Despite efforts to identify the impact of the street environment on the movements and behaviors of children, research has focused on only a few streets because the methods employed were time-consuming and labor-intensive, which limits extending the analysis to multiple streets over a long period. With the development of vision-based human behavior analysis, researchers can record the overall movements of children automatically and obtain a large amount of movement behavior information. This study aims to analyze the physical environments of streets by monitoring the movement behaviors of children on streets through computer vision, using CCTV video footage collected from Osan, South Korea. The proposed methods identify child movement behaviors through their trajectories from video footage, identify child-friendly street indices based on child movement behaviors, and classify street environments. By performing a statistical analysis, this study analyzes the relationships between the movement behaviors of children and the physical environments of the streets. Potential interventions to encourage diverse and independent child movement behaviors through the improvement of school zone streets, visual exposure to varying environments, and safe zones at the entrances adjacent to child-related facilities are suggested. This study will help to improve the design of street environments to allow children to move and behave freely and independently.

With steep increases in the number of tourists, the term 'overtourism' has emerged to describe the direct negative impact of tourism on the lives of residents in popular destinations. To address these challenges, it is necessary to undertake precise spatial assessments to formulate effective management strategies. Accordingly, this study presents a novel spatial analysis-based research framework for overtourism management, focusing on Bukchon, Seoul, South Korea. This framework investigates spatial conflicts to identify urban areas in need of overtourism management strategies. By utilizing various open-source datasets with fine-grained resolution, we apply a stepwise spatial analysis to delineate tourism-residence conflict zones. Through contextual exploration, we identify common attributes among the five detected potential conflict zones, shedding light on the following two predominant street types: vibrant commercial districts and quaint alleys emblematic of the tourist destination. These findings provide a nuanced understanding of the street environments contributing to overtourism, thus offering an objective assessment of how overtourism spatially unfolds in urban settings.

This study investigates the effectiveness of strengthened penalty policies in South Korean school zones by analyzing the changes in road users' behaviors, focusing on pedestrian-vehicle interactions. This study employed three surrogate safety measurements: vehicle speed, Pedestrian Safety Margins (PSM), and Predicted Collision Risk (PCR) level. The comprehensive analysis covers a spectrum of behaviors, from simple to complex, assessing the policy's impact in urban environments. The findings reveal several important insights. First, the policy enforcement resulted in a positive impact on vehicle speeds, with average speeds aligning with posted speed limits. Second, an increase was observed in yielding behaviors, particularly in school zones. However, much of this behavior appeared to be cosmetic, emphasizing the need for more safety-oriented yielding practices. Finally, the policy enforcement had a mixed impact on PCR levels, with a reduction in danger levels in school zones, yet an unexpected increase in danger levels in non-school zones. This study provides valuable insights into the effectiveness of strengthened penalty policies in school zones, particularly the development of a safe and sustainable urban environment.

As analyzing and controlling human movements is a key task in urban planning and management process, decision-makers have developed the computer-aided methodologies for pedestrian volume estimation. However, the existing pedestrian volume estimation studies have faced several constraints, such as the high cost of manual counting. This study aims to develop a street-level pedestrian volume estimation model by utilizing a computeraided approach to evaluate data acquired through closed-circuit television (CCTV). We employ spatial characteristics and pedestrian volume data derived from sensed points to estimate pedestrian volumes on unsensed streets. The Bayesian Neural Network (BNN)-based model exhibited the best performance for the pedestrian volume estimation, with a mean hourly error of 21.305 individuals for 10-fold cross-validation and total 77.394 individuals for test estimation on four streets representing diverse pedestrian volume, surpassing the performance of regression and deep neural network (DNN)-based models. The BNN-based model also exhibits considerable performance in terms of estimating the hourly pedestrian volumes of entire streets within the study site (i.e., Bukchon Hanok Village in Seoul), and we examine the potential of the stochastic model to assess the uncertainty of a probabilistic artificial intelligence (AI) model. Additionally, a scenario-based test that assumes two specific road control situations presents a numeric and visual aid for predicting the aftermath of an urban management policy. The methodology proposed in this study can contribute to the exploitation of computeraided techniques and trustworthy AI for supporting decision-making processes, thereby providing urban planners with more quantitative evidence.

Understanding how children use streets will allow urban planners to create streets that are livelier and more child-friendly. However, it is still challenging to investigate detailed information on children's movement behaviors on individual streets due to labor-intensive observations, the difficulty of integrating qualitative data, and privacy concerns. With the wide application of vision-based technologies, Researchers can analyze the movements and behaviors of large groups of children on streets automatically and continuously. This study aims to analyze children's movement behaviors on streets utilizing computer vision techniques in urban surveillance systems. The proposed methods automatically extract children's trajectories, calculate movement behavioral features, and classify the movement behavioral characteristics of children. Our results identify five movement behaviors of children on streets: walking, staying, running, accelerating, and decelerating, demonstrating their use of streets. In addition, the results showed the feasibility of computer vision techniques to identify differences in children's movement behaviors under varying conditions, such as day and time, and the potential of street environments to positively influence children's use of streets from the daily street lives of children. Utilizing vision-based analysis of children's movement behaviors provides meaningful information for improving streets for children toward a better understanding of the use of streets.

Collecting reliable pedestrian trajectories in pedestrian behavior analysis, trajectories broken by frame sampling and trajectories crossing in multi-object conditions often hinder their performance of existing pedestrian tracking models. Despite attempts to address these issues by performing detection and tracking simultaneously using deep learning algorithms, previous methods still struggle with errors such as mistaking a single pedestrian for multiple pedestrians. We propose a novel approach to efficiently collect and correct pedestrian trajectories with minimized practical errors in multi-object conditions for urban surveillance systems. Our system utilizes a single vision sensor to automatically collects trajectories of multiple pedestrians and employ simple, low-computational algorithms, particularly the Deep simple online real-time tracking (Deep SORT) method, to calibrate the trajectories from tracking-by-detection models. Additionally, our system identifies and merges broken pedestrian trajectories, treating them as potential single trajectories, while considering their spatiotemporal ranges. We evaluate the proposed system by implementing it on real testbed video footage. Our method significantly improves practical errors and achieves more accurate pedestrian trajectories compared to existing models, and exhibits robust characteristics, effectively handling complex situations such as occlusions and crowds.

The conceptual framework of child-friendly cities guarantees children's equal access to public urban services. Despite the widespread application of geographical information systems (GISs) and pedestrian network analysis, studies have yet to analyze children's comprehensive pedestrian access to urban services in a large-scale city. This study demonstrates GIS-based approaches to measuring children's pedestrian access to urban services using a pedestrian path layer and the spatial layers of social infrastructure locations in Seoul, South Korea. We show the spatial inequities in children's access to urban services, which depend on the locational characteristics of social infrastructures and the urban development patterns around children. We analyze how children's access to social infrastructures is differentiated by land use composition. Our statistical analysis finds that low-rise residential areas, consisting of impermeable street patterns, increase children's walking distance and restrict children from accessing urban services within their walkable area. In addition, there is potential for key infrastructures such as schools and local community centers to promote pedestrian access to urban services for children. Considering pedestrian access at the street level will help pinpoint vulnerable areas with children who have less access overall and maximize the users served within the service areas of infrastructures.

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.

To efficiently manage a time-consuming and expensive process in traditional urban planning, we aim to develop an artificial intelligence (AI) advisor that assists nonprofessionals participating in the urban planning process by using a generative adversarial network (GAN). This study presents the process of developing the AI model, which suggests appropriate land use plans for user-targeted sites and urban density scenarios. We first create an image dataset that embeds land use, floor area ratio (FAR), and building cover ratio (BCR) information in the RGB channel. Then, an algorithm is developed for establishing an optimized training set with 1000 images and methods for validating the obtained results from both quantitative and visual perspectives. We set up a pilot test to generate three urban density scenarios in Sewoon-Sangga district by constructing urban data-encoded image datasets of Seoul. The pilot test results reveal that our proposed model successfully suggests appropriate land use plans according to the three scenarios. Based on the pilot test, the AI advisor improves its output, training performance, and usability by reflecting block morphology with the Hamming distance and accepting user-designed road patterns. We expect that the novel approach developed in our study will contribute to research on AI-based urban planning.

Different functions of urban spaces generate travel demand, and urban travel inversely influences urban spaces. Collective human mobility patterns can reveal land use patterns and urban structure although previous studies constrained the area around transportation stations. This paper explores the relationship between urban movements and land use patterns of a whole city using taxi ridership data in Seoul, South Korea. This study takes advantages of taxi datasets that are not bounded by routes and stations and contain more precise information in terms of space and time. In addition, we verify the connection between land use and travel patterns using a statistical method. To reveal the interaction between trip mobility patterns and land use, taxi ridership patterns are grouped by a clustering method. Each clustered pattern is then analyzed with the corresponding urban context. We also statistically verify the relationship between land use and the pattern clusters, and show that the seven obtained clusters can be categorized into four types and were significantly correlated with land use. Furthermore, more detailed land use can be classified using the spatiotemporal properties of taxi ridership patterns. The proposed method is expected to monitor change in land use development.

Purpose: In declining cities, it is vulnerable to disaster risks due to the complex reasons such as aging of the buildings, decreases in population and slowly development of infrastructure. When making regeneration plans for declining areas, it is necessary to analyze and predict disaster risks. Therefore, this paper aims to predict the potentials of disaster risks in small-scale declining areas in cities that show risks in the future. Method: This study predicts potential of disaster risk by using cellular automata. By simulating the heavy rainfall disaster using RCP scenarios, 10-year unit change simulation results for heavy rainfall disaster risk are acquired. After simulating disasters that affect to the heavy rainfall disaster, we applied cellular automata to obtain a 10-year unit disaster risk potential. Result: By considering the yearly change values of disaster risk elements in small-scale urban regeneration regions, this study suggests a prediction method of disaster risk potential simulation for heavy rainfall. Our research contributes to the analysis and forecasting of disaster risks in aging small declining areas settings. We expect that the suggested model would be helpful to identify disaster risks and prioritize urgent areas in small decline cities in the future.

Purpose: Declining cities are vulnerable to disasters due to a decrease in population, a decrease in the number of businesses, and the aging of buildings. Residents of declining areas are susceptible to disaster risks, and from an urban planning perspective, it is necessary to identify various disaster risks before establishing regeneration plans for declining areas in various cities. Thus, this paper aims to analyze the risk of disasters in small-scale declining areas in cities that reflect the weights for each disaster risk indicator. Method: This study calculates disaster risk according to the IPCC disaster risk assessment framework. Based on field surveys and spatial statistics, the collected data are integrated into the grid according to the small-scale declining area to classify the grades and obtain risk scores. Next, after evaluating the importance of each risk indicator through expert advice, the disaster risk is analyzed by assigning weights using AHP analysis. In the developed model, a disaster risk analysis was conducted using weights for each disaster risk factor. Result: As a result, the quantified evaluation results were displayed in a 10 m grid for each city. The results of considering the weights for each risk indicator were obtained, and the risk of heavy rainfall and snow disasters specific to each region were obtained. It is expected to use the developed methods in this study as a disaster risk analysis for a small declining area in a city.

With the spread of an aging society, the mobility constrained seniors' need for public transportation is increasing. This paper investigates accessibility to welfare facilities for elderly people, that is, whether elderly people can easily access the welfare facilities that they frequently visit. We consider welfare facilities including residential welfare institutions, medical and welfare institutions, leisure and welfare institutions, and commuting-system welfare facilities. We verify spatial distribution of welfare facilities and relationship to the unequal development of urban or public transportation networks. Inequality of accessibility was found between the urban and suburban areas of the study area. This paper identified the spatial inequity of facilities for elderly people by measuring spatial accessibility considering public transportation accessibility and walking time to the facilities. This paper suggested that improving public transportation accessibility would increase the spatial equity of welfare services for elderly people.

Purpose: Declining cities are vulnerable due to their aging populations and decreasing economic activity. The risk assessment framework developed by IPCC is a representative method of analyzing hazardous impacts in urban spaces. However, most previous research on declining cities and risk assessments are based on macro-scale evaluations. As risk assessment must be included in the urban planning process, development of a risk assessment model for smaller areas is necessary. Thus, this study aims to develop a rainfall disaster risk assessment model for a small area in a declining city. Method: This study selected the appropriate indicators for a small area based on the IPCC disaster risk assessment framework. As suggested by IPCC, multidimensional data are used as indicators of the physical, social, and climate-related health of a city. The indicators include not only computational data based on GIS but practical data obtained through a field study involving a site-specific evaluation. In the developed model, the risk of a rainfall disaster at the study site is quantified with a numerical score. Result: The numerated assessment result is allocated into the grid of the case study site in Daegu, South Korea. The rasterized image visually represents the risk impact score of a rainfall disaster in a small area. Consequently, this study proposes a high-resolution risk impact assessment model that can be applied to urban design and provides the results of realistic and practical rainfall disaster analysis.

Though the technological advancement of smart city infrastructure has significantly improved urban pedestrians’ health and safety, there remains a large number of road traffic accident victims, making it a pressing current transportation concern. In particular, unsignalized crosswalks present a major threat to pedestrians, but we lack dense behavioral data to understand the risks they face. In this study, we propose a new model for potential pedestrian risky event (PPRE) analysis, using video footage gathered by road security cameras already installed at such crossings. Our system automatically detects vehicles and pedestrians, calculates trajectories, and extracts frame-level behavioral features. We use k-means clustering and decision tree algorithms to classify these events into six clusters, then visualize and interpret these clusters to show how they may or may not contribute to pedestrian risk at these crosswalks. We confirmed the feasibility of the model by applying it to video footage from unsignalized crosswalks in Osan city, South Korea.

Efficient evacuation planning is important for quickly navigating people to shelters during and after an earthquake. Geographical information systems are often used to plan routes that minimize the distance people must walk to reach shelters, but this approach ignores the risk of exposure to hazards such as collapsing buildings. We demonstrate evacuation route assignment approaches that consider both hazard exposure and walking distance, by estimating building collapse hazard zones and incorporating them as travel costs when traversing road networks. We apply our methods to a scenario simulating the 2016 Gyeongju earthquake in South Korea, using the floating population distribution as estimated by a mobile phone network provider. Our results show that balanced routing would allow evacuees to avoid the riskiest districts while walking reasonable distances to open shelters. We discuss the feasibility of the model for balancing both safety and expediency in evacuation route planning.

Pedestrian-vehicle accidents are the cause of many human injuries and deaths. To address this challenge, vision-based traffic systems have focused on detecting traffic-related objects' behaviors, such as vehicle position and velocity relative to pedestrians. In this paper, we propose a new and simple model for effectively recognizing overhead front point of vehicles, while only using a single stationary camera capturing from an oblique angle. The proposed system uses faster R-CNN model for detecting object bounding box and mask, projects the mask's extreme points down to find the car's ground front point, and transforms these coordinates from oblique to overhead frame of reference. Our experimental result shows that this method is effective for recognizing overhead front point of car (accuracy: 92.4%) within a certain tolerance.