Crowd Gathering Research Articles

Intragroup dynamics and splitting characteristics of social group movement through a funnel-shaped bottleneck

Social groups are common in crowd gatherings and events. In this study we explore the trajectories and interactions of groups consisting of more than 4 pedestrians. Although previous studies indicate that such groups have a tendency to split in free space, precise empirical data for bottleneck movement is still limited. Here we fill this gap by performing exploring experiments on the dynamics of groups of 5, 6 and 7 persons through funnel-shaped bottlenecks of two different widths. The intragroup dynamics and the splitting characteristics of groups are analyzed on macroscopic, mesoscopic and microscopic scales. The macroscopic scale results reveal that increasing bottleneck width improves group speed and reduces severe deformations. The bottleneck width significantly affects group shape, while group size significantly affects group speed in narrow bottlenecks. The mesoscopic scale results confirm the existence of mesoscopic components by identifying the heterogeneous bonds between each pair of adjacent sequence members, i.e. stable and unstable bonds. The sub-individual is only connected by unstable bonds and subgroup is formed of stable bonds. The reduction in bottleneck width negatively affects the average speed of subgroups but has little impact on their spatial requirements and members’ relative positions. So the stable spatial characteristics of subgroups are measured as the spatial threshold for the splitting of unstable bonds. The microscopic scale results investigate the split state of unstable bonds within the group. The spatial distributions of split states differ evidently between 5-person and 6, 7 persons groups. The split behavioral strategies of groups are more sensitive to intragroup spatial restriction for narrow bottleneck (0.8 m), but more sensitive to environmental restriction for wide bottleneck (1.5 m). The intragroup disturbance caused by splitting is assessed exploringly based on the concept of ‘energy conversion-spillover’. The results can help improve crowd management, better design of bottlenecks, and more realistic description of group dynamics in modeling approaches.

An unsupervised group detection method for understanding group dynamics in crowds

Pedestrian groups arrive in large numbers in crowd gatherings, especially of a spiritual nature. Various studies have been done on crowd control in public spaces by analysing the behaviour of pedestrian groups. Understanding group dynamics can help better plan pedestrian facilities and large events. Many existing group sensing models primarily determine social bonding between pedestrians using spatiotemporal parameters, such as distance, directional movement, and overlapping time. However, social bonding determined based on these parameters assumes the bonding to be symmetric, spatially and temporally static and is unaffected by neighbourhood. Our study addresses the issue by relaxing such assumptions and developing an unsupervised group detection model based on potential candidates. The proposed model can handle temporal and spatial variations more effectively than those based on simple spatiotemporal parameters. The model developed is assessed both quantitatively and qualitatively. New metrics are introduced for quantitative evaluation, comparing predicted groups and ground truth instead of pedestrian pairs with ground truth. A visualisation method is developed for the qualitative assessment. Group splits and group merges are calculated to assist in understanding crowd movement patterns. Overall, this study helps in further exploring and assessing groups, which can improve understanding of crowd dynamics.

Enhancing Urban Resilience Through Spatial Interaction-Based City Management Zoning

Good city management is essential in mitigating the impact of various crisis events and thus enhances urban resilience. The current zoning system that underlies China’s city management system, however, hardly meets the resilient requirements in failing to appropriately and flexibly allocate patrol personnel under normal and various crisis scenarios. We propose an optimization method that gives rise to a resilient city management zoning system by introducing spatial interaction. We realize this through community detection in a spatially embedded patrol passage cost network. Illustrated by a case in Jiangbei District, Ningbo, China, we show that the optimized zoning system has significant advantages in terms of event coverage, response efficiency, and workload balance in the normal scenario as compared to the status quo with up to 3.6 percent, 5.9 percent, and 34.0 percent performance improvement, respectively. Moreover, the method also achieves similar performance improvement in three typical shock scenarios in city management: epidemic, typhoon, and crowd gathering. We conclude the article with discussions on the findings’ significance in urban resilience building and their methodological and theoretical implications in applied urban sciences.

RETRACTED: Multi-scale fusion public gathering recognition based on residual network

This article has been retracted. A retraction notice can be found at https://doi.org/10.3233/JIFS-219433.

An Innovative Multi-Objective Rescheduling System for Mitigating Pandemic Spread in Aviation Networks

The novel coronavirus outbreak has significantly heightened environmental costs and operational challenges for civil aviation airlines, prompting emergency airport closures in affected regions and a substantial decline in ridership. The consequential need to reassess, delay, or cancel flight itineraries has led to disruptions at airports, amplifying the risk of disease transmission. In response, this paper proposes a spatial approach to efficiently address pandemic spread in the civil aviation network. The methodology prioritizes the use of a static gravity model for calculating route-specific infection pressures, enabling strategic flight rescheduling to control infection levels at airports (nodes) and among airlines (edges). Temporally, this study considers intervals between takeoffs and landings to minimize crowd gatherings, mitigating the novel coronavirus transmission rate. By constructing a discrete space–time network for irregular flights, this research generates a viable set of routes for aircraft operating in special circumstances, minimizing both route-specific infection pressures and operational costs for airlines. Remarkably, the introduced method demonstrates substantial savings, reaching almost 53.4%, compared to traditional plans. This showcases its efficacy in optimizing responses to pandemic-induced disruptions within the civil aviation network, offering a comprehensive solution that balances operational efficiency and public health considerations in the face of unprecedented challenges.

Research on spatiotemporal characteristics of urban crowd gathering based on Baidu heatmap

With the rapid development of urban transportation and the increase in per capita car ownership, the problem of urban traffic congestion is becoming increasingly prominent. Due to the uneven distribution of crowd in different regions of the city, it is difficult to determine and solve the traffic dynamics congestion. In order to solve the problem that it is difficult to determine the dynamics of traffic congestion areas caused by uneven distribution of vitality in different regions of mountainous cities, a crowded mega mountainous city is selected as research object and it proposes a model to calculate the change characteristics of regional crowd gathering. Baidu Heatmap is used as it could distinguish crowd gathering in certain urban core area. The heat map pictures in dozens of consecutive days is extracted and researchers conducted pixel statistical classification on thermal map images. Based on the pixel data of different levels of the pictures, the calculation model is established and an algorithm based on particle swarm optimization is proposed. The calibration of the relative active population equivalent density is conducted, and the distribution characteristics of crowd gathering in time and space are analyzed. The results show that there are obvious spatiotemporal characteristics for this selected city. In time, holidays have an important impact on crowd gathering. The peak time of crowd gathering on weekdays is different from that on rest days. The research in this paper has a direct practical value for the identification of traffic congestion areas and the corresponding governance measures. The dynamic identification of population gathering areas in mountainous mega cities, demand prediction for various transportation regions, and future population OD(Origin—Destination) planning are of great significance.

A chaotic-based watermarking scheme for ensuring integrity of a face recognition system in public large gathering scenario

Secure data transmission is critical in public video surveillance applications such as tracking missing persons in large crowd gatherings. It may help the security personnel to monitor & record the live feeds from multiple cameras. However, this live streaming could be falsified. Various cryptographic techniques have been proposed to ensure the integrity of transmitted images, but continuous attacks still pose serious challenges to ensuring the integrity and authenticity of transmitted images. This work proposes a new chaotic-based watermarking technique to ensure the integrity of the transmitted video frames, and since the frames come from multiple cameras, the scheme should be as simple as possible and as fast as possible, the simplicity manifests in designing a fragile watermarking scheme therefore, the watermark is to be embedded in the spatial domain by direct manipulation of the values of the cover’s pixels. To ensure the security of the proposed scheme, chaotic maps have been deployed to reach an acceptable level of security with minimum computational overhead. The proposed scheme is novel in design in terms of the chaotic maps used and it is tested in challenging large gathering scenario. Experimental results demonstrate that the proposed method successfully locates any falsification in the transmitted images, and the performance of the scheme evaluates different types of attacks. Our testing results in good rate tamper of detection with TPR=99.2005, TNR=99.0777 and with a good quality of imperceptibility for the watermarked image with PSNR=43.82dB, SSIM=00.9955.

Risk analysis of crowd gathering on metro platforms during large passenger flow

With the continuous development of urbanization, metro systems have become an essential means of travel for urban residents. However, the increase of passengers using the metro system also increases the risk of occupant instability and stampede, which is especially significant on metro platforms considering the mixed passenger flow of boarding, waiting and getting off. How to grasp the overcrowding of passengers and gathering risk on the metro platform in time is the critical problem of the current research. To solve the difficulties in acquiring the number of passengers at metro platforms, a 7-phase model is constructed to estimate the risk of crowd gathering in each platform according to the Origin-Destination (OD) data. The proposed model divides a travel process into 7 components in detail and evaluates the crowd-gathering risk accurately to the platform. The travel time of each passenger is calculated by the station structure parameters and pedestrian parameters. Then, the travel stages of passengers are acquired by the travel time. Thirdly, the number of passengers on any platform and at any time is calculated. Finally, the temporal and spatial distribution of crowd-gathering risk on all metro platforms of a metro network is analyzed. The constructed passenger travel model and the crowd-gathering risk analysis method can effectively identify the high-risk platform and high-risk periods, which provides method and data support for metro operation safety improvement and crowd accident prevention.

Evaluation of Urban Rail Transit Station Effectiveness Based on TOD Principle -Taking Shanghai as an Example

The rapid development of China's cities has led to a variety of traffic problems. Building a public transportation-oriented urban space development model and playing the role of rail transit in urban development are important measures for the upgrading and construction of China's cities. In order to evaluate the development effect of TOD mode of existing rail transit stations, this paper combines the concept of TOD development and 5D principles to establish a set of urban rail transit station effectiveness evaluation index system, determines the four first-level indicators and 13 second-level indicators based on transportation, society, economy and environment, uses the analytic hierarchy method to determine the weights, and analyzes 30 stations of Shanghai Metro Line 2 as a case study.Based on the Arcgis spatial analysis technology, the data is organized and the indexes are calculated. The sites with the strongest TOD ability are Nanjing East Road and Jing'an Temple, the worst sites are Pudong International Airport and Haitian Third Road, their gap mainly lies in the level of economic development around the stations and the degree of crowd gathering,and finally the paper combined with the evaluation results to give development suggestions, this method can be used to study the level of development of rail transit stations in different cities and different regions. This method can be used to study the development level of rail transit stations in different cities and regions, and give suggestions for improving the TOD situation of stations.

The Monumental Crowd: Collective Attention in Symbolic and Material Spaces

Abstract: Although often ignored in one's daily routine, monuments can serve as potent magnets for crowd gatherings in unveilings, preservation, and protest. The materiality of the monument provides a point of collective attention that binds participants in common cause within a civic space. When successful, these gatherings depend on enacting a set of processes: planning, organizing, gathering, performing, and dispersing. Each occurs within the context of state sponsorship or opposition. As is evident from cases in colonial New York and contemporary Virginia, monuments, no longer mute, can speak to crowds with shared images of morality and justice in their metal and marble materiality.

Crowd Characterization in Surveillance Videos Using Deep-Graph Convolutional Neural Network.

Crowd behavior is a natural phenomenon that can provide valuable insight into the crowd characterization process. Modeling the visual appearance of a large crowd gathering can reveal meaningful information about its dynamics. Parametric modeling can be used to develop efficient and robust crowd monitoring systems. A crowd can be structured or unstructured based on the organization. In this article, crowd characterization has been mapped to a graph classification problem to classify movements based on order parameter ( ϕ ), active force components, and steadiness (Reynolds number). The graphs are constructed from the motion groups obtained using an active Langevin framework. These graphs are processed using a deep graph convolutional neural network for crowd characterization. For experimentation, we have prepared a dataset comprising of videos from popular publicly available datasets and our own recorded videos. The proposed framework has been compared with the latest deep learning-based frameworks in terms of accuracy and area under the curve (AUC). We have obtained a 4%-5% improvement in accuracy and AUC values over the existing frameworks. The insights obtained from the proposed framework can be used for better crowd monitoring and management.

A crowd clustering prediction and captioning technique for public health emergencies.

The COVID-19 pandemic has come to the end. People have started to consider how quickly different industries can respond to disasters due to this public health emergency. The most noticeable aspect of the epidemic regarding news text generation and social issues is detecting and identifying abnormal crowd gatherings. We suggest a crowd clustering prediction and captioning technique based on a global neural network to detect and caption these scenes rapidly and effectively. We superimpose two long convolution lines for the residual structure, which may produce a broad sensing region and apply our model's fewer parameters to ensure a wide sensing region, less computation, and increased efficiency of our method. After that, we can travel to the areas where people are congregating. So, to produce news material about the present occurrence, we suggest a double-LSTM model. We train and test our upgraded crowds-gathering model using the ShanghaiTech dataset and assess our captioning model on the MSCOCO dataset. The results of the experiment demonstrate that using our strategy can significantly increase the accuracy of the crowd clustering model, as well as minimize MAE and MSE. Our model can produce competitive results for scene captioning compared to previous approaches.

Survey of Crowd Crush Disasters and Countermeasures

Introduction:On the night of October 29, 2022, a crowd crush occurred during Halloween festivities in the Itaewon neighborhood of Seoul, Korea. At least 156 people were killed and at least 173 others were injured. In this study, the author tried to learn a lesson by investigating the worldwide crowd crush disaster and analyzing the differences and results.Method:First, the crowd crush disasters were investigated and summarized through literature and internet searches. Second, based on this, the prevention and management of crowd crush disasters, emergency medical response, and necessary research/development contents were derived through a Delphi survey of experts.Results:Crowd crush disasters have been experienced from developed countries to developing countries since the 1800s. Commonly the crowd density was high, and the crowds continued to move and then the crowd collapsed above a certain limit. The biggest casualty occurred during a pilgrimage to Mecca in Saudi Arabia in 2015, but the theme of the event, such as concerts, sporting events, and funerals, was varied. Experts survey was that the manager's efforts not to increase the crowd density, and efforts to maintain order and prevent contingencies were important. They said that it is important to comply with the principles of disaster medicine, but it is difficult to access the patient in the crowd crush state, so the management of the crowd may be more important. They said that it is necessary to establish a realistic guideline and a real-time crowd density monitoring system using CCTV or drones.Conclusion:Crowd crush disasters can occur in any type of crowd gathering events where the crowd density increases, and prevention through crowd management and real-time crowd density monitoring should be implemented.

Exploring the Quantitative Assessment of Spatial Risk in Response to Major Epidemic Disasters in Megacities: A Case Study of Qingdao.

With the global spread of various human-to-human epidemics, public health issues have become a focus of attention. Therefore, it is of great importance to improve the quantitative risk assessment of the construction of resilient cities in terms of epidemic disasters. Starting with the dimensions of social activities and material space, this paper took Qingdao, China, with a population of 5 million, as an example, and took its seven municipal districts as the research scope. In this paper, five risk factors, including the Population density index, Night light index, Closeness index of roads, Betweenness index of roads and Functional mixed nuclear density index were selected for weighted superposition analysis. We conducted a quantitative assessment of the spatial risk of epidemic disaster so as to obtain the classification and spatial structure of the epidemic disaster risk intensity. The results show that: ① The roads with a large traffic flow are most likely to lead to the risk of urban spatial agglomeration, and the areas with a large population density and large mixture of infrastructure functions are also important factors causing the risk of epidemic agglomeration. ② The analysis results regarding the population, commerce, public services, transportation, residence, industry, green space and other functional places can reflect the high-risk areas for epidemic diseases with different natures of transmission. ③ The risk intensity of epidemic disasters is divided into five risk grade areas. Among them, the spatial structure of epidemic disasters, composed of the first-level risk areas, is characterized by "one main area, four secondary areas, one belt and multiple points" and has the characteristics of spatial diffusion. ④ Catering, shopping, life services, hospitals, schools and transportation functional places are more likely to cause crowd gathering. The management of these places should be focused on prevention and control. At the same time, medical facilities should be established at fixed points in all high-risk areas to ensure the full coverage of services. In general, the quantitative assessment of the spatial risk of major epidemic disasters improves the disaster risk assessment system in the construction of resilient cities. It also focuses on risk assessment for public health events. It is helpful to accurately locate the agglomeration risk areas and epidemic transmission paths that are prone to outbreak or cause epidemic transmission in cities so as to assist the relevant practitioners in containing the epidemic from the initial stage of transmission in a timely manner and prevent the further spread of the epidemic.

Sparrow Search Optimization with Transfer Learning-Based Crowd Density Classification

Due to the rapid increase in urbanization and population, crowd gatherings are frequently observed in the form of concerts, political, and religious meetings. HAJJ is one of the well-known crowding events that takes place every year in Makkah, Saudi Arabia. Crowd density estimation and crowd monitoring are significant research areas in Artificial Intelligence (AI) applications. The current research study develops a new Sparrow Search Optimization with Deep Transfer Learning based Crowd Density Detection and Classification (SSODTL-CD2C) model. The presented SSODTL-CD2C technique majorly focuses on the identification and classification of crowd densities. To attain this, SSODTL-CD2C technique exploits Oppositional Salp Swarm Optimization Algorithm (OSSA) with EfficientNet model to derive the feature vectors. At the same time, Stacked Sparse Auto Encoder (SSAE) model is utilized for the classification of crowd densities. Finally, SSO algorithm is employed for optimal fine-tuning of the parameters involved in SSAE mechanism. The performance of the proposed SSODTL-CD2C technique was validated using a dataset with four different kinds of crowd densities. The obtained results demonstrated that the proposed SSODTL-CD2C methodology accomplished an excellent crowd classification performance with a maximum accuracy of 93.25%. So, the proposed method will be highly helpful in managing HAJJ and other crowded events.

Characteristics of air quality changes in Quanzhou during the COVID-19 epidemic

On January 8, 2020, there was an outbreak of coronavirus disease 19 (COVID-2019), and on January 24, 2020, Fujian Province issued the "Fujian Provincial Emergency Plan for Public Health Emergencies", which restricted crowd gathering activities such as markets, greatly reducing transportation and industry. The difference between air quality before and after the epidemic was studied, and the results show that the mass concentrations of SO2, NO2, CO, PM10, and PM2.5 in Quanzhou City in February 2020 all decreased in January, with decline rates of 5.8–8.8%, 34.7–46.7%, 14.9–25.1%, 9.4–15.9%, and 5.4– 14.3%; O3 did not show the same change pattern. NO2, CO, O3, PM10, and PM2.5 decreased compared with the same period in 2018 and 2019, of which SO2 and NO2 declined the most, and CO and O3 declined the least. The spatial distribution map of air quality in Quanzhou City in the first half of 2020 was obtained by the Arcgis anti-distance weighting method, and the results showed that the Qingyuanshan area had the best air quality in Quanzhou and was the least affected by the epidemic; the air quality in Tushan Street and Jintoupu changed the most, showing the most obvious impact of the epidemic. The changes in air quality in Quanzhou during the epidemic period show that traffic and industrial pollution sources were more likely to be affected by the epidemic, and the sources of domestic pollution were relatively stable.

Implementing caregiver management measures in general hospitals to prevent the COVID-19 pandemic.

To summarize the whole process management measures for caregivers of inpatients in a non-new coronavirus pneumonia designated hospital in China during the novel coronavirus pneumonia epidemic. The implementation of these measures is mainly to prevent the virus pandemic caused by crowd gathering. A quasi-experimental study. Novel coronavirus pneumonia prevention and control measures were implemented in the 'pre-hospital link, hospitalization link and discharge link' for the whole process of the hospitalized patients' caregivers. To evaluate the effects by time point inspection, the results were fed back using information technology for quality improvement. The results of three time point inspections indicated that the management quality of many projects improved continuously (p < 0.05). From January to June of 2021, 20 departments implemented a facial recognition information management system. The ratio of patients/caregivers admitted was 1:0.528~1:0.965; It was found that it is effective to implement the whole process management measures. The facial recognition system plays a key role in the comprehensive management of the caregivers.

Review of analyses on crowd-gathering risk and its evaluation methods

Overcrowding and stampedes may occur in public places with the gathering of crowds. To mitigate and prevent risk, the accident mechanism and methods for monitoring and evaluating crowd-gathering risk were investigated. Related studies are reviewed and summarized in this paper. The evolution process of crowd-gathering risk and precipitating factors were explained systematically. Risk monitoring methods are classified into three types according to the key technologies adopted. Articles exploring risk evaluation methods for crowd gathering are outlined, and the three main paradigms were formed. Finally, the shortcomings and future research points are summarized to promote more in-depth and comprehensive studies on crowd-gathering risk, develop monitoring technologies, and build an integrated system of risk management.

How do urban services facilities affect social segregation among people of different economic levels? A case study of Shenzhen city

Social segregation hinders the development of cities and has become a hot topic in urban research. Existing studies have focused on the difference in the distribution of crowd activities to measure segregation but have ignored the impact of the urban environment on crowd gathering and segregation. To study the impact and understand social segregation more comprehensively, we coupled mobile phone datasets and housing price data to divide city dwellers into three socioeconomic levels. Considering that spatial colocation is a necessary condition for interaction among various social groups, spatial colocation probability was proposed to quantitatively describe the degree of social segregation at the community scale. Point-of-interest (POI) data were introduced to represent the urban service facilities. The effect of urban service facilities on the segregation of different groups was analyzed by using geographically weighted regression (GWR). The results indicate three points, as follows. (1) Significant social segregation in Shenzhen mostly occurs in suburban and downtown areas, and the interaction segregation of people mainly occurs between people with high and low socioeconomic levels. (2) More economically inclusive and necessary service facilities (e.g., medical and insurance companies) can promote crowd interaction and ease the segregation of social activities. (3) The impact of service facilities on the interaction of various social groups is related to the development of the area where the activities occur, and the most significant impact is in high-tech industrial zones. This study quantitatively calculated the impacts of different service facilities on different groups of people in different communities and times. From the results, detailed and reasonable suggestions were made for city planners.

Third- and First-Person Effects of COVID News in HBCU Students’ Risk Perception and Behavioral Intention: Social Desirability, Social Distance, and Social Identity

ABSTRACT With COVID-19’s unprecedented impact and ongoing debates on prosocial social distancing and antisocial beach crowd gathering, this study examined the roles of social desirability, social distance marked by race, and social identity in the third-person effect (TPE) of COVID-related news concerning risk perception among Historically Black College and University (HBCU) students. The findings confirmed third- and first-person effects with the perceived influence of antisocial (beach partying) news greater for others than themselves, whereas the self was perceived to be more responsive to the influence of prosocial (social distancing) news. A symmetric interaction between social desirability and social identity indicated that the racial out-group, non-HBCU students in Florida were perceived as more affected by antisocial beach party news, whereas the in-group HBCU peers were perceived as more influenced by desirable social distancing warning news and COVID news in general. The first-person perception and shrunk self-other gap in COVID news influence on HBCU students” concerns may be associated with perceived personal vulnerability and responsibility as well as a political reflection on the government’s handling of COVID. While they presumed favorable media influence on self and the in-group to preserve collective self-esteem, they made realistic threat judgments from perceiving undesirable media influence on their in-group peers’ risky behavior tendency and intended to take preventive actions (staying home) to avoid the health risk. Further, it was the first-person perceptual gap between themselves and the out-group, non-HBCU students that influenced them to heed social distancing warnings.