• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제12권8호
• /
• pp.3769-3789
• /
• 2018

Social pedestrian groups are the basic elements that constitute a crowd; therefore, detection of such groups is scientifically important for modeling social behavior, as well as practically useful for crowd video understanding. A social group refers to a cluster of members who tend to keep similar motion state for a sustained period of time. One of the main challenges of social group detection arises from the complex dynamic variations of crowd patterns. Therefore, most works model dynamic groups to analysis the crowd behavior, ignoring the existence of stationary groups in crowd scene. However, in this paper, we propose a novel unified framework for detecting social pedestrian groups in crowd videos, including dynamic and stationary pedestrian groups, based on spatiotemporal-oriented energy measurements. Dynamic pedestrian groups are hierarchically clustered based on energy flow similarities and trajectory motion correlations between the atomic groups extracted from principal spatiotemporal-oriented energies. Furthermore, the probability distribution of static spatiotemporal-oriented energies is modeled to detect stationary pedestrian groups. Extensive experiments on challenging datasets demonstrate that our method can achieve superior results for social pedestrian group detection and crowd video classification.

• 한국재난정보학회 논문집
• /
• 제7권1호
• /
• pp.12-20
• /
• 2011

본 연구는 군중집회 및 군중눌림 현상의 문헌조사 및 분석에 근거하여 군중집회 및 군중집회에서의 인명피해의 특성을 알아보고자 수행되었다. 대규모 군중집회 시 인명 피해의 특성은 참가자 수와 군중 밀도가 군중집회 종류와 특성에 따라 다른 영향을 미치고 있었다. 군중 집회에 영향을 주는 변수와 그 원인들은 상황에 따라 다양한 정 혹은 부의 영향을 미치고 있었으며 그 변수들은 날씨, 참가자 수, 행사 기간, 실외와 실내, 착석과 이동, 행사 유형, 군중 감정상태, 술 혹은 약물, 군중 밀도, 관련 시설, 참가자 연령 등이었다. 이 중 군중 눌림현상은 실험적으로도 연구가 가능하였고, 사고가 유발되는 물리적 기전으로 보아 군중 압력과 군중 밀도 및 압력의 지속 시간에 영향을 받았으나 사망에 이르는 구체적인 압력 수치를 도출하려면 인간 신체와 관련된 여러 외부적 영향으로 인하여 추가적 연구가 더 필요하다.

• 한국멀티미디어학회논문지
• /
• 제17권9호
• /
• pp.1104-1114
• /
• 2014

In large-scale crowd simulations, it is very important for the decision-making system of manipulating interactive behaviors to minimize the computational cost for controlling realistic behaviors such as collision avoidance. In this paper, we propose a large-scale realtime crowd simulation method using the affordance and navigation potential fields such as attractive and repulsive forces of electromagnetic fields. In particular, the model that we propose locally handles the realistic interactions between agents, and thus radically reduces the cost of expensive computation on interactions which has been the most problematic in crowd simulation. Our method is widely applicable to the expression and analysis of various crowd behaviors that are needed in behavior control in computer games, crowd scenes in movies, emergent behaviors of evacuation, etc.

• 한국방재안전학회논문집
• /
• 제16권1호
• /
• pp.49-60
• /
• 2023

2022년 10월 29일 발생한 이태원 참사 이후 군중압박 사고로 인한 인명피해에 대한 관심이 높아졌으나 국내에서 군중압박과 관련된 학술적, 실제적 기반이 미약함이 지적되었다. 이에 본 연구에서는 군중압박과 관련된 용어와 개념을 조사하고 가능한 한글 용어 후보들을 제안하였으며, 국내외에서 발생한 대표적인 군중압박 사고 사례를 조사하여 정리하였다. 일부 대표적 사례를 기반으로 한 선진국의 접근법들을 조사하였고, 그 중 대표적으로 영상분석, 시뮬레이션 및 설문과 인터뷰 방법을 요약 도출하였다. 이를 통하여 군중압박 사고의 한글 용어 표준화와 개념 정립, 평가 및 접근 방법의 체계화가 이루어지기를 기대하고 있다.

• International Journal of Computer Science & Network Security
• /
• 제24권4호
• /
• pp.26-34
• /
• 2024

Most human emotions are conveyed through facial expressions, which represent the predominant source of emotional data. This research investigates the impact of crowds on human emotions by analysing facial expressions. It examines how crowd behaviour, face recognition technology, and deep learning algorithms contribute to understanding the emotional change according to different level of crowd. The study identifies common emotions expressed during congestion, differences between crowded and less crowded areas, changes in facial expressions over time. The findings can inform urban planning and crowd event management by providing insights for developing coping mechanisms for affected individuals. However, limitations and challenges in using reliable facial expression analysis are also discussed, including age and context-related differences.

• 한국안전학회지
• /
• 제39권3호
• /
• pp.50-59
• /
• 2024

In recent years, building architecture has become increasingly complex and larger in scale to accommodate many people. In densely populated facilities, the interiors are becoming more intricate and high-rise, with narrow corridors, hallways, and stairs. This poses challenges for evacuating occupants in case of emergencies such as fires, making it crucial to assess the evacuation safety in advance. In evacuation safety research, there are significant limitations to theoretical studies owing to their association with crowd behavior and human evacuation characteristics, as well as the risks associated with experiments involving human participants. Consequently, evacuation experiments conducted using simulation-based methodologies are gaining recognition worldwide. However, crowd simulations face validation difficulties because of variations in crowd movement and evacuation characteristics across different cases and scenarios, as well as the challenge of accurately reflecting human characteristics during evacuations. In this study, we investigated validation methods for evacuation simulations using the SAFEGUARD validation data set (SGVDS) provided by the University of Greenwich, UK. The SGVDS collects data on crowd evacuations through actual evacuation tests conducted on ColorLine's large RO-PAX ferry and Royal Caribbean International's cruise ships. The accuracy of the crowd simulations can be validated by comparing SGVDS and crowd simulation results. This study will contribute to the development of highly accurate crowd simulations by verifying various crowd simulations.

• 한국안전학회지
• /
• 제38권2호
• /
• pp.87-95
• /
• 2023

On October 29, 2022, a very large number of people gathered in Itaewondong, Yongsan-gu, Seoul, Korea for a Halloween festival, and as crowds pushed through narrow alleys, 159 deaths and 195 injuries occurred, making it the largest crushing incident in Korea. There have been a number of stampede deaths where crowds gathered at large-scale festivals, event venues, and stadiums, both at home and abroad. When the density increases, the physical contact between bodies becomes very strong, and crowd turbulence occurs when the force of the crowd is suddenly added from one body to another; thus, the force is amplified and causes the crowd to behave like a mass of fluid. When crowd turbulence occurs, people cannot control themselves and are pushed into he crowd. To prevent a stampede accident, investigation and management of areas expected to be crowded and congested must be systematically conducted, and related ministries and local governments are planning to establish a crowd management system to prepare safety management measures to prevent accidents involving multiple crowds. In this study, based on national data, a continuous digital topographic map is modeled in 3D to analyze the risk of crowding and present a plan for selecting high-risk walking routes. Areas with a high risk of crowding are selected in advance based on various data (numerical data, floating population, and regional data) in a realistic and feasible way, and the analysis is based on the visible results from 3D modeling of the risk area. The study demonstrates that it is possible to prepare measures to prevent cluster accidents that can reflect the characteristics of the region.

• 한국안전학회지
• /
• 제38권2호
• /
• pp.96-103
• /
• 2023

Generally, human stampedes and crowd collapses occur when people press against each other, causing falls that may result in death or injury. Particularly, crowd accidents have become increasingly common since the 1990s, with an average of 380 deaths annually. For instance, in Korea, a stampede occurred during the Itaewon Halloween festival on October 29, 2022, when several people crowded onto a narrow, downhill road, which was 45 meters long and between 3.2 and 4 meters wide. Precisely, this stampede was primarily due to the excessive number of people relative to the road size. Essentially, stampedes can occur anywhere and at any time, not just at events, but also in other places where large crowds gather. More specifically, the likelihood of accidents increases when the crowd density exceeds a turbulence threshold of 5-6 /m². Meanwhile, festivals and events, which have become more frequent and are promoted through social media, garner people from near and far to a specific location. Besides, as cities grow, the number of people gathering in one place increases. While stampedes are rare, their impact is significant, and the uncertainty associated with them is high. Currently, there is no scientific system to analyze the risk of stampedes due to crowd concentration. Consequently, to prevent such accidents, it is essential to prepare for crowd disasters that reflect social changes and regional characteristics. Hence, this study proposes using digital topographic maps and crowd-density risk simulations to develop a 3D model of the region. Specifically, the crowd density simulation allows for an analysis of the density of people walking along specific paths, which enables the prediction of danger areas and the risk of crowding. By using the simulation method in this study, it is anticipated that safety measures can be rationally established for specific situations, such as local festivals, and preparations may be made for crowd accidents in downtown areas.

• 한국콘텐츠학회논문지
• /
• 제17권6호
• /
• pp.643-649
• /
• 2017

이 연구는 다중운집행사에 필요한 경찰력을 통원하는데 영향을 미치는 요인들을 비교하는 연구로서 안전한 다중운집행사를 진행하기 위하여 경찰력을 동원하는데 장애요인의 우선순위를 파악하여 경찰력 동원에 과학적인 근거를 마련하기 위한 목적을 가지고 있다. 다중운집행사는 많은 사람들이 특정한 장소에 집중적으로 모여들기 때문에 사소한 원인에 의하여 많은 피해가 발생할 수 있는 특성을 가지고 있다. 이러한 위험을 통제하기 위하여 경찰력을 동원하곤 하는데 위험을 유발할 수 있는 우선순위에 대한 기준은 아직 부족하여 이에 대한 분석이 필요하다. 이 연구에서는 다중운집행사의 안전에 위협에 영향을 미치는 요인을 행사 내부 위험요소, 행사 외부 위험 요소, 교통안전 위협요소로 분류한 기준과 함께 다중운집행사에 동원된 민간경비인력 수준과 참여자를 독립변수로 설정하여 2013년에서부터 2014년까지 동원된 경찰력을 종속변수로 하여 위계적 회귀분석으로 살펴보았다. 그 결과, 다중운집행사의 안전 확보를 위한 경찰력 동원에 가장 크게 영향을 미친 요인은 참가 인원이었으며, 그 다음으로 자체질서요원 인원, 경찰력 동원 사유가 영향을 미치는 것으로 나타났다. 따라서, 다중운집행사의 위험성은 참가인원의 규모가 큰 영향을 미치는 것으로 볼 수 있다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제13권2호
• /
• pp.912-928
• /
• 2019

Crowd behavior analysis research has revealed a central role in helping people to find safety hazards or crime optimistic forecast. Thus, it is significant in the future video surveillance systems. Recently, the growing demand for safety monitoring has changed the awareness of video surveillance studies from analysis of individuals behavior to group behavior. Group detection is the process before crowd behavior analysis, which separates scene of individuals in a crowd into respective groups by understanding their complex relations. Most existing studies on group detection are scene-specific. Crowds with various densities, structures, and occlusion of each other are the challenges for group detection in diverse crowded scenes. Therefore, we propose a group detection approach called Collective Interaction Filtering to discover people motion interaction from trajectories. This approach is able to deduce people interaction with the Expectation-Maximization algorithm. The Collective Interaction Filtering approach accurately identifies groups by clustering trajectories in crowds with various densities, structures and occlusion of each other. It also tackles grouping consistency between frames. Experiments on the CUHK Crowd Dataset demonstrate that approach used in this study achieves better than previous methods which leads to latest results.