Pedestrian Network Research Articles

How do crosswalk delays affect pedestrian access in zoning areas? Walking access reduction by signalized crosswalks in Seoul, South Korea

Long waiting times at signalized crosswalks delay pedestrians and reduce the effective walksheds around urban facilities. In this study, we identify to what extent crosswalk delays at signalized crosswalks reduce pedestrians’ effective walking access and how the impact of crosswalk delays is differentiated by zoning types in Seoul, Korea. We demonstrate a pedestrian network analysis method considering crosswalk delay times, using a detailed pedestrian path geospatial layer and dataset of signal timings. Our results show significant reductions in walkshed size and served users. Employing multiple regression modeling, we identified that these reductions were not uniform and depended on zoning patterns in urban areas. In addition, our findings suggest that commercial zoning areas and recent residential developments, comprised of high-rise towers in mega-blocks, create greater delays to pedestrians and constrict effective walksheds. We discuss the need for further studies and the potential contribution of zoning system and its land use managements on improving walking conditions such as underground space developments and mixed residential density patterns.

Development of a GIS-based walking route planner with integrated comfort walkability parameters

Walking is a simple physical activity that is available to almost everyone and effective in reducing pollution, saving resources, and enhancing the general health of the population. Previous research has revealed the need to produce a WI that takes into account real-time comfort conditions. The objective of this study was to design a GIS-based algorithm that automated real time and customized WI computation and proposed a comfortable walking route using these inputs. Two comfort factors were considered; shaded areas and temperature. To this end, we devised a novel method for projecting shadows on the ground and a new way of creating a network of sidewalk routes. We designed an algorithm that successfully computed WI values which took the user’s comfort preferences into account. These values were then used to create the most comfortable route for the user to take along the sidewalks, in line with real-time weather conditions. Key findings unfold in two directions; firstly, GIS functions are enriched by the methods developed for creating a pedestrian network along sidewalks and road crossings, and for calculating the shadows cast by adjacent buildings. Secondly, the proposed algorithm constitutes a significant step forward in the development of a route planner that incorporates personalized walking comfort parameters based on real-time data. This in turn could potentially increase the frequency and duration of walks, thereby improving the general health of the population. Validation was based on an assessment of the accuracy of the model in calculating shadows and on the analysis of the routes calculated under different conditions.

Compression of Vehicle and Pedestrian Detection Network Based on YOLOv3 Model

Compression of Vehicle and Pedestrian Detection Network Based on YOLOv3 Model

The Pedestrian Network Concept: A Systematic Literature Review

The design of urban spaces that foster sustainable practices requires new analytical and structural approaches to spatial planning. An appropriate pedestrian network could significantly contribute to sustainable urban development goals, particularly by promoting sustainable mobility and pedestrian friendliness. With such goals, several attempts have been made to develop suitable models for pedestrian networks. However, something that is missing from the current literature is a framework that incorporates the main findings of the various studies as an integrated concise concept of the pedestrian network. To address this knowledge gap, this paper reviews studies on pedestrian networks and evaluates this concept based on the systematic 3W1H analysis method, which asks where, what, who, and how. In essence, the following questions are thus analyzed: Where is the pedestrian network located, What criteria play a role in the pedestrian network's performance, Who uses the pedestrian network, and How can the pedestrian network be analyzed? In this context, a systematic literature review is carried out by investigating studies conducted during the period 2001 to 2023 that appear in the Scopus database. The paper presents the results of the review of a selection of 67 papers dealing with pedestrian networks. Findings show that different models have been developed based on particular characteristics. Overall, researchers aimed to identify the most suitable network based on specific criteria for optimizing the walking experience in urban areas. By synthesizing the findings reported in these papers, this paper arguably contributes to a more comprehensive understanding of pedestrian networks, provides insights into the prioritization of design phases, facilitates the use of pedestrian network assessment models for future research, and creates a bigger picture for urban planners with a multidimensional view to a new sustainable urban structure.

Do footbridge and underpass improve pedestrian safety? A Hong Kong case study using three-dimensional digital map of pedestrian network

Hong Kong is a compact city with high activity and travel intensity. In the past decades, many footbridges and underpasses were installed to reduce the pedestrian-vehicle conflicts on urban roads. However, it is rare that the effects of configuration of pedestrian network on pedestrian crashes are investigated. In Hong Kong, many footbridges and underpasses are connected to major transport hubs and commercial building development and become parts of giant elevated and underground walkway systems. It is challenging to characterize such a complicated pedestrian network. In this study, a three-dimensional digital map is applied to estimate the connectivity and accessibility of pedestrian network, and measure the relationship between pedestrian network characteristics and pedestrian safety at the macroscopic level. Hence, the effects of footbridge and underpass on pedestrian safety are examined. For example, comprehensive built environment, pedestrian network, traffic, and crash data are aggregated to 379 grids (0.5 km × 0.5 km). Then, multivariate Poisson lognormal regression approach is applied to model fatal and severe injury (FSI) and slight injury pedestrian crashes, with which the effects of unobserved heterogeneity, spatial correlation, and correlation between crash counts are accounted. Results indicate that population density, traffic volume, walking trip, footpath density, node density, number of vertices per footpath segment, bus stop, metro exit, residential area, commercial area, and government and utility area are positively associated with pedestrian crashes. In contrast, average gradient, accessibility of footbridge, accessibility of underpass, and number of crossings per road segment are negatively associated with pedestrian crashes. In other word, pedestrian safety would be improved when footbridge and underpass are more accessible. Findings have implications for the design and planning of pedestrian network to promote walkability and improve pedestrian safety.

Global interest in walking accessibility: a scoping review

ABSTRACT We conduct a systematic scoping review of the academic literature concerning pedestrian accessibility. We distinguish “walking accessibility” from the broader topic of “walkability” by two criteria: papers must consider one or more destination type(s), and papers must address the issue of distance or impedance. After searching Web of Science, TRID, and Google Scholar databases and conducting screening, we identify 85 papers meeting these criteria. We organise the literature review according to the four components of accessibility identified by Geurs and van Wee [(2004). Accessibility evaluation of land-use and transport strategies: Review and research directions. Journal of Transport Geography, 12(2), 127–140]: (1) Land use; (2) transport; (3) temporal; and (4) individual and also add a section on the topic of impedance. Walking accessibility studies address a much greater range of land uses or destination types than is typically found for other modes. The transportation component is relatively undeveloped, as pedestrian infrastructure includes many influential elements not currently tracked in GIS systems. Few studies address the temporal component of walk accessibility, which varies according to climatic and nighttime conditions. Most papers do not account for the significant variation across individual capabilities and preferences regarding walking. We note that developing detailed pedestrian networks is a key first step, as most published analysis is conducted on roadway networks. A second major recommendation is to consider individual variations in walk accessibility across demographic classifications, accounting for varying levels of physical mobility.

Correlation-and-Correction Fusion Attention Network for Occluded Pedestrian Detection

As a significant issue in computer vision, pedestrian detection has achieved certain achievements with the support of deep learning. However, pedestrian detection in congested scenes still encounters the challenging problem of feature loss and obfuscation. To address the issue, we propose a pedestrian detection network based on a correlation-and-correction fusion attention mechanism. First, a multimask correction attention module is proposed, which generates visible part masks of pedestrians, enhancing the visible region’s features and correcting the false one. Besides, the module preserves the features of multiclass pedestrians by generating multiple masks. Then, we fuse a correlation channel attention module to enhance the correlation of various pedestrians’ body features. Next, we studied three fusion methods of correlation and correction attention mechanisms and found that the serial connection of “correlation first and correction behind” works best. Finally, we extend our method to multiclass pedestrian detection in congested scenes. Experimental results on the CityPersons, Caltech, and CrowdHuman datasets demonstrate the effectiveness of our method. On the CityPersons dataset where more than 70% of pedestrians are occluded, our method outperforms the baseline method by 1.12% on the heavy occlusion subset and surpasses many outstanding methods.

Measuring Individual Walking Accessibility based on Preference for Urban Walking Environment

Walking is the most universal mode of transport for human beings. As walking accessibility is significantly associated with social equity, many urban planners have focused on the improvement of walking accessibility by developing pedestrian environments. However, the pedestrians can have different levels of walking accessibility depending on individual characteristics. However, walking accessibility tends to be measured with distance or travel time, ignoring individual aspects. This paper proposes a method to measure individual walking accessibility based on the preference for the urban walking environment. For an individual-based measurement, we suggest a preference-based IWAI (Individual Walking Accessibility Index) which quantifies individual level of walking accessibility by summing environmental factors multiplied by preference-based weights. We measure the visible proportion of street greenness and sky openness using street view images and obtain the last two factors by integrating traffic data with pedestrian network. Finally, by comparing the street images of the shortest and the preference-based walking accessible path from IWAI, the result highlights that walking accessibility can be measured by individual preference for the environmental factors. This study provides a novel approach for measuring individual walking accessibility based on preference, which support implementing public polices or urban planning for improving walking accessibility.

MLFFCSP: a new anti-occlusion pedestrian detection network with multi-level feature fusion for small targets

MLFFCSP: a new anti-occlusion pedestrian detection network with multi-level feature fusion for small targets

Mapping the walk: A scalable computer vision approach for generating sidewalk network datasets from aerial imagery

While cities around the world are increasingly promoting streets and public spaces that prioritize pedestrians over vehicles, significant data gaps have made pedestrian mapping, analysis, and modeling challenging to carry out. Most cities, even in industrialized economies, still lack information about the location and connectivity of their sidewalks, making it difficult to implement research on pedestrian infrastructure and holding the technology industry back from developing accurate, location-based Apps for pedestrians, wheelchair users, street vendors, and other sidewalk users. To address this gap, we have designed and implemented an end-to-end open-source tool— Tile2Net —for extracting sidewalk, crosswalk, and footpath polygons from orthorectified aerial imagery using semantic segmentation. The segmentation model, trained on aerial imagery from Cambridge, MA, Washington DC, and New York City, offers the first open-source scene classification model for pedestrian infrastructure from sub-meter resolution aerial tiles, which can be used to generate planimetric sidewalk data in North American cities. Tile2Net also generates pedestrian networks from the resulting polygons, which can be used to prepare datasets for pedestrian routing applications. The work offers a low-cost and scalable data collection methodology for systematically generating sidewalk network datasets, where orthorectified aerial imagery is available, contributing to over-due efforts to equalize data opportunities for pedestrians, particularly in cities that lack the resources necessary to collect such data using more conventional methods.

A Pedestrian Detection Network Model Based on Improved YOLOv5

Advanced object detection methods always face high algorithmic complexity or low accuracy when used in pedestrian target detection for the autonomous driving system. This paper proposes a lightweight pedestrian detection approach called the YOLOv5s-G2 network to address these issues. We apply Ghost and GhostC3 modules in the YOLOv5s-G2 network to minimize computational cost during feature extraction while keeping the network's capability of extracting features intact. The YOLOv5s-G2 network improves feature extraction accuracy by incorporating the Global Attention Mechanism (GAM) module. This application can extract relevant information for pedestrian target identification tasks and suppress irrelevant information, improving the unidentified problem of occluded and small targets by replacing the GIoU loss function used in the bounding box regression with the α-CIoU loss function. The YOLOv5s-G2 network is evaluated on the WiderPerson dataset to ensure its efficacy. Our proposed YOLOv5s-G2 network offers a 1.0% increase in detection accuracy and a 13.2% decrease in Floating Point Operations (FLOPs) compared to the existing YOLOv5s network. As a result, the YOLOv5s-G2 network is preferable for pedestrian identification as it is both more lightweight and more accurate.

Bi-objective analytics of 3D visual-physical nature exposures in high-rise high-density cities for landscape and urban planning

Urban dwellers enjoy nature exposure in the neighborhood built environment through visual and physical ways, such as window views and outdoor activities. However, existing studies and analytics examine these pathways separately, leading to underinformed urban planning practices such as difficult prioritizing urban areas with both low-level nature exposures. The underinformation problem is particularly severe for high-rise, high-density cities that embrace high-level vertical diversity. This study aims to propose bi-objective analytics of 3D visual-physical nature exposures, for holistic – rather than separated – assessments. First, a floor-level Nature Exposure Index (NEI) is defined with visual and physical components. The visual component NEIv is assessed by window view imagery and deep transfer learning, while the physical component NEIp reflects the mean time from the floor to the nearest natural sites (e.g., nature parks and seaside) through the 3D pedestrian network. Then, bi-objective optimization-based analytics is designed for (i) identifying buildings and blocks with holistically low-level visual-physical nature exposures using NEI and (ii) examining probabilistic outputs and robustness of linear weighting schemes. A case study of 519 buildings showed that the NEI-enabled bi-objective analytics is automatic, effective, and inexpensive. Interviews with field experts confirmed that the analytics provides comprehensive evidence for a holistic identification of high-rise, high-density areas in need of nature exposure for landscape management and urban planning.

PERMEABILITY FACTORS OF OFFICE BLOCK PEDESTRIAN ACCESS AT TRANSIT ORIENTED DEVELOPMENT AREA - KARET BENDUNGAN HILIR

The development of public transportation in big cities in general, and Jakarta in particular, has set new trends in regional development by referring to the concept of Transit Oriented Development (TOD). Some superblock area development, bundled as a TOD area ignores several principles, especially pedestrian-friendly related values. Limited regulations and inconsistencies in understanding the application of TOD concepts and theories have resulted in the inappropriate application of TOD principles in transit-oriented areas. This study aims to find the implementation of connecting routes and the use of pedestrian paths with descriptive-qualitative methods by field observation. The research will be conducted in the Transit-Oriented development area Karet-Bendungan Hilir. The study shows that although the area being researched provides an ideal permeable block for pedestrian access, a problem with the management of the access prevents the pedestrian access to be fully functioning as a pedestrian networks system inside a Transit-Oriented Development area.

Investigating the application of deep learning to identify pedestrian collision-prone zones

The main objective of this study is to understand the factors that contribute to the frequency of both the total pedestrian-vehicle collisions and collisions that involve pedestrian violations and identify collision-prone areas. The two Full Bayes (FB) macro-level models were applied to historical collision records of the City of Hamilton to identify the collision-prone zones and the key factors that contribute to collision occurrence in TAZs. Finally, a self-organizing map (SOM) deep learning model was developed to identify collision-prone zones for the two collision classes. The results showed that the SOM model identified collision-prone zones with a high accuracy that exceeded the traditional Bayesian approach, based on the developed consistency test. As for the total collisions, the SOM model revealed that intersection density is the most important factor in distinguishing between collision-prone and non-collision-prone zones, followed by the pedestrian network directness and the proportion of residential land uses. As for the collisions that involved pedestrian violations, intersection density was also found to be the most important factor, followed by the density of bike-share stations and parking lots in a TAZ. The results of this study could aid planners in designing pedestrian-friendly networks and develop specific recommendations to enhance safety in unsafe zones.

The “15-minutes station”: a case study to evaluate the pedestrian accessibility of railway transport in Southern Italy

The recent sustainability challenges that our world is facing have raised, more than ever, the attention to the mobility of passengers and freight in the European and international agendas. The energy transition that has begun globally requires identifying and adopting safe, resilient, and increasingly sustainable mobility solutions. In this perspective, the modal split of passengers plays an essential role. One of the main encouraged policies is to promote an efficient mass rapid transit in urban and suburban areas. More in detail, when considering rail transport, it is necessary to analyze and evaluate the role of stations from at least two points of view: i) the ease of access to the station; ii) the opportunities that can be easily reached in its surroundings, following the concept of the “15-minutes” city. These two issues should be properly addressed to guarantee the role of railway stations as an access point to the transport system and an infrastructural element that can enhance a territory. Starting from these considerations, this research proposes a GIS-based methodology able to analyse railway stations from two points of view: i) walkability, considering the main functional characteristics of the transport network, and ii) impact on the territory, by identifying the services located in an area corresponding to “15-minutes” distances using active modes. For each railway station, the main activities in a 15-minute walking isochrone can be evaluated, both considering the walking distance on the pedestrian network and taking into account the current walkability of each link based on arc characteristics. This allows to study the accessibility of railway stations based on the current pedestrian network and the potential one with ideal characteristics. The method is applied to a case study located in Sicily (Italy), in the case of some urban stations. The final scope is to design a decision-support framework useful for railway station operators and local decision-makers to support strategic decisions regarding the railway system and the planning of appropriate pedestrian transport networks to increase railway station accessibility.

Spatiotemporal Attention-Based Pedestrian Trajectory Prediction Considering Traffic-Actor Interaction

Accurate pedestrian trajectory prediction is significant and challenging for traffic-actor protection and intelligent driving. However, most of the existing methods focus on pedestrian-pedestrian interaction ignoring the pedestrian-vehicle interaction and containing much ineffective interaction. To tackle the problem, this paper proposes a spatiotemporal LSTM attention network (STLAN) for pedestrian trajectory prediction based on a spatiotemporal dual attention module (STDAM). First, a pedestrian social force model is established to analyze the effective pedestrian-pedestrian interaction and pedestrian-vehicle interaction in mixed traffic scene. Second, a pedestrian trajectory prediction network architecture is constructed to integrate temporal and spatial information. By temporal attention STDAM captures underlying dependencies between time steps in individual pedestrian, assigning different weights to the input sequence of it. By spatial attention STDAM extracts latent interaction between traffic actors, assigning different weights to every traffic actor. In addition, random noise is introduced to simulate the uncertainty of pedestrian trajectory. Experimental results on a real-world traffic actors dataset show that STLAN outperforms several mainstream methods. Specifically, STLAN achieves 0.17/0.11 ADE/FDE and 3.79/0.38 AMD/AMV in the representative pedestrian-pedestrian interaction scene, with 0.15/0.17 ADE/FDE and 2.36/0.21 AMD/AMV in pedestrian-vehicle interaction scene. Furthermore, ablation experiments validate the effectiveness of STDAM in improving trajectory prediction accuracy.

신속통합기획을 적용한 공공보행통로 가이드라인 개발

The Public Pedestrian Routes, a representative space in the city, is a road centered on the daily lives of local residents, and not only provides living benefits but also cultural activities to form homogeneity of local communities, helping local residents enjoy walking, convenience of living, and revitalize local communities. Recognizing the importance of Public Pedestrian Routes, one of the alternatives to solving urban problems such as human alienation and disconnection, this study aims to present guidelines for Public Pedestrian Routes Wooshin villa maintenance in Gungdong, Seoul, applying a rapid integration plan. The plan will focus on the function of a circular pedestrian network so that not only the residents of Wooshin villa but also the residents of nearby areas can use the public space in the complex. To achieve the purpose, open parks and biotops in the complex are connected to Public Pedestrian Routes to form a green circulation system. The outer space that circulates through Biotop, open parks, and Public Pedestrian Routes is planned as a park for communication with local residents such as pedestrian-centered slit expansion, creation of three-dimensional spaces for various purposes, unit welfare facilities, and parking lots.

Connectivity analysis in pedestrian networks: A case study in Wuhan, China

Pedestrian network is an important part of urban transportation systems. The connectivity among networks substantially influences the residential access to surrounding amenity resources. Although studies to date have explored the evaluation and analysis of connectivity in pedestrian networks, the impact of mobility associated with network, e.g., public transport transfer, was not captured. In this study, we built a novel model that integrates network structural complexity and public transport transfer into connectivity index (CIi). The structural complexity of pedestrian networks as the first factor for the proposed model is calculated using the LNR (Link- Node Ratio) method. The degree of public transport transfer freedom as the second factor is quantified based on the number of public transport stations and their categories. We applied this model to the city of Wuhan, and the correlation between pedestrian network connectivity and urban amenities is further explored using Spearmans’ rank correlation coefficient. This study provides two findings: first, difference in connectivity was identified in some residential areas after considering public transport transfer freedom, compared with traditional LNR. Denser pedestrian network and transit stops were revealed in the locations with higher CIi, with differences between old and new districts. Second, newly proposed CIi improved the significance in the association between connectivity and urban amenities. In residential areas with advanced connectivity, urban planners are expected to expand urban amenities to improve the living environmental.

IPD-Net: Infrared Pedestrian Detection Network via Adaptive Feature Extraction and Coordinate Information Fusion.

Infrared pedestrian detection has important theoretical research value and a wide range of application scenarios. Because of its special imaging method, infrared images can be used for pedestrian detection at night and in severe weather conditions. However, the lack of pedestrian feature information in infrared images and the small scale of pedestrian objects makes it difficult for detection networks to extract feature information and accurately detect small-scale pedestrians. To address these issues, this paper proposes an infrared pedestrian detection network based on YOLOv5, named IPD-Net. Firstly, an adaptive feature extraction module (AFEM) is designed in the backbone network section, in which a residual structure with stepwise selective kernel was included to enable the model to better extract feature information under different sizes of the receptive field. Secondly, a coordinate attention feature pyramid network (CA-FPN) is designed to enhance the deep feature map with location information through the coordinate attention module, so that the network gains better capability of object localization. Finally, shallow information is introduced into the feature fusion network to improve the detection accuracy of weak and small objects. Experimental results on the large infrared image dataset ZUT show that the mean Average Precision (mAP50) of our model is improved by 3.6% compared to that of YOLOv5s. In addition, IPD-Net shows various degrees of accuracy improvement compared to other excellent methods.

Research on Pedestrian Detection Algorithm Based on MobileNet-YoLo.

To address the problem that large pedestrian detection networks cannot be directly applied to small device scenarios due to the heavyweight and slow detection speed, this paper proposes a pedestrian detection and recognition model MobileNet-YoLo based on the YoLov4-tiny target detection framework. To address the problem of low accuracy of YoLov4-tiny, MobileNetv3 is used to optimize its backbone feature extraction network, and the MFF model is proposed to fuse the output of the first two layers to solve the information loss problem, and the attention mechanism CBAM is introduced after strengthening the feature extraction network to further improve the detection efficiency; then the 3 × 3 convolution is replaced by the depth separable convolution, which greatly reduces the number of parameters and thus improves the detection rate, then propose Ordinary data augmentation to efficiently augment the dataset and dynamically adjust the target detection anchor frame using the k-means++ clustering algorithm. Finally, the model weights trained by the VOC2007 + 2012 dataset were applied to the pedestrian dataset for retraining by the transfer learning method, which effectively solved the problem of scarce samples and greatly shortened the training time. The experimental results on the VOC2007 + 2012 dataset show that the average means accuracy of the MobileNet-YoLo model compared to YoLov4-tiny, MobileNet-YoLov4, MobileNet-YoLov3, and YoLov5s by 5.00%, 1.30%, 3.23%, and 0.74%, respectively and have reached the level to realize the landed application.