reveal that the proposed strategy (i) improves the traffic intensity balance degree by up to 52.9 under high demand levels; (ii) reduces delay by up to 80.56% and stops by up to 89.35% with the increase in demand level and CAV penetration rate; (iii) guarantees absolute bus priority under various demand levels and CAV penetration rates. The proposed strategy performs well even when CAV penetration is low. This indicates that the proposed strategy has the potential for real-world application. Full article

This paper addresses the growing importance of efficient street lighting management, driven by rising electricity costs and the need for municipalities to implement cost-effective solutions. Central to this study is the UNI 11248 Italian regulation, which extends the European EN 13201-1 standard introduced in 2016. These standards provide guidelines for designing, installing, operating, and maintaining lighting systems in pedestrian and vehicular traffic areas. Specifically, the UNI 11248 standard introduces the possibility to dynamically adjust light intensity through two alternative operating modes: (a) Traffic Adaptive Installation (TAI), which dims the light based solely on real-time traffic flow measurements; and (b) Full Adaptive Installation (FAI), which, in addition to traffic measurements, also requires evaluating road surface luminance and meteorological conditions. In this paper, we first present the general architecture and operation of an FAI-enabled lighting infrastructure, which relies on environmental sensors and a heterogeneous wireless communication network to connect intelligent, remotely controlled streetlights. Subsequently, we examine large-scale, in-field FAI infrastructures deployed in Vietnam and Italy as case studies, providing substantial measurement data. The paper offers insights into the measured energy consumption of these infrastructures, comparing them to that of conventional light-control strategies used in traditional installations. The measurements demonstrate the superiority of FAI as the most efficient solution. Full article

Vehicle detection is essential for road traffic surveillance and active safety management. Deep learning methods have recently shown robust feature extraction capabilities and achieved improved detection results. However, vehicle detection models often perform poorly under abnormal lighting conditions, especially in highway tunnels. We proposed an adaptive vehicle detection model that accounts for varying luminance intensities to address this issue. The model categorizes the image data into abnormal and normal luminance scenarios. We employ an improved CycleGAN with edge loss as the adaptive luminance adjustment module for abnormal luminance scenarios. This module adjusts the brightness of the images to a normal level through a generative network. Finally, YOLOv7 is utilized for vehicle detection. The experimental results demonstrate that our adaptive vehicle detection model effectively detects vehicles under abnormal luminance scenarios in highway tunnels. The improved CycleGAN can effectively mitigate edge generation distortion. Under abnormal luminance scenarios, our model achieved a 16.3% improvement in precision, a 1.7% improvement in recall, and a 9.8% improvement in mAP_0.5 compared to the original YOLOv7. Additionally, our adaptive luminance adjustment module is transferable and can enhance the detection accuracy of other vehicle detection models. Full article

The development of unmanned ground vehicles and unmanned aerial vehicles requires high-precision navigation due to the autonomous motion and higher traffic intensity. The existing L1 band GNSS receivers are a good and cheap decision for smartphones, vehicle navigation, fleet management systems, etc., but their accuracy is not good enough for many civilian purposes. At the same time, real-time kinematic (RTK) navigation allows for position precision in a sub-centimeter range, but the system cost significantly narrows this navigation to a very limited area of applications, such as geodesy. A practical solution includes the integration of dual-band GNSS receivers and inertial sensors to solve high-precision navigation tasks, but GNSS position accuracy may significantly affect IMU performance due to having a great impact on Kalman filter performance in unmanned vehicles. The estimation of dilution-of-precision (DOP) parameters is essential for the filter performance as the optimality of the estimation in the filter is closely connected to the quality of a priori information about the noise covariance matrix and measurement noise covariance. In this regard, the current paper analyzes the DOP parameters of the latest generation dual-band GNSS receivers and compares the results with the L1 ones. The study was accomplished using two types of antennas—L1/L5 band patch and wideband helix antennas, which were designed and assembled by the authors. In addition, the study is extended with a comparison of GNSS receivers from different generations but sold on the market by one of the world’s leading GNSS manufacturers. The analyses of dilution-of-precision (DOP) parameters show that the introduction of dual-band receivers may significantly increase the navigation precision in a sub-meter range, in addition to multi-constellation signal reception. The fast advances in the performance of the integrated CPU in GNSS receivers allow the number of correlations and tracking satellites to be increased from 8–10 to 24–30, which also significantly improves the position accuracy even of L1-band receivers. Full article

As air traffic intensity increases and stochastic uncertainties, such as wind direction and speed, continue to impact air traffic controllers’ workload significantly, airlines are increasingly pressured to reduce costs by flying via straighter/more direct trajectories. Due to these changes, it is important to search for new means/solutions for aircraft conflict resolution to ensure the required level of safety and rational flight trajectory. Such a solution could be the implementation of Dubin’s method of flight trajectories. This paper aims to propose and deeply analyze a new mathematical model for two-aircraft conflict resolution where the Dubins method is applied in a dynamic conflict scenario. In this model, at a certain moment, the flight trajectory of one aircraft follows a path similar to a moving circle’s tangential line. Upon that, the conflict detection and resolution (CDR) model considers wind uncertainty. The proposed CDR method could be applied when uncertainty such as wind direction and speed are inconstant (stochastic) throughout the simulation. Full article

With the continuous development of new power systems, the intelligence of distribution networks has been increasingly enhanced. However, network security issues, especially distributed denial-of-service (DDoS) attacks, pose a significant threat to the safe operation of distribution networks. This paper proposes a novel DDoS attack defense mechanism based on software-defined network (SDN) architecture, combining Rényi entropy and multi-level convolutional neural networks, and performs fine-grained analysis and screening of traffic data according to the amount of calculation to improve the accuracy of attack detection and response speed. Experimental verification shows that the proposed method excels in various metrics such as accuracy, precision, recall, and F1-score. It demonstrates significant advantages in dealing with different intensities of DDoS attacks, effectively enhancing the network security of user-side devices in power distribution networks. Full article

The electrification of mobility has introduced considerable challenges to distribution networks due to varying demand patterns in both time and location. This underscores the need for adaptable tools to support strategic investments, grid reinforcement, and infrastructure deployment. In this context, the present study employs real-world datasets to propose a comprehensive spatial–temporal energy model that integrates a traffic model and geo-referenced data to realistically evaluate the flexibility potential embedded in the light-duty transportation sector for a given study region. The methodology involves assessing traffic patterns, evaluating the grid impact of EV charging processes, and extending the analysis to flexibility services, particularly in providing primary and tertiary reserves. The analysis is geographically confined to the Lombardy region in Italy, relying on a national survey of 8.2 million trips on a typical day. Given a target EV penetration equal to 2.5%, corresponding to approximately 200,000 EVs in the region, flexibility bands for both services are calculated and economically evaluated. Within the modeled framework, power-intensive services demonstrated significant economic value, constituting over 80% of the entire potential revenues. Considering European markets, the average marginal benefit for each EV owner is in the order of 10 € per year, but revenues could be higher for sub-classes of users better fitting the network needs. Full article

In both developed and developing countries, atmospheric pollution with particulate matter (PM) remains an important issue. Despite the health effects of poor air quality, studies on air pollution are often limited by the high costs of continuous monitoring and the need for extensive sampling. Furthermore, these particles are often enriched with potentially toxic trace elements and organic pollutants. This study evaluates both the composition of atmospheric dust accumulated during a certain timespan on Hedera helix and Senecio cineraria leaves and the potential for their use as bio-monitors. The test plants were positioned near automatic air quality monitoring stations at four different sites with respectively high, moderate and low traffic intensity. The gravimetric deposition of PM10 and PM2.5 on leaves was compared with data recorded by the monitoring stations and related to the weather conditions reported by Argentina’s National Meteorological Service. To determine the presence of trace elements enriching the PM deposited on leaves, two analytical techniques were applied: XRF (not destructive) and ICP (destructive). The results indicated that only in the unpaved street location (site 2) did PM10 and PM2.5 concentrations (90 µg m⁻³ and 9 µg m⁻³) in the air exceed more than five times WHO guidelines (15 µg m⁻³ and 5 µg m⁻³). However, several trace elements were found to be enriching PM deposited on leaves from all sites. Predominantly, increased concentrations of Cd, Cu, Ti, Mn, Zn and Fe were found, which were associated with construction, traffic and unpaved street sources. Furthermore, based on its capability to sequester above 2800 µg cm⁻² of PM10, 2450 µg cm⁻² of PM2.5 and trace elements, Senecio cineraria can be taken into consideration for adoption as a bio-monitor or even for PM mitigation. Full article