Pedestrian Counting Research Articles

The Impact of PM10 Levels on Pedestrian Volume: Findings from Streets in Seoul, South Korea.

Although many studies have revealed that both air quality and walking activity are dominant contributors to public health, little is known about the relationship between them. Moreover, previous studies on this subject have given little consideration to the day-to-day atmospheric conditions and floating populations of surrounding areas even though most pedestrian count surveys are not conducted on a single day. Against this backdrop, using the 2015 Pedestrian Volume Survey data and quasi-real-time weather, air quality, and transit ridership data in Seoul, this study investigates the relationship between particulate matter (PM)10 and pedestrian street volumes empirically. The regression results suggest that PM10 concentration determines people’s intention to walk and affects the volume of street-level pedestrians. The three regression models, which adopted different spatial aggregation units of air quality, demonstrated that PM10 elasticity of pedestrian volume is the largest in the borough-level (the smallest spatial unit of air quality alert) model. This means that people react to the most accurate information they can access, implying that air quality information should be provided in smaller spatial units for public health. Thus, strengthening air quality warning standards of PM is an effective measure for enhancing public health.

A novel framework for automated monitoring and analysis of high density pedestrian flow

Pedestrian traffic is an important subject of surveillance to ensure public safety and traffic management, which may benefit from intelligent and continuous analysis of pedestrian videos. State-of-the-art methods for intelligent pedestrian surveillance have a number of limitations in automating and deriving useful information of high-density pedestrian traffic (HDPT) using closed circuit television (CCTV) images. This work introduces an automatic and improved HDPT surveillance system by integrating and optimizing multiple computational steps to predict pedestrian distribution from input video frames. A fast and efficient particle image velocimetry (PIV) technique is proposed to yield pedestrian velocities. A machine learning regressor model, boosted Ferns, is used to improve pedestrian count and density estimation: an essential metric for HDPT analysis. A camera perspective model is proposed to improve the speed and position estimates of HDPT by projecting 2D image pixels to 3D world-coordinate dat. All these functional improvements in HDPT velocity and displacement estimations are used as inputs to a sophisticated pedestrian flow evolution model, PEDFLOW to predict HDPT distribution at a future time point, which is a crucial information for pedestrian traffic management. The predicted and simulated HDPT properties (density, velocity) obtained using the proposed framework show low errors when compared to the ground truth data. The proposed framework is computationally efficient, suitable for multiple camera feeds with HDPT videos, and capable of rapidly analyzing and predicting flows of thousands of pedestrians. The paper shows one of the first steps towards fully integrated CCTV-based automated HDPT management system.

MEASURING ATTRIBUTES AND BEHAVIOURAL PATTERNS OF STREET LIVELINESS IN URBAN AREA

Streets are essential elements of the form and function of the urban environment. A growing body of research has pointed out that the more human activities in the street, the more interesting the city will be. Streets showcase the community and connect people. They are the most comfortable social environments that provide aesthetical and interactional pleasure for everyone. This paper presents the outcomes of a study which measured street liveliness based on the attributes and behaviours of the pedestrians. The aim of this study was to examine how people experience the liveliness at various times for various activities on the street. To achieve this, the study applied a case study method by looking at one of the streets that showed a diversity of activities with a unique shopping experience and liveliness in Kuala Lumpur called Jalan Masjid India. The method involved two types of data collection techniques, which were site observations, and pedestrian counts. Results which consisted of behavioural mapping and activity analysis revealed that Jalan Masjid India has its own merits in terms of street liveliness, function, and urban activities.

MEASURING ATTRIBUTES AND BEHAVIOURAL PATTERNS OF STREET LIVELINESS IN URBAN AREA

Streets are essential elements of the form and function of the urban environment. A growing body of research has pointed out that the more human activities in the street, the more interesting the city will be. Streets showcase the community and connect people. They are the most comfortable social environments that provide aesthetical and interactional pleasure for everyone. This paper presents the outcomes of a study which measured street liveliness based on the attributes and behaviours of the pedestrians. The aim of this study was to examine how people experience the liveliness at various times for various activities on the street. To achieve this, the study applied a case study method by looking at one of the streets that showed a diversity of activities with a unique shopping experience and liveliness in Kuala Lumpur called Jalan Masjid India. The method involved two types of data collection techniques, which were site observations, and pedestrian counts. Results which consisted of behavioural mapping and activity analysis revealed that Jalan Masjid India has its own merits in terms of street liveliness, function, and urban activities.

Understanding volume and correlations of automated walk count: Predictors for necessary, optional, and social activities in Dilworth Park

In this paper, we explore the potential use of automated pedestrian walk count data in urban design research. The Center City District (CCD) research group used computer vision to collect automated pedestrian walk data from Dilworth Park, Philadelphia. By comparing the count data and participant observations of social activities in the park, we found that the frequencies of social activities in the park could be predicted by the pedestrian count when considering the outdoor thermal comfort index and the types of events taking place in Dilworth Park. By examining correlations among multiple sensors, we found that the entry–exit correlation is a useful indicator to assess how people use public space by estimating the ratio of necessary-to-optional activities.

IoT Enabled Intelligent Sensor Node for Smart City: Pedestrian Counting and Ambient Monitoring

An Internet of Things (IoT) enabled intelligent sensor node has been designed and developed for smart city applications. The fabricated sensor nodes count the number of pedestrians, their direction of travel along with some ambient parameters. The Field of View (FoV) of Fresnel lens of commercially available passive infrared (PIR) sensors has been specially tuned to monitor the movements of only humans and no other domestic animals such as dogs, cats etc. The ambient parameters include temperature, humidity, pressure, Carbon di Oxide (CO2) and total volatile organic component (TVOC). The monitored data are uploaded to the Internet server through the Long Range Wide Area Network (LoRaWAN) communication system. An intelligent algorithm has been developed to achieve an accuracy of 95% for the pedestrian count. There are a total of 74 sensor nodes that have been installed around Macquarie University and continued working for the last six months.

Daily Bicycle and Pedestrian Activity as an Indicator of Disaster Recovery: A Hurricane Harvey Case Study.

Changes in levels and patterns of physical activity might be a mechanism to assess and inform disaster recovery through the lens of wellbeing. However, few studies have examined disaster impacts on physical activity or the potential for physical activity to serve as an indicator of disaster recovery. In this exploratory study, we examined daily bicycle and pedestrian counts from four public bicycle/pedestrian trails in Houston, before and after Hurricane Harvey landfall, to assess if physical activity returned to pre-Harvey levels. An interrupted time series analysis was conducted to examine the immediate impact of Harvey landfall on physical activity; t-tests were performed to assess if trail usage returned to pre-Harvey levels. Hurricane Harvey was found to have a significant negative impact on daily pedestrian and bicycle counts for three of the four trails. Daily pedestrian and bicycle counts were found to return to pre-Harvey or higher levels at 6 weeks post-landfall at all locations studied. We discuss the potential for further research to examine the trends, feasibility, validity, and limitations of using bicycle and pedestrian use levels as a proxy for disaster recovery and wellbeing among affected populations.

A convolutional neural‐network‐based pedestrian counting model for various crowded scenes

AbstractPedestrian counting from unconstrained images is an important task in various applications such as resource management, transportation engineering, urban design, and advertising, but it is greatly challenged by some factors such as interocclusion, cross‐scene, scale, and scene perspective distortion. Traditional image‐based methods suffer from them, and the performance of conventional sensor‐based methods such as Kinect and LASER degrades gradually with the increase in pedestrian count and distance from the device to pedestrians. Based on these challenges, this paper proposes a new network model making use of stacked multicolumn convolutional neural networks (CNNs) for pedestrian counting. The human's head features are used to replace the whole body for solving the problem of serious occlusion and choose multicolumn CNNs for dealing with scale and scene perspective distortion. Also, pretrained VGG‐16 is used to generate deeper detailed features and expand the receptive field of the model. Extensive analysis and experiments on current major pedestrian counting datasets show that the proposed network model has considerable advantages in pedestrian counting tasks compared to other state‐of‐the‐art models, and the proposed network model has an improvement effect for the training process. Moreover, the visual differences between the generated density map and ground‐truth density map are visualized and analyzed quantitatively to demonstrate the feasibility of the model.

The Impact of Twenty Four-Hour Public Transport in Melbourne, Australia: An Evaluation of Alcohol-Related Harms

Transporting people out of nightlife districts is often cited as a major issue associated with alcohol-related harm. The Victorian Government introduced 24-hour public transport (24hr PT) in Melbourne, Australia, on Friday and Saturday nights in January 2016. After the 1-year trial period, funding was extended for a further 4 years, at a cost of more than AU$300 million to date. The current study aimed to determine whether 24hr PT reduced harms associated with the nightlife of Melbourne and whether there has been an increased number of people using the transport and visiting Melbourne city on Friday and Saturday nights. Police assault data, ambulance attendance data, crash data, public transport use data, and pedestrian counting data were analyzed to determine the impact of 24hr PT on harms in the nightlife of Melbourne, as well as changes in the number of people using public transport and attending the city. There was no change from 2015 to 2016 in the number of police-recorded assaults, ambulance attendances, or crashes for the entirety of the night. There were significantly more people out in the city later in the evening, and more people using trains and trams during the 24hr PT time (i.e., 1 A.M.-5 A.M.). The initiative did not decrease harm in Melbourne nightlife, in contrast to industry, government, and expert predictions. With expenditures of more than AU$300 million, the costs and benefits of this initiative require further consideration and research, especially when it is proposed in opposition to evidence-based solutions, such as closing venues earlier.

Pedestrian Flow Tracking and Statistics of Monocular Camera Based on Convolutional Neural Network and Kalman Filter

Pedestrian flow statistics and analysis in public places is an important means to ensure urban safety. However, in recent years, a video-based pedestrian flow statistics algorithm mainly relies on binocular vision or a vertical downward camera, which has serious limitations on the application scene and counting area, and cannot make use of the large number of monocular cameras in the city. To solve this problem, we propose a pedestrian flow statistics algorithm based on monocular camera. Firstly, a convolution neural network is used to detect the pedestrian targets. Then, with a Kalman filter, the motion models for the targets are established. Based on these motion models, data association algorithm completes target tracking. Finally, the pedestrian flow is counted by the pedestrian counting method based on virtual blocks. The algorithm is tested on real scenes and public data sets. The experimental results show that the algorithm has high accuracy and strong real-time performance, which verifies the reliability of the algorithm.

Pedestrian Count Expansion Methods: Bridging the Gap between Land Use Groups and Empirical Clusters

Count expansion methods are a useful tool for creating long-term pedestrian or cyclist volume estimates from short-term counts for safety analysis or planning purposes. Expansion factors can be developed based on the trends from automated counters set up for long periods of time. Evidence has shown that the activity patterns can vary between sites so that there is potential to create more accurate estimates by grouping similar long-term count trends into factor groups. There are two common approaches to developing factor groups in pedestrian and cyclist count expansion studies. The land use classification approach has the advantage of being simple to apply to short-term count locations based on attributes of the surrounding area, but it requires assumptions by the researchers about which characteristics correlate with different activity patterns. Empirical clustering approaches can potentially create more distinct clusters by effectively matching locations with similar patterns, but they do not present an easy way to apply the resulting factor groups to appropriate short-term count sites. This study connects the two approaches and takes advantage of the benefits of both by using objective measures of the surrounding land use to model membership in the empirical cluster groups.

Pedestrian counting estimation based on fractal dimension

Counting the number of pedestrians in urban environments has become an area of interest over the past few years. Its applications include studies to control vehicular traffic lights, urban planning, market studies, and detection of abnormal behaviors. However, these tasks require the use of intelligent algorithms of high computational demand that need to be trained in the environment being studied. This article presents a novel method to estimate pedestrian flow in uncontrolled environments by using the fractal dimension measured through the box-counting algorithm, which does not require the use of image pre-processing and intelligent algorithms. Four scenarios were used to validate the method presented in this article, of which the last scene was a low-light surveillance video, showing experimental results with a mean relative error of 4.92% when counting pedestrians. After comparing the results with other techniques that depend on intelligent algorithms, we can confirm that this method achieves improved performance in the estimation of pedestrian traffic.

Have walking and bicycling increased in the US? A 13-year longitudinal analysis of traffic counts from 13 metropolitan areas

Local, state, and federal governments promote walking and bicycling in order to reduce emissions and improve public health. Tracking rates of bicycling and walking over time is important for assessing progress towards this goal. In the United States, most data are limited to cross-sectional self-report surveys (e.g., National Household Travel Surveys [NHTS]) or capture only the main commute mode (e.g., American Community Survey [ACS]). This study examines temporal trends (while controlling for spatial factors) of active travel in 13 US metropolitan areas between 2004 and 2016 (with 78% of counts occurring between 2010 and 2016) using repeated counts of bicycle and pedestrian traffic (n = 1319 count locations; 5554 bicycle and 5166 pedestrian counts). We used multilevel mixed-effects models to examine the multi-year trend in bicycling and walking during morning and afternoon peak periods. In the 5 out of 8 models where the temporal trend was statistically significant, we found that, on average, traffic volumes increased at a rate of 2–6% (bicycle) and 2–3% (pedestrian) per year among count locations, holding other variables constant. Presence of bicycle facilities (e.g., bicycle lanes, off-street trails) was positively associated with higher levels of bicycle traffic. Our results based on observed traffic patterns suggest larger increases in bicycling and walking as compared to the trends reported from the NHTS and ACS data. Quantifying the temporal trend from observed counts of traffic may aid policy makers and urban planners in assessing progress towards the goal of increasing bicycling and walking to reduce emissions and increase physical activity.

Cross-Line Pedestrian Counting Based on Spatially-Consistent Two-Stage Local Crowd Density Estimation and Accumulation

This paper proposes a scalable approach for counting pedestrians crossing a virtual line when the crowd is highly dynamic and possibly extremely dense. The approach mainly consists of two parts: local crowd density estimation and pedestrian counting based on accumulating local densities across the line. To obtain a fine estimation of local crowd densities, we divide the neighborhood at the line into a number of blocks. We enforce spatial consistency between local counts in the blocks and those in the enclosing regions to guarantee consistent estimation of local crowd densities. For scalability to various density levels in crowd density estimation, we propose a two-stage strategy: pre-classification of density levels and subsequent regression with overlapped operational ranges. To count pedestrians crossing the virtual line, we accumulate the crowd densities across the line according to the locally estimated velocities. Extensive experimental results demonstrate the effectiveness of the proposed approach and its scalability to crowdedness.

Bicycle and Pedestrian Count Programs: Scan of Current U.S. Practice

As bicycling and walking have become more integrated into transportation agencies’ processes of planning, design, and operations, some state, regional, and local agencies have established nonmotorized data collection programs of varying scopes and with varying methods. The purpose of this study was to identify ways to plan and implement a nonmotorized count program in Virginia, and the scope included reviewing existing U.S. national-level guidance and examples from state departments of transportation (DOTs) other than Virginia’s to determine the most effective ways of implementing such a program. Study tasks included synthesizing the literature to obtain relevant information with regard to nonmotorized travel monitoring programs, practices, and technologies, as well as obtaining information from representatives of three states through interviews of public agency staff and researchers involved in each state’s program. The study found a large volume of recent research on the topic of nonmotorized travel monitoring. The study concluded that the practice of nonmotorized travel monitoring has evolved and expanded in recent years; that many commercially available counting technologies exist and have been evaluated; that the practice of nonmotorized travel monitoring, as with motorized travel monitoring, has several aspects beyond purchase and installation of automatic count equipment; and that several states are developing nonmotorized count programs and have begun putting their data to use. The findings provide a foundational resource for state DOTs that are considering developing state-level counting programs.

An efficient algorithm for tracking and counting pedestrians based on feature points in video surveillance applications

An efficient algorithm for tracking and counting pedestrians based on feature points in video surveillance applications

Analysis of the influence of urban built environment on pedestrian flow in an intermediate-sized city in the Andes of Ecuador

This study explores the potential influence of the urban built environment on pedestrian flow in an intermediate Latin American city, Cuenca-Ecuador. Data from samples of 48 street segments were analyzed to explore how physical and spatial features of streets and sidewalks, as well as land use and occupation conditions influence the number of pedestrians. A quantitative approach was used to model the individual and combined influences of several variables on pedestrian counts using multiple regression models. Results from statistical modeling indicate that variables related to the physical features of the streets, such as sidewalk width, are positively correlated to pedestrian counts, whereas parking space and front setback are negatively correlated. We conclude that streets in our study area are largely inadequate for pedestrians due to their poor design, scarce walking infrastructure, and the prevalence of all sorts of obstacles on the sidewalks. Pedestrians, however, seem to cope with this adverse environment by adopting different strategies. We expect that this study will help city planners design better environments for non-motorized travel in intermediate cities.

Land-Use, Street Configuration and Pedestrian Volume: The Case of a Historic Town, Mymensingh, Bangladesh

This study explores the relative connections among pedestrian movement patterns, land use and street configuration by analyzing the pedestrian volume, existing land use pattern as well as the street configuration in different streets of Mymensingh. Mymensingh is a historic town of Bangladesh which was established by the British Colonists in more than 200 years ago along the river Brahmaputra. The street patterns of Mymensingh is very unique as it was developed by the fusion of the streets made by British Colonists and the narrow streets made by the local inhabitants. By using the method of Space Syntax the configurational values of street segments are compared with the corresponding pedestrian counts. The study result shows that the correlation is very poor within this two factors as commercial land uses affects the pedestrians more.

Predicting Pedestrian Counts for Crossing Scenario Based on Fused Infrared-Visual Videos

Estimating the number of pedestrians based upon surveillance videos and images has been a critical task while implementing intelligent signal controls at intersections. However, this has been a difficult task considering the pedestrian waiting area is an outdoor scenario with complex and time-varying surrounding environment. In this study, a method for estimating pedestrian counts based on multisource video data has been proposed. First, the partial least squares regression (PLSR) model is developed to estimate the number of pedestrians from single-source video (either visible light video or infrared video). Meanwhile, the temporal feature of the scenario (daytime or nighttime) is identified based on visible light video. According to the recognized time periods, pedestrian count detection results from the visible light and infrared video data can be obtained with preset corresponding confidence levels. The empirical experiments showed that this fusion method based on environment perception holds the benefits of 24-hour monitoring for outdoor scenarios at the pedestrian waiting area and substantially improved accuracy of pedestrian counting.

Street life and the built environment in an auto-oriented US region

Urban planners and designers believe that the built environment at various geographic scales affects pedestrian activity, but have limited empirical evidence at the street scale, to support their claims. We are just beginning to identify and measure the qualities that generate active street life, and this paper builds on the first few studies to do so. This study measures street design qualities and surrounding urban form variables for 881 block faces in Salt Lake County, Utah, and relates them to pedestrian counts. This is the largest such study to date and includes suburbs as well as cities. At the neighborhood scale, we find that D variables – development density, accessibility to destinations, and distance to transit – are significantly associated with the pedestrian activity. At the street scale, we find significant positive relationships between three urban design qualities – imageability, human scale, and complexity – and pedestrian counts, after controlling for neighborhood-scale variables. Finally, we find that pedestrian counts are positively associated with seven of twenty streetscape features – historic buildings, outdoor dining, buildings with identifiers, less sky view, street furniture, active uses, and accent building colors. This study provides implications for streetscape projects that aim to create walkable places in typical auto-oriented, medium-sized cities.