Search

With the rapid economic development and high concentration of urban population, people's income level and quality of life continue to improve, resulting in more and more crowded scenes caused by people going out. Especially in urban commercial centers, transportation hubs, sports venues during important events, tourist attractions, etc., crowd gatherings occur frequently. However, accidents involving crowd gatherings in public places occur frequently, causing heavy casualties and property losses. Therefore, for crowd recognition, this paper proposes a new method to accurately estimate the number of dense crowds. In this method, a density map with accurate pedestrian locations is first generated using the focal inverse distance transform and used as ground truth labels for network training. Then, a multi-scale feature fusion algorithm based on residual network is designed, combining spatial and channel attention mechanisms to improve the accuracy and stability of crowd density estimation. In dense crowds, the phenomenon of overlapping and occlusion of people is very common and serious, making it difficult for existing pedestrian detection methods to distinguish each individual and accurately count the flow of people. To solve this problem, this paper proposes a density map-based method that uses a local maximum detection strategy and a K-nearest neighbor algorithm to convert the density map into the corresponding dense head bounding box. This method can effectively reduce the impact of occlusion and improve the accuracy of people counting. In order to further improve the estimation accuracy, a pattern recognition density peak clustering algorithm is introduced to study the clustered crowds. By treating the head bounding box as an element point, the distance between each element point is calculated, and the density of each point is calculated. Then perform clustering to find the cluster center with the highest density in each class. Finally, by comparing the density of each cluster center with the corresponding density threshold and adopting the corresponding decision-making method, the accuracy of people counting is further improved. [ABSTRACT FROM AUTHOR]

The zero-velocity update (ZUPT) algorithm is a pivotal advancement in pedestrian navigation accuracy, utilizing foot-mounted inertial sensors. Its key issue hinges on accurately identifying periods of zero-velocity during human movement. This paper introduces an innovative adaptive sliding window technique, leveraging the Fourier Transform to precisely isolate the pedestrian's gait frequency from spectral data. Building on this, the algorithm adaptively adjusts the zero-velocity detection threshold in accordance with the identified gait frequency. This adaptation significantly refines the accuracy in detecting zero-velocity intervals. Experimental evaluations reveal that this method outperforms traditional fixed-threshold approaches by enhancing precision and minimizing false positives. Experiments on single-step estimation show the adaptability of the algorithm to motion states such as slow, fast, and running. Additionally, the paper demonstrates pedestrian trajectory localization experiments under a variety of walking conditions. These tests confirm that the proposed method substantially improves the performance of the ZUPT algorithm, highlighting its potential for pedestrian navigation systems. [ABSTRACT FROM AUTHOR]

Pedestrian tracking in surveillance videos is crucial and challenging for precise personnel management. Due to the limited coverage of a single video, the integration of multiple surveillance videos is necessary in practical applications. In the realm of pedestrian management using multiple surveillance videos, continuous pedestrian tracking is quite important. However, prevailing cross-video pedestrian matching methods mainly rely on the appearance features of pedestrians, resulting in low matching accuracy and poor tracking robustness. To address these shortcomings, this paper presents a cross-video pedestrian tracking algorithm, which introduces spatial information. The proposed algorithm introduces the coordinate features of pedestrians in different videos and a linear weighting strategy focusing on the overlapping view of the tracking process. The experimental results show that, compared to traditional methods, the method in this paper improves the success rate of target pedestrian matching and enhances the robustness of continuous pedestrian tracking. This study provides a viable reference for pedestrian tracking and crowd management in video applications. [ABSTRACT FROM AUTHOR]

Walkability is generally defined by the extent to which the built environment supports and encourages walking by promoting pedestrian comfort, connectivity and convenience. This paper scrutinizes the steps taken in Johannesburg to develop walkability policies and projects. It draws on empirical evidence from the pedestrian policies (e.g., the Corridors of Freedom) and pedestrian projects (e.g., the Grayston Pedestrian Bridge) recently introduced. Rather than focusing on the merits of walkability or the ease of its application, this paper highlights the misunderstandings of introducing walkability in Johannesburg. The concept is particularly problematic in Johannesburg, where the "unpleasantness of walking" and "walkability-with-people" is insufficiently considered. In a broader sense, this discussion aims to first understand urban planning in an African/Global South city, and then to decolonize transport planning by bringing the experiences in Johannesburg to the forefront. Such a critical reading of the intertwined and overlapping practices of policymaking provides additional insights into the processes of urban planning and transport in Johannesburg and beyond. [ABSTRACT FROM AUTHOR]

In recent years, real-time vehicle congestion detection has become a hot research topic in the field of transportation due to the frequent occurrence of highway traffic jams. Vehicle congestion detection generally adopts a vehicle counting algorithm based on object detection, but it is not effective in scenarios with large changes in vehicle scale, dense vehicles, background clutter, and severe occlusion. A vehicle object counting network based on a feature pyramid split attention mechanism is proposed for accurate vehicle counting and the generation of high-quality vehicle density maps in highly congested scenarios. The network extracts rich contextual features by using blocks at different scales, and then obtains a multi-scale feature mapping in the channel direction using kernel convolution of different sizes, and uses the channel attention module at different scales separately to allow the network to focus on features at different scales to obtain an attention vector in the channel direction to reduce mis-estimation of background information. Experiments on the vehicle datasets TRANCOS, CARPK, and HS-Vehicle show that the proposed method outperforms most existing counting methods based on detection or density estimation. The relative improvement in MAE metrics is 90.5% for the CARPK dataset compared to Fast R-CNN and 73.0% for the HS-Vehicle dataset compared to CSRNet. In addition, the method is also extended to count other objects, such as pedestrians in the ShanghaiTech dataset, and the proposed method effectively reduces the misrecognition rate and achieves higher counting performance compared to the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

• Visual behaviour of distracted and non-distracted pedestrians was examined. • Post-encroachment time was used to assess risk of pedestrians-vehicle interactions. • Mobile phone usage increased chances of unsafe pedestrian-vehicle interactions. • Visual metrics parameters showed a significant impact on unsafe interactions. Use of mobile phones while crossing the road has become a growing safety concern among pedestrians. Though recently, some studies have attempted to understand risky behaviour of distracted pedestrians by analyzing impacts of distraction on pedestrians' walking and visual characteristics, the impacts were measured individually. Moreover, most of the existing findings are based on lab-based experiments, whereas the real-world conditions are quite dynamic in nature. Hence, to overcome these gaps, the current study aims to understand to what extent different potential factors (such as distraction, walking characteristics, visual behaviour, etc.) affect the risk of unsafe pedestrian-vehicle interactions in real-world settings. For this purpose, a field experiment was designed to understand the impact of distraction on pedestrians' risky behaviour using an eye-tracker at an unsignalized intersection in Delhi. The risk assessment was done using Post-Encroachment Time (PET) measure. The results indicate that the use of mobile phone while crossing the road increased the chances of unsafe (PET < 1.5 s) pedestrian-vehicle interactions by approximately eight folds. Surprisingly, pedestrians who crossed the intersection at high speed had higher chances of unsafe interactions, whereas their increased crossing initiation time reduced the probability of unsafe interactions. Moreover, increased pupil diameter and fixating more on vehicles approaching from the right side resulted in a reduced probability of unsafe interactions. However, the high frequency of saccades and higher fixations on the crosswalk's right side negatively affected unsafe interactions. This study highlights that it is important to consider visual characteristics along with demographic and walking and traffic characteristics as they can provide in-depth insights into pedestrians' road crossing behaviour that are required to design and implement effective countermeasures and policies. [ABSTRACT FROM AUTHOR]

Pedestrian self-organizing movement plays a significant role in evacuation studies and architectural design. Lane formation, a typical self-organizing phenomenon, helps pedestrian system to become more orderly, the majority of following behavior model and overtaking behavior model are imprecise and unrealistic compared with pedestrian movement in the real world. In this study, a pedestrian dynamic model considering detailed modelling of the following behavior and overtaking behavior is constructed, and a method of measuring the lane formation and pedestrian system order based on information entropy is proposed. Simulation and analysis demonstrate that the following and avoidance behaviors are important factors of lane formation. A high tendency of following results in good lane formation. Both non-selective following behavior and aggressive overtaking behavior cause the system order to decrease. The most orderly following strategy for a pedestrian is to overtake the former pedestrian whose speed is lower than approximately 70% of his own. The influence of the obstacle layout on pedestrian lane and egress efficiency is also studied with this model. The presence of a small obstacle does not obstruct the walking of pedestrians; in contrast, it may help to improve the egress efficiency by guiding the pedestrian flow and mitigating the reduction of pedestrian system orderliness. [ABSTRACT FROM AUTHOR]

This research is based on the adaptation and validation in French of the Pedestrian Receptivity Questionnaire for Fully Autonomous Vehicles (PRQF). Faced with the emergence of numerous issues related to interactions between pedestrians and fully autonomous vehicles (FAV), this questionnaire is a reliable and scientifically valid means of collecting pedestrian experiences when they cross a road in front of an autonomous vehicle. The translation methodology of the PRQF was based on a committee approach, which consisted of asking several bilingual experts to come to an agreement on a consensual translation. The validation of the French PRQF (F-PRQF) was based on the completion of 584 questionnaires. Different psychometric tests were carried out, including an accuracy analysis, a principal component factor analysis (PCA) and a confirmatory factor analysis (CFA). Our results show that the F-PQRF displays satisfactory psychometric qualities, in particular in terms of internal consistency (α = 0.932 and ω = 0.935). We also find a scale with five components, which confirms the theoretical model structure of the original scale. The F-PRQF showed good psychometric qualities, which should thus facilitate research in fully autonomous vehicles among French-speaking audiences. [ABSTRACT FROM AUTHOR]

• Child pedestrian injuries in Greater London are studied. • A host of exposure measures, built environment and socio-economic factors are considered. • School enrolment was found to be a better proxy exposure measure compared to number of schools. • Social inequities are found to be strongly associated with child pedestrian safety. Introduction: Child pedestrian safety remains a challenge despite the remarkable progress that has been attained in recent years, particularly, in high income jurisdictions such as London. This study sought to identify and quantify the magnitude of the effects of various explanatory variables, from the domains of transport, built and natural environment, socio-demographic and economic factors, on ward level child pedestrian injury frequencies in Greater London. Method: We adopted a multilevel random parameters model to investigate the factors associated with child pedestrian injuries given the hierarchical nature of the data comprising of wards nested within boroughs. Results: We found that crime, the Black, Asian, and Minority Ethnic (BAME) population, school enrollment, and the proportion of the population who walk five times a week had an increasing effect on the number of child pedestrian casualties. Conversely, the proportion of the population with a level 4 qualification and the number of cars per household had a decreasing effect. Conclusions: Our study identified high child pedestrian injury frequency wards and boroughs: Stratford and New Town had the highest expected child pedestrian injury frequencies followed by Selhurst, Westend, and Greenford Broadway. Some inner London boroughs are among the highest injury frequency areas; however, a higher number of high child pedestrian injury boroughs are in outer London. Practical Applications: The paper provides recommendations for policy makers for targeted child pedestrian safety improvement interventions and prioritization to optimize the utilization of often constrained resources. The study also highlights the importance of considering social inequities in policies that aim at improving child traffic safety. [ABSTRACT FROM AUTHOR]

The city corridor is a strategic area with its own aesthetic and functional value. In connection with the city's beauty, streetscape is one of the determining aspects in terms of architecture and spatial planning. The existence of streetscape tends to be overlooked, resulting in the area lacking identity. The presence of streetscape with character can provide a strong identity for the area, especially when the corridor in question has strategic value in terms of function and location. Proximity to the coastal area is indeed one of the advantages, although currently, it is still hidden, especially architecturally. This research aims to analyze streetscape elements with a coastal area approach. The research location is in the corridor of Piere Tendean Street, Manado, divided into 3 segments. The observed units include pedestrians, street furniture, activities, and buildings on the roadside. The conclusion is obtained that there are soft elements, hard elements, mobile elements, and natural elements. Elements closely related to the coastal area are mobile elements, such as activities enjoying the beauty of the beach equipped with breakwaters, activities in pedestrian and park locations, as well as fishing activities conducted in the empty space where the river meets the beach. [ABSTRACT FROM AUTHOR]

The pedestrian attribute recognition aims to generate a structured description of pedestrians, which serves an important role in surveillance. Current works usually assume that the images and the specific pedestrian states, including pedestrian occlusion and pedestrian orientation, are given. However, we argue that the current works ignore the guidance of the pedestrian state and cannot achieve the appropriate performance since the appearance feature will become unreliable due to the variance of the pedestrian state, which is common in practice. Therefore, this paper proposes the Explicit State Representation (ExSR) Guided Pedestrian Attribute Recognition to improve the accuracy through state learning and attribute fusion among frames. Firstly, the pedestrian state is explicitly represented by concatenating the pedestrian orientation and occlusion, which can be accurately determined via analyzing the pose. Secondly, the state-aware pedestrian attribute fusion method is proposed and divided into two cases, namely the inter-state case and the intra-state case. In the intra-state case, the appearance feature will remain stable and the attribute relations are propagated to refine. The method of exploiting attribute relations within a single frame is the Graph Neural Network. In the inter-state case, the state changes, the attribute relationship propagation is prevented, and the advantages of attribute recognition in each frame are complemented to make a reliable judgment on the invisible region. The experimental results demonstrate that the ExSR outperforms the state-of-the-art methods on two public databases, benefiting from the explicit introduction of the state into the attribute recognition. [ABSTRACT FROM AUTHOR]

Pedestrian injuries from falls are an understudied cause of morbidity. Here, we compare the burden of pedestrian injuries from falls occurring on streets and sidewalks with that from motor vehicle collisions. Data on injurious falls on streets and sidewalks, and pedestrian-motor vehicle collisions, to which Emergency Medical Services responded, along with pedestrian and incident characteristics, were identified in the 2019 National Emergency Medical Services Information System database. In total, 118,520 injurious pedestrian falls and 33,915 pedestrians-motor vehicle collisions were identified, with 89% of the incidents occurring in urban areas. Thirty-two percent of pedestrians struck by motor vehicles were coded as Emergent or Critical by Emergency Medical Services, while 19% of pedestrians injured by falls were similarly coded. However, the number of pedestrians whose acuity was coded as Emergent or Critical was 2.1 times as high for injurious falls as compared with pedestrians-motor vehicle collisions. This ratio was 3.9 for individuals 50 years and older and 6.1 for those 65 years and older. In conclusion, there has been substantial and appropriate policy attention given to preventing pedestrian injuries from motor vehicles, but disproportionately little to pedestrian falls. However, the population burden of injurious pedestrian falls is significantly greater and justifies an increased focus on outdoor falls prevention, in addition to urban design, policy, and built environment interventions to reduce injurious falls on streets and sidewalks, than currently exists across the USA. [ABSTRACT FROM AUTHOR]

Effective collision risk reduction in autonomous vehicles relies on robust and straightforward pedestrian tracking. Challenges posed by occlusion and switching scenarios significantly impede the reliability of pedestrian tracking. In the current study, we strive to enhance the reliability and also the efficacy of pedestrian tracking in complex scenarios. Particularly, we introduce a new pedestrian tracking algorithm that leverages both the YOLOv8 (You Only Look Once) object detector technique and the StrongSORT algorithm, which is an advanced deep learning multi-object tracking (MOT) method. Our findings demonstrate that StrongSORT, an enhanced version of the DeepSORT MOT algorithm, substantially improves tracking accuracy through meticulous hyperparameter tuning. Overall, the experimental results reveal that the proposed algorithm is an effective and efficient method for pedestrian tracking, particularly in complex scenarios encountered in the MOT16 and MOT17 datasets. The combined use of Yolov8 and StrongSORT contributes to enhanced tracking results, emphasizing the synergistic relationship between detection and tracking modules. [ABSTRACT FROM AUTHOR]

Frequent occlusion of tracking targets leads to poor performance of tracking algorithms. A common practice in multi-target tracking algorithms is to re-identify the occluded tracking targets, which increases the number of identity switching occurrences. This paper focuses on online multi-object tracking and designs an anti-occlusion, robust association strategy, and feature extraction model. Specifically, the least squares algorithm and the Kalman filter are used to predict the trajectory of the tracking target, while the two-way self-attention mechanism is employed to extract the features of the tracking target, as well as positive and negative samples. After the tracking target is occluded, the association strategy is used to assign the identity information from before the occlusion. The experimental results demonstrate that the algorithm proposed in this paper has achieved excellent tracking performance on the MOT dataset. [ABSTRACT FROM AUTHOR]

Machine learning methods and agent-based models enable the optimization of the operation of high-capacity facilities. In this paper, we propose a method for automatically extracting and cleaning pedestrian traffic detector data for subsequent calibration of the ingress pedestrian model. The data was obtained from the waiting room traffic of a vaccination center. Walking speed distribution, the number of stops, the distribution of waiting times, and the locations of waiting points were extracted. Of the 9 machine learning algorithms, the random forest model achieved the highest accuracy in classifying valid data and noise. The proposed microscopic calibration allows for more accurate capacity assessment testing, procedural changes testing, and geometric modifications testing in parts of the facility adjacent to the calibrated parts. The results show that the proposed method achieves state-of-the-art performance on a violent-flows dataset. The proposed method has the potential to significantly improve the accuracy and efficiency of input model predictions and optimize the operation of high-capacity facilities. [ABSTRACT FROM AUTHOR]

Background: Multiple distraction indicators have been applied to measure street-crossing distraction but their validities in predicting pedestrian safety are poorly understood. Methods: Based on a video-based observational study, we compared the validity of four commonly used distraction indicators (total duration of distraction while crossing a street, proportion of distracted time over total street-crossing time, duration of the longest distraction time, and total number of distractions) in predicting three pedestrian safety outcomes (near-crash incidence, frequency of looking left and right, and speed crossing the street) across three types of distraction (mobile phone use, talking to other pedestrians, eating/drinking/smoking). Change in Harrell's C statistic was calculated to assess the validity of each distraction indicator based on multivariable regression models including only covariates and including both covariates and the distraction indicator. Results: Heterogeneous capacities in predicting the three safety outcomes across the four distraction indicators were observed: 1) duration of the longest distraction time was most predictive for the occurrence of near-crashes and looks left and right among pedestrians with all three types of distraction combined and talking with other pedestrians (Harrell's C statistic changes ranged from 0.0310 to 0.0335, P < 0.05), and for the occurrence of near-crashes for pedestrians involving mobile phone use (Harrell's C statistic change: 0.0053); 2) total duration of distraction was most predictive for speed crossing the street among pedestrians with the combination and each of the three types of distraction (Harrell's C statistic changes ranged from 0.0037 to 0.0111, P < 0.05), frequency of looking left and right among pedestrians distracted by mobile phone use (Harrell's C statistic change: 0.0115), and the occurrence of near-crash among pedestrians eating, drinking, or smoking (Harrell's C statistic change: 0.0119); and 3) the total number of distractions was the most predictive indicator of frequency of looking left and right among pedestrians eating, drinking, or smoking (Harrell's C statistic change: 0.0013). Sensitivity analyses showed the results were robust to change in grouping criteria of the four distraction indicators. Conclusions: Future research should consider the pedestrian safety outcomes and type of distractions to select the best distraction indicator. [ABSTRACT FROM AUTHOR]

Robust and accurate three-dimensional localization is essential for personal navigation, emergency rescue, and worker monitoring in indoor environments. For localization technology to be employed in various applications, it is necessary to reduce infrastructure dependence and limit the maximum error bound. This study aims to accurately estimate the location of various people using smartphones in a building with a cloud platform-based localization system. The proposed technology is modularized in a hierarchical structure to sequentially estimate the floor and location. This system comprises four localization modules: course level detection, fine level detection (FLD), fine location tracking (FLT), and level change detection (LCD). Each module operates organically according to the current user status. The position estimation range is defined as a total of three phases, and an appropriate location estimation module suitable for the corresponding phase operates to estimate the user's location gradually and precisely. When the user's floor is determined by an FLD, the two-dimensional position of the user is estimated by an FLT module that tracks the user's position by comparing the received signal strength indicator vector sequence and radio map. Also, LCD recognizes the user's floor change and converts the user's phase. To verify the proposed technology, various experiments were conducted in a six-story building, and an average accuracy of less than 2 m was obtained. [ABSTRACT FROM AUTHOR]

The popularity of new structural systems and prestressing technology has led to the widespread use of the large-space floor structures in large buildings such as high-speed rail terminals, conference centers, and sports stadiums. The reduction of nonessential load-bearing elements and the increase in span of the structure result in a reduction in the natural frequency and damping ratio of the floor structure, while the floor is a crowded area with disorderly flow between people, which may lead to human-induced vibration problems. In order to assess the dynamic performance of the large-span floor structure under crowd load, the random crowd-floor vertical interaction equation is derived, and the correctness of the equation is verified by comparing it with the test. For the stochastic nature of walking crowds, a formulation modeling method for random crowd is proposed, including pedestrian-dynamics parameters, formulation model, and response parameters. The model is characterized by considering inter- and intrasubject variability and reflects the vertical interaction between pedestrians and the floor system. According to the random crowd-floor dynamic equation, the variation of modal parameters and acceleration response of the floor during random crowd walking are also analyzed. The research in this paper will help in analyzing the comfort of large-span floor structures under pedestrian excitation and better meet the needs of the development of lightweight large-span structures. [ABSTRACT FROM AUTHOR]

Introduction: This study evaluated drivers' delays and yields when making right turns at the onset of a circular green traffic signal, in response to pedestrian behaviors, during daylight and at night with streetlamp lighting, in urban and suburban locations. Methods: Researchers conducted trials at three intersections, two typical urban and one typical suburban intersection. Driver movement data were collected (N = 302) in three conditions: no pedestrian present (control); a pedestrian displaying a cane; and a pedestrian flagging a cane. Driver delay was recorded in seconds from the onset of the green traffic signal to when the vehicle crossed the middle of the width of the pedestrian's crosswalk. Results: The primary result was that there are no statistical differences between day and night driver delay. Secondarily, statistically significant differences were found for urban versus suburban drivers when flagging the cane versus displaying the cane, with urban sites and flagging a cane garnering greater delays. Yield calculations, based on the time required for a pedestrian to nearly clear the first half of the crosswalk, confirmed the patterns of drivers' behaviors. Discussion: Authors anticipated negative effects on yielding in nighttime lower-light environments; however, there was similar driver delay and yields compared to daylight hours. Greater driver delay and yields, for flagging a cane and at urban settings, confirmed the results in previous studies. Implications for Practitioners: When assessing travel options, informed pedestrians and orientation and mobility specialists can consider the effectiveness of active cane techniques to increase the likelihood of yields from turning drivers, even at night where streetlamp lighting is the source of illumination. When instructing cane techniques for use at crosswalks, specialists can consider the options for mitigating risk when crossing typical streets in urban and suburban areas. [ABSTRACT FROM AUTHOR]

• The ADNMT uses two asymmetric networks to generate pseudo-labels for each other by clustering and iterative training. • Based on alternate training and mutual teaching to optimize the pseudo-labels. • The SCIC compensates the similarity distance of images captured by inter-camera. • The SSIC further suppresses the similarity distance by calculate the similarity distribution. Unsupervised person re-identification (Re-ID) has always been challenging in computer vision. It has received much attention from researchers because it does not require any labeled information and can be freely deployed to new scenarios. Most unsupervised person Re-ID research studies produce and optimize pseudo-labels by iterative clustering algorithms on a single network. However, these methods are easily affected by noisy labels and feature variations caused by camera shifts, which will limit the optimization of pseudo-labels. In this paper, we propose an Asymmetric Double Networks Mutual Teaching (ADNMT) architecture that uses two asymmetric networks to generate pseudo-labels for each other by clustering, and the pseudo-labels are updated and optimized by alternate training. Specifically, ADNMT contains two asymmetric networks. One network is a multiple granularity network, which extracts pedestrian features of multiple granularity that correspond to numerous classifiers, and the other network is a conventional backbone network, which extracts pedestrian features that correspond to a classifier. Furthermore, because the camera style changes seriously affect the generalization ability of the proposed model, this paper designs Similarity Compensation of Inter-Camera (SCIC) and Similarity Suppression of Intra-Camera (SSIC) according to the camera ID of the pedestrian images to optimize the similarity measure. Extensive experiments on multiple Re-ID benchmark datasets show that our proposed method achieves superior performance compared with the state‐of‐the‐art unsupervised person re-identification methods. [ABSTRACT FROM AUTHOR]

Statistical mechanics of walks defined on the spatial graphs of the city of Lubbock (10,421 nodes) and the Texas Tech University (TTU) campus pedestrian network (1466 nodes) are used for evaluating structural isolation and the integration of graph nodes, assessing their accessibility and navigability in the graph, and predicting possible graph structural modifications driving the campus evolution. We present the betweenness and closeness maps of the campus, the first passage times to the different campus areas by isotropic and anisotropic random walks, as well as the first passage times under the conditions of traffic noise. We further show the isolation and integration indices of all areas on the campus, as well as their navigability and strive scores, and energy and fugacity scores. The TTU university campus, a large pedestrian zone located close to the historical city center of Lubbock, mediates between the historical city going downhill and its runaway sprawling body. [ABSTRACT FROM AUTHOR]

Pedestrian detection has always been a difficult and hot spot in computer vision research. At the same time, pedestrian detection technology plays an important role in many applications, such as intelligent transportation and security monitoring. In complex scenes, pedestrian detection often faces some challenges, such as low detection accuracy and misdetection due to small target sizes and scale variations. To solve these problems, this paper proposes a pedestrian detection network PT-YOLO based on the YOLOv5. The pedestrian detection network PT-YOLO consists of the YOLOv5 network, the squeeze-and-excitation module (SE), the weighted bi-directional feature pyramid module (BiFPN), the coordinate convolution (coordconv) module and the wise intersection over union loss function (WIoU). The SE module in the backbone allows it to focus on the important features of pedestrians and improves accuracy. The weighted BiFPN module enhances the fusion of multi-scale pedestrian features and information transfer, which can improve fusion efficiency. The prediction head design uses the WIoU loss function to reduce the regression error. The coordconv module allows the network to better perceive the location information in the feature map. The experimental results show that the pedestrian detection network PT-YOLO is more accurate compared with other target detection methods in pedestrian detection and can effectively accomplish the task of pedestrian detection in complex scenes. [ABSTRACT FROM AUTHOR]

Analysis of pedestrians' head and lower limb injuries at the tissue level is lacking in studies of tram–pedestrian collisions. The purpose of this paper therefore to investigate the impact response process and severity of pedestrians' injuries in tram–pedestrian collisions, using the Total Human Model for Safety (THUMS) pedestrian human body model together with the tram FE model. Two full-scale tram–pedestrian dummy crash tests were performed to validate the FE model, and the total correlation and analysis (CORA) score of head acceleration yielded values of 0.840 and 0.734, confirming a strong agreement between the FE-simulated head responses and the experimental head kinematics. The effects of different tram speeds and impact angles on pedestrians' impact response injuries and the differences were further analyzed. The results indicate that direct impact of the lower limb with the tram's obstacle deflector leads to lower limb bone shaft fractures and knee tissue damage. Neck fling contributed to worsened head injury. Coup contusions were the predominant type of brain contusion, surpassing contrecoup contusions, while diffuse axonal injury was mainly concentrated in the collision-side region of the brain. Pedestrians' injuries are influenced by tram velocity and impact angle: higher tram velocities increase the risk of lower limb and head injuries. The risk of head injury for pedestrians is higher when the impact angle is negative, while lower limb injuries are more significant when the impact angle is 0°. This study provides practical guidance for enhancing tram safety and protecting pedestrians. [ABSTRACT FROM AUTHOR]

In Africa, 40% of traffic fatalities are pedestrians -- the highest proportion globally. Yet little is known about driver characteristics that are associated with unsafe driving in African countries. We aimed to explore associations between driving practices that endanger pedestrian safety and motorists' attitudes to the law (i.e. lawlessness and normlessness), controlling for sociodemographic and personality factors. We used the Response Time Method, based on Russell Fazio's attitude paradigm, to collect information about driver behaviours, attitudes, and personality traits among a sample of 440 motorists. Male gender was associated with unsafe driving, even when controlling for the effects of personality and attitudes to the law. Unsafe driving was also associated with four dimensions of motorists' personality, namely aggression, impulsivity, risk tolerance, and altruism, even when controlling for sociodemographic factors. Lawlessness (defined as a general disregard for the law) is also an important determinant of unsafe driving, even when controlling for sociodemographic and personality factors. These findings suggest that efforts to improve pedestrian safety in South Africa should focus on changing motorists' attitudes to the law. Significance: This study addresses pedestrian safety in the context of South Africa. The fact that 40% of traffic fatalities are pedestrians highlights the urgent need to understand the factors contributing to unsafe driving practices. The study delves into uncharted territory by examining driver characteristics associated with unsafe driving. Through exploring associations between driving practices that endanger pedestrian safety and motorists' attitudes to the law, this article provides valuable insights that can inform targeted interventions. [ABSTRACT FROM AUTHOR]

• This paper studies pedestrian behavior interacting with AVs on four-lane roads. • Roadway design and AV operations and signaling impact pedestrian behavior. • AV with yellow signal reduced pedestrian waiting time at curb and in the middle. • Interaction effects exist between roadway centerline features and AV operations. • Older individuals tend to wait longer before making crossing decisions. One of the main challenges autonomous vehicles (AVs) will face is interacting with pedestrians, especially at unmarked midblock locations where the right-of-way is unspecified. This study investigates pedestrians' crossing behavior given different roadway centerline features (i.e., undivided, two-way left-turn lane, and median) and various AV operational schemes portrayed to pedestrians through on-vehicle signals (i.e., no signal, yellow negotiating indication, and yellow/blue negotiating/no-yield indications). This study employs virtual reality (VR) to simulate an urban unmarked midblock environment where pedestrians interact with AVs as they cross a four-lane arterial roadway. Results demonstrate that both roadway centerline design features and AV operations and signaling significantly impact pedestrians' unmarked midblock crossing behavior, including the waiting time at the curb, waiting time in the middle of the road, and the total crossing time. But only the roadway centerline design features significantly impact the walking time. Participants in the undivided scene spent a longer time waiting at the curb and walking on the road than in the median and TWLTL scenes, but they spent a shorter time waiting in the middle of the road. Compared to the AV without a signal, the design of yellow signal significantly reduced pedestrian waiting time at the curb and in the middle. But yellow/blue signals significantly increased the pedestrian waiting time. Interaction effects between roadway centerline design features and AV operations and signaling are significant only for waiting time in the middle of the road. Other factors, such as demographics, past behaviors, and walking exposure of pedestrians, are also explored. Results indicate that older individuals tend to wait longer before making crossing decisions, and pedestrians' past crossing behaviors and past walking exposures do not significantly impact pedestrian walking behavior interacting with AV. [ABSTRACT FROM AUTHOR]

''Using eye-tracking technology, this study examined hazard detection at night. Using a 2 (younger versus older) x 2 (simulator versus on road) repeated-measures mixed design, 16 older adults and 17 younger adults drove their own vehicle and on a driving simulator under nighttime conditions wearing eye tracking technology. Both driving conditions had three roadway hazards of pedestrians looking at their cell phone while posed to cross the roadway. Pupil glances were recorded using outcome measures of total fixation duration, number of fixations, and time to first fixation. Results showed older adults detected hazards similarly to younger adults, especially during on-road performance. Night hazard detection was similar across driving conditions except for time to first fixation, which was faster on-road for both age groups. Results support potential use of driving simulators as a proxy for on-road with night driving needed for research and practice. [ABSTRACT FROM AUTHOR]

Objective of the Study: Present and investigate the impacts of an innovative strategy called superilla, implemented in Barcelona city (Spain). Methodology: The research is divided into two stages: i) Theoretical exploration of the superilla concept and its sustainable principles (pedestrian priority, reduction in vehicle traffic, urban mobility, creation of open spaces and green areas, urban furniture, reduction of air pollution, among others) and ii) Impacts analysis of the superilla implemented in Barcelona, through case studies: PobleNou and Sant Antoní. Originality/Relevance: Although the superilles strategy in Barcelona represents a milestone for sustainability, there is a scarcity of discussions on the subject in the Brazilian context. Main Results: The study outlined the implementation phases of the superilles and highlighted six key points of the intervention, such as: i) modal hierarchy; ii) regional issue, iii) coordination between projects; iv) popular participation, v) exportable model and vi) renaturalization of the city. Contributions: This research focuses on investigating the processes occurring in the urban intervention of superilles, making their understanding more in-depth and disseminating this process in Brazil, in order to contribute to the new urban strategies dissemination. [ABSTRACT FROM AUTHOR]

The pedestrian mode is gaining recognition as a basic building block in urban system design. Increasing attention is being given to developing vehicle-free zones to reduce urban pollution and return the inner city to its former role as a place for personal interaction. Attempts are being made to improve the walking experience, to make it more safe, convenient, and attractive especially in urban milieu to promote sustainable development. To provide the best pedestrian environment, it is necessary to understand pedestrian level of service. Conventional techniques of LOS determination consider the basic mobility characteristics of the pedestrian. The objective of this study is to develop a more accurate method of estimating level of service of pedestrians by developing composite indicators that integrate both qualitative and quantitative factors affecting both pedestrian and roadway conditions in the pedestrian movement. The relative importance of these multifaceted factors is assessed using Saaty scale and an Analytic Hierarchy Process is applied to determine the level of service. Field investigations include road network characterization and inventory studies, vehicular / pedestrian traffic characterization studies, pedestrian opinion surveys, expert opinion surveys. The approach has been applied to a major arterial road in Hyderabad city, Telangana, India in Mehdipatnam zone where pedestrian interactions are relatively high due to the commercial activity along the corridors. The stretches considered in the study exhibited Level of service "B" and "C" which implied necessary improvements to promote a good level of service. [ABSTRACT FROM AUTHOR]

Since pedestrians are impacted by solar radiation differently, urban designers must evaluate solar radiation exposure of pedestrian paths adopting an inclusive approach. This paper proposes a maximum threshold of direct solar radiation exposure for pedestrians based on activity, user profile and environmental conditions, defined as the difference between the energy consumption before feeling exhausted and the energy cost of walking. Two users of diverse walking abilities, a young adult and an elderly person with mobility impairment, were characterised by metabolic activity, walking speed and maximum energy capacity. Based on the theoretical framework, the energy budget of young adults to cope with thermal stress was set as three times higher than for the elderly. This framework was used to quantify the contribution of direct solar radiation to energy balance and then classify walkability during clear-sky summer hours; the term 'walkable' referred to environmental conditions allowing users to walk without feeling exhausted. The methodology was tested on an open area and an urban canyon in Milan; applicability by urban designers was key in developing a simplified way to evaluate shading needs. This approach could be applied to evaluate solar radiation exposure of pedestrian paths adopting diverse user experiences as an evaluation criterion. [ABSTRACT FROM AUTHOR]

The article presents selected results of research on improving pedestrian traffic safety. Based on annually-updated accident statistics made available by the police, as well as the new pedestrian traffic regulations in force, detailed work was undertaken to assess the level of visibility of pedestrians by drivers in pedestrian crossing areas. The research was carried out by analyzing several characteristic cases of pedestrian crossings occurring in Poland, in which there was only dedicated lighting for crossings, only street lighting, and a variant of coexistence of both of the above lighting solutions. Illuminance measurements were made in the horizontal and vertical planes of pedestrian crossings, and the results were confronted with the relevant guidelines. The next step involved a complementary measurement of the luminance distribution of the vertical plane containing the pedestrian and a portion of the sub- and super-horizontal background. Visibility pedestrians was considered in positive and negative contrast variants, and was then related to the obtained results of the illumination distribution. The analysis of the results of the study indicated the possibility of limited visibility of pedestrians at the crossings despite the satisfactory results obtained from measurements of the illuminance distribution within the crossings. [ABSTRACT FROM AUTHOR]

Across the world, economically marginalized people employ inventive strategies to encourage pedestrians to part with small sums of money by offering goods or services rather than directly requesting cash. People who use drugs wander urban streets selling magazines that some will buy but few will read. The homeless clean car windows that need no cleaning or provide token items instead of requesting donations because asking for money is contentious to the extent that selling stuff is not. Surveying existing ethnographic research, this article explores why. I analyse how informal income‐generating practices adapt to the cultural assumptions of majority populations, including the notion that adults should become valuable through gainful employment and the idea that the unreturned gift humiliates givers and receivers. [ABSTRACT FROM AUTHOR]

Lighting pedestrian crossings play an important role in ensuring the safety of pedestrians on the road, especially at night or in conditions of reduced visibility. Currently functioning normative and formal requirements, which are the basis for the design of lighting for pedestrian crossings, are described by criteria of lighting intensity. Each of the existing solutions operates in an urban environment, and at the stage of design and operation, the obtained values of the produced luminance contrast of the pedestrian silhouette and its background are not known. The main purpose of the article is to compare, based on luminance parameters, the three lighting solutions used at pedestrian crossings. This article presents and describes the method of determining the luminance contrast based on luminance measurements of the pedestrian silhouette and its background. Detailed results of measurements of luminance parameters at selected pedestrian crossings are presented. An analysis of the results was made with a breakdown of typical lighting solutions used in urban conditions. The differences between standard lighting were discussed, as well as supplementary and dedicated lighting. The obtained ranges of luminance contrast levels occurring in urban conditions for individual solutions were determined. The energy indicator for each solution was determined. [ABSTRACT FROM AUTHOR]

With daily traffic of 260 000 vehicles, the San Francisco–Oakland Bay Bridge is a major connection in the San Francisco Bay area (CA, USA). The new bridge is a designated lifeline structure (to remain open for emergency traffic after a major seismic event) with a design life of 150 years. The new bridge is 3.6 km long and consists of four distinct structures: a low-rise post-tensioned concrete box girder near the Oakland shore; a 2.4 km long segmental concrete box girder (skyway); a first-of-its-kind self-anchored suspension (SAS) bridge with a 385 m main span over the navigational channel; and a post-tensioned concrete box girder that connects to the east portal of the Yerba Buena Island tunnel. Opened in 2013, the signature span of the bridge is the SAS bridge with a length of 624 m and a total deck width of 79 m accommodating ten lanes of traffic in addition to a bike/pedestrian path. The US$6.4 billion megaproject was procured under multiple contracts and was delivered using the traditional design–bid–build method. This paper describes the key design innovations and construction methods which address the unique challenges on this project. [ABSTRACT FROM AUTHOR]

This paper presents an econometric modelling framework to unravel the communication and mobility issues of visually impaired pedestrians in the context of connected autonomous vehicles (CAVs). The research uses a dataset collected through a tailor-made stated-preference survey given to visually impaired pedestrians and provides evidence-based recommendations on communication techniques. The recommendations are based on the findings of a structural equation model (SEM) estimated using the survey data. The latent factors 'safety and security' and 'importance of hearing' are generated using a confirmatory factor analysis embedded in the SEM. The results from the model show that these two factors have negative influences on how much a visually impaired pedestrian trusts the use of CAVs. [ABSTRACT FROM AUTHOR]

Focusing on the crossing scenarios of pedestrians and autonomous vehicles (AVs) at unsignalized road sections, this study proposes an evolutionary game model considering the related benefits of AVs, pedestrians and traffic managers, and it explores the impact of policy intervention. The game analysis results indicate that strategies of the three agents are closely associated with pedestrians' trust in driverless technology, human-vehicle communication and government regulation. In cases where regulation exists, increasing the AV supervision threat would facilitate the equilibrium to evolve toward pedestrians crossing and AVs yielding. In addition, we unexpectedly find that the introduction of crash probability information cannot boost the equilibrium outcomes in the expected direction, which may be related to information overload. Finally, this study verifies the validity of the model through numerical simulations and provides managerial implications with respect to driverless technology promotion, communication device design and regulatory policy formulation. [ABSTRACT FROM AUTHOR]

This paper proposes a novel prediction model termed the social and spatial attentive generative adversarial network (SSA-GAN). The SSA-GAN framework utilizes a generative approach, where the generator employs social attention mechanisms to accurately model social interactions among pedestrians. Unlike previous methodologies, our model utilizes comprehensive motion features as query vectors, significantly enhancing predictive performance. Additionally, spatial attention is integrated to encapsulate the interactions between pedestrians and their spatial context through semantic spatial features. Moreover, we present a novel approach for generating simulated multi-trajectory datasets using the CARLA simulator. This method circumvents the limitations inherent in existing public datasets such as UCY and ETH, particularly when evaluating multi-trajectory metrics. Our experimental findings substantiate the efficacy of the proposed SSA-GAN model in capturing the nuances of pedestrian interactions and providing accurate multimodal trajectory predictions. [ABSTRACT FROM AUTHOR]

A simple approach to the Hamiltonian theory of radiation phenomena is proposed. It is shown that the so-called "Trautman-Bondi-mass", known to a rather narrow circle of specialists in general relativity, appears naturally in any special relativistic field theory. The structure of the "radiation data phase space" for the field and its isomorphism with the "Cauchy data phase space" are thoroughly analyzed. [ABSTRACT FROM AUTHOR]

Signal control, as an integral component of traffic management, plays a pivotal role in enhancing the efficiency of traffic and reducing environmental pollution. However, the majority of signal control research based on game theory primarily focuses on vehicular perspectives, often neglecting pedestrians, who are significant participants at intersections. This paper introduces a game theory-based signal control approach designed to minimize and equalize the queued vehicles and pedestrians across the different phases. The Nash bargaining solution is employed to determine the optimal green duration for each phase within a fixed cycle length. Several simulation tests were carried out by SUMO software to assess the effectiveness of this proposed approach. We select the actuated signal control approach as the baseline to demonstrate the superiority and stability of the proposed control strategy. The simulation results reveal that the proposed approach is able to reduce pedestrian and vehicle delay, vehicle queue length, fuel consumption, and CO2 emissions under different demand levels and demand patterns. Furthermore, the proposed approach consistently achieves more equalized queue length for each lane compared to the actuated control strategy, indicating a higher degree of fairness. [ABSTRACT FROM AUTHOR]

Pedestrian detection based on deep learning methods have reached great success in the past few years with several possible real-world applications including autonomous driving, robotic navigation, and video surveillance. In this work, a new neural network two-stage pedestrian detector with a new custom classification head, adding the triplet loss function to the standard bounding box regression and classification losses, is presented. This aims to improve the domain generalization capabilities of existing pedestrian detectors, by explicitly maximizing inter-class distance and minimizing intra-class distance. Triplet loss is applied to the features generated by the region proposal network, aimed at clustering together pedestrian samples in the features space. We used Faster R-CNN and Cascade R-CNN with the HRNet backbone pre-trained on ImageNet, changing the standard classification head for Faster R-CNN, and changing one of the three heads for Cascade R-CNN. The best results were obtained using a progressive training pipeline, starting from a dataset that is further away from the target domain, and progressively fine-tuning on datasets closer to the target domain. We obtained state-of-the-art results, M R − 2 of 9.9, 11.0, and 36.2 for the reasonable, small, and heavy subsets on the CityPersons benchmark with outstanding performance on the heavy subset, the most difficult one. [ABSTRACT FROM AUTHOR]

As micro-electro-mechanical systems (MEMS) technology continues its rapid ascent, a growing array of smart devices are integrating lightweight, compact, and cost-efficient magnetometers and inertial sensors, paving the way for advanced human motion analysis. However, sensors housed within smartphones frequently grapple with the detrimental effects of magnetic interference on heading estimation, resulting in diminished accuracy. To counteract this challenge, this study introduces a method that synergistically employs convolutional neural networks (CNNs) and support vector machines (SVMs) for adept interference detection. Utilizing a CNN, we automatically extract profound features from single-step pedestrian motion data that are then channeled into an SVM for interference detection. Based on these insights, we formulate heading estimation strategies aptly suited for scenarios both devoid of and subjected to magnetic interference. Empirical assessments underscore our method's prowess, boasting an impressive interference detection accuracy of 99.38%. In indoor environments influenced by such magnetic disturbances, evaluations conducted along square and equilateral triangle trajectories revealed single-step heading absolute error averages of 2.1891° and 1.5805°, with positioning errors averaging 0.7565 m and 0.3856 m, respectively. These results lucidly attest to the robustness of our proposed approach in enhancing indoor pedestrian positioning accuracy in the face of magnetic interferences. [ABSTRACT FROM AUTHOR]

Pedestrian re-identification aims to retrieve pedestrians across various cameras and scenes. However, the accuracy of re-identification is often affected by factors such as low-quality images of pedestrians and environmental conditions. Consequently, it is crucial for machine learning models to learn features from multiple dimensions. In response to these challenges, this paper proposes a Multi-Dimensional Attention Feature Fusion (MDAFF) method for pedestrian re-identification based on the NFormer approach. This method enables the model to learn and fuse pedestrian features from multiple dimensions, enriching the expressive power of the feature maps and improving the discrimination among pedestrians. By incorporating a PA module into the ResNeXt network for feature extraction, the model enhances its global perception and integrates pedestrian position information into the feature maps. This increases the model's sensitivity to pedestrian positions and reduces the impact of noise on re-identification accuracy. Furthermore, the method extracts channel and spatial correlations from the fused position feature maps and performs feature fusion, facilitating the fusion of multi-dimensional attention features. This alleviates the influence of varying scenarios and poses on re-identification, thereby enhancing the model's performance. Compared to the Res50+NFormer method, which directly models the relationships among different pedestrians after feature extraction, MDAFF integrates multi-dimensional features into the feature maps, improving the model's expressive power and capturing the relationships among different pedestrians more effectively. The proposed MDAFF method achieves a 1.3% increase in mAP and a 1.9% increase in Rank-1 on the Market1501 dataset, as well as a 1.7% increase in mAP and a 0.5% increase in Rank-1 on the DukeMTMC-reID dataset. Therefore, the MDAFF method effectively improves the accuracy of pedestrian re-identification. [ABSTRACT FROM AUTHOR]

This study develops a framework of pedestrian evacuation microsimulation modeling that considers pedestrians' social-physiological behavior in assessing an airport evacuation. The study implements social force model within a simulation platform enabling the articulation of stochastic pedestrian walking behavior realistically and reliably. It performs a sensitivity analysis of pedestrian behavior parameters to identify the candidate parameters required to capture pedestrian behavior under different levels of panic conditions. The study considers the case study of the Ottawa International Airport and tests and evaluates contrasting evacuation scenarios under low panic, medium panic, and high panic situations. Results indicate that under the low panic evacuation scenario, the pedestrians yield their movements with an increase in network bottleneck, potentially exhibit cooperative behavior, and control their speed with the rise of crowd density. On the contrary, individuals in high panic evacuation scenarios exhibit aggressive behavior indicated by their average speed, which is approximately 1.15 and 3.5 times the average compared with medium panic and low panic evacuation scenarios, respectively. Results suggest that it takes 5.38 min to evacuate 1300 passengers under high panic conditions compared with 9.75 min for a low panic evacuation scenario. However, in the case of a high panic evacuation scenario, the average speed keeps increasing even with the increase in crowd density. This framework can develop and evaluate strategies for safely evacuating the airport in the case of an emergency. [ABSTRACT FROM AUTHOR]

In the ostensibly unceasing prioritization of motorized infrastructure, walking has remained a ubiquitous mode of mobility for a large proportion of Nairobi's urban commuters. Planning for motorized mobility has historically been at a higher level of consideration although a much larger percentage of the population travels on foot. The conspicuous pedestrian has been and continues to be masked under the spotlight of the motor vehicle with a discernible outcome of spatial injustices. Using secondary data, historical literature and expert interviews, this paper examines how walking as a mode of mobility has developed over time and the challenges experienced by pedestrians in Nairobi. Linking to the notion of justice, the paper attempts to assess the association between walking and spatial justice using three dimensions—spatial, modal and individual dimensions—that are used as a framework to assess how injustices unfold and are experienced by Nairobi's pedestrians. The historical path dependency that has restricted and attempted to replace walkability by prioritizing motor vehicle use as well as the technical engineering design that lacks integration of social aspects of mobility has presented challenges in provision of safe non-motorized infrastructure in the contemporary urban travel in Nairobi, enduringly dismissing walking as a valid mode of mobility. Advancing spatial justice in Nairobi's urban mobility will require more than a technical process of extending the side of the road by a metre or two but rather deliberate effort in understanding the pedestrians' mobility needs that can best be understood by attuning to the everyday realities of travelling on foot. [ABSTRACT FROM AUTHOR]

Pedestrian trajectory prediction has many real-world applications, such as crowd video surveillance and self-driving cars. However, this is a challenging problem due to the complexity of modeling social interactions between road agents (pedestrians and vehicles). Previous studies that addressed trajectory prediction applied traditional deep learning approaches, including CNN and RNN. Meanwhile, the applications of graph neural networks have drawn increasing interest recently, and significant progress has been made in extracting features from complex and unstructured data. In this paper, we propose STM-GCN, a spatiotemporal multi-graph convolutional network for pedestrian trajectory prediction. Our approach consists of collecting information about the social interactions between pedestrians in a crowd that are position-based and velocity-based interactions integrated into a multi-graph. Then we apply Graph Neural Network learning on the obtained multi-graph for predictions. Through experiments on the ETH and UCY benchmarking datasets, our proposed method outperforms the state of the art by 6% average displacement error (ADE) and 10% final displacement error (FDE). Our implementation is available at https://github.com/YoussefTaki/STMGCN. [ABSTRACT FROM AUTHOR]