Crowd Events Research Articles

ReCAP: A distributed CAPTCHA service at the edge of the network to handle server overload

AbstractWeb server overload resulting from an application layer–based distributed denial‐of‐service (DDoS) attack or a flash crowd event continues to be a major problem in today's internet because it renders the Web server unavailable in both cases. In this paper, we propose a novel system, called ReCAP, that handles server overload resulting from application layer–based DDoS attacks or flash crowd events. The system is envisioned as a service that can be provided to websites that have limited resources with no infrastructure in place to handle these events. The main goal of ReCAP is to filter attack traffic in case of a DDoS attack event and to provide users with basic information during a flash crowd event. The proposed system is composed of 2 main modules: (1) the HTTPredirect module, which is a stateless Hypertext Transfer Protocol server that redirects Web requests destined to the targeted Web server to the second module, and (2) the distributed Completely Automated Public Turing Test To Tell Computers and Humans Apart (CAPTCHA) service, which comprises a large number of powerful nodes geographically and suitably distributed in the internet acting as a large distributed firewall. All requests to the origin Web server are redirected to the CAPTCHA nodes, which can segregate legitimate clients from automated attacks by requiring them to solve a challenge. Upon successful response, legitimate clients (humans) are forwarded through a given CAPTCHA node to the Web server. These CAPTCHA proxies are envisioned to be placed intrinsically at the edge of the network in the proximity of the clients to curb communication delays, and thus perceived response times, and to relieve the core network from further traffic congestion. In particular, such organization fits squarely in the fifth use case scenario presented in the European Telecommunications Standards Institute Mobile Edge Computing Industry Specification Group's introductory technical paper on Mobile‐Edge Computing. In conclusion, the performance evaluation shows that the proposed system is able to mitigate application‐layer DDoS attacks while incurring acceptable delays for legitimate clients as a result of redirecting them to and via CAPTCHA nodes.

SPARTA: A survival performance degradation framework for identity federations

Identity federations simplify user’s access control across different networks, domains or systems. These federations allow users to seamlessly access data from another domain and they avoid the need of a completely redundant user administration. Federations rely on Identity Providers (IdPs) to manage user’s identities. However, IdPs are prone to Distributed Denial-of-Service (DDoS) attacks and flash crowd events. Those attacks and events can severely compromise the performance of IdPs, affecting legitimate users. Existing solutions either ignore such events, statically improving the performance of only specific IdP operations, or tolerate a predetermined number of failures, employing extra hardware resources purchased to replicate IdPs services. This article presents SPARTA, a Survival Performance degrAdation fRamework for idenTity federAtions. SPARTA offers identity federation survivability employing the collective intelligence principles. We showcase the framework over a real identity management system. Results from the experiments show the improvements of the system under attacks. We measure improvements by identity authentication latency (i.e., the time interval between the authentication request and its response) and throughput. As future works, we intend to evaluate our solution using large-scale identity federations.

Abnormal Prediction of Dense Crowd Videos by a Purpose–Driven Lattice Boltzmann Model

AbstractIn the field of intelligent crowd video analysis, the prediction of abnormal events in dense crowds is a well-known and challenging problem. By analysing crowd particle collisions and characteristics of individuals in a crowd to follow the general trend of motion, a purpose-driven lattice Boltzmann model (LBM) is proposed. The collision effect in the proposed method is measured according to the variation in crowd particle numbers in the image nodes; characteristics of the crowd following a general trend are incorporated by adjusting the particle directions. The model predicts dense crowd abnormal events in different intervals through iterations of simultaneous streaming and collision steps. Few initial frames of a video are needed to initialize the proposed model and no training procedure is required. Experimental results show that our purpose-driven LBM performs better than most state-of-the-art methods.

A Survey of Video-Based Crowd Anomaly Detection in Dense Scenes

Population growth has made the probability of incidents at large-scale crowd events higher than ever. In the past decades, automated crowd scene analysis done by computer vision has attracted attention. However, severe occlusions and complex crowd behaviors make such analysis a challenge. As a key aspect of crowd scene analysis, a number of works dealing with dense crowd anomaly detection based on computer vision have been presented. This work is a survey of computer vision techniques for analyzing dense crowd scenes. It covers two aspects: crowd density estimation and abnormal event detection. Some problems and perspectives are discussed at the end.

Pleasures of the Vanguard

Pleasures of the Vanguard

Exploiting Delay-Aware Load Balance for Scalable 802.11 PSM in Crowd Event Environments

This paper presents ScaPSM (i.e., Scalable Power-Saving Mode Scheduler), a design that enables scalable competing background traffic scheduling in crowd event 802.11 deployments with Power-Saving Mode (PSM) radio operation. ScaPSM prevents the packet delay proliferation of previous study, if applied in the crowd events scenario, by introducing a new strategy of adequate competition among multiple PSM clients to optimize overall energy saving without degrading packet delay performance. The key novelty behind ScaPSM is that it exploits delay-aware load balance to control judiciously the qualification and the number of competing PSM clients before every beacon frame’s transmission, which helps to mitigate congestion at the peak period with increasing the number of PSM clients. With ScaPSM, the average packet delay is bounded and fairness among PSM clients is simultaneously achieved. ScaPSM is incrementally deployable due to only AP-side changes and does not require any modification to the 802.11 protocol or the clients. We theoretically analyze the performance of ScaPSM. Our experimental results show that the proposed design is practical, effective, and featuring with significantly improved scalability for crowd events.

Abnormal crowd event detection by fusing saliency information and social force model

Abnormal event detection plays an important role in intelligent video surveillance. A new abnormal behavior detection algorithm is presented by fusing spatiotemporal features. We first extract SI as the feature representation in the spatial domain. Then, by combining the high precision optical flow algorithm with social force model, we calculate the interaction force as the feature representation in the temporal domain. A novel motion feature descriptor, i.e., Histogram of Interaction Force (HOIF) is proposed, which is combined with SI as temporal-spatial features to be input to the Support Vector Machine (SVM) to identify the crowd events. The effectiveness of the proposed algorithm is put to test on the UMN dataset, and the experimental results indicate that the presented method offers more reliable performance than some existing algorithms in terms of accuracy and robustness.

A review on technological advancements in crowd management

With more and more people attending public gatherings, and with a continuous rise in crowd density in urban areas, the management of crowd has become more challenging than ever before. Every year, many people lose their lives due to inefficient crowd planning and management. Crowd management is an interdisciplinary area, and it requires understanding of engineering and technological aspects, along with an understanding of crowd behavior and crowd flow management, i.e. psychological and sociological aspects. This paper presents a broad, but not exhaustive overview of the recent technological advancements in the area of crowd planning and monitoring techniques for an effective crowd management system. It discusses the crowd modeling aspects during the planning of crowded scenario, and the technological advancements in crowd data acquisition techniques [based on Vision, Wireless/Radio-Frequency (RF) and Web/Social-media data mining technologies] during execution of crowded event. The paper also considers technological applications in some highly crowded scenarios on earth as case studies, along with future research directions in the area.

Zika Virus Disease.

The Zika virus is an arbovirus belonging to the virus family Flaviviridae. The virus was isolated in 1947 from a rhesus monkey in the Zika Forest of Uganda. The virus causes sporadic mild human infections in Africa and later in Asia. However, by 2007 a major shift in its infection pattern was noticed and thousands of human infections were reported in the State of Yap and Federated States of Micronesia. In the last 3 years, major outbreaks have continued to occur and the virus has spread to several Pacific and American countries. These outbreaks were mostly asymptomatic; however, there were more severe clinical signs associated with the infections. Those signs included microcephaly and Guillain–Barre syndrome. It is believed that various species of mosquitoes can biologically transmit the virus. However, Aedes aegypti is most widely associated with the Zika virus. Recently, new modes of virus transmission have been reported, including mother-to-fetus, sexual, blood transfusion, animal bites, laboratory exposure and breast milk. Differential diagnosis is very important as some other arboviruses such as yellow fever virus, West Nile virus, dengue virus, and chikungunya virus have similar clinical manifestations to the Zika virus infection as well as relating serologically to some of these viruses. Established laboratory diagnostic tests to detect the Zika virus are limited, with reverse transcription polymerase chain reaction being the most widely used test. Taking into consideration the quickness of the spread of infection, size of the infected population and change of the infection severity pattern, the Zika virus infection merits collective efforts on all levels to prevent and control the disease. Limited research work and data, concurrent infection with other arboviruses, involvement of biological vectors, mass crowd events, human and trade movements and lack of vaccines are some of the challenges that we face in our efforts to prevent and control the Zika virus infection.

When is policing fair? Groups, identity and judgements of the procedural justice of coercive crowd policing

ABSTRACTProcedural justice theory (PJT) is now a widely utilised theoretical perspective in policing research that acknowledges the centrality of police ‘fairness’. Despite its widespread acceptance this paper asserts that there are conceptual limitations that emerge when applying the theory to the policing of crowd events. This paper contends that this problem with PJT is a result of specific assumptions that are highlighted by two studies using a novel experimental approach. Study 1 systematically manipulated the social categories used to describe crowd participants subjected to police coercion. The experiment demonstrates how these social categories dramatically affected participants’ perceptions of the same police action and that it was participants’ relational identification with the police, rather than a superordinate category, that mediated the association between judgements of procedural fairness and intentions to cooperate. In Study 2, using a quasi-experimental design, we then replicated and extended these findings by demonstrating how perceptions of procedural fairness are also influenced by levels of in-group identification. The paper concludes by exploring the implications of the data for reconceptualising the social psychological processes mediating these judgements and impacts of police legitimacy.

On current crowd management practices and the need for increased situation awareness, prediction, and intervention

Recent accidents (News, 2006, 2010, 2013, 2015) show that crowded events can quickly turn into tragedies. The goal of crowd management is to avoid such accidents through careful planning and implementation. Crowd management practices are collaborative efforts between the different actors of the crowd management team and the crowd that depend on effective handling, sharing, and communication of information. Safety and comfort of a crowd depend on the success of such efforts. We have studied current practices and the role of technology through interviews to crowd managers. Our findings show that event planning and monitoring can be complex and sophisticated, but are operated with little support from technology. Crowd managers intend to increase their use of technology, but they have been so far dissatisfied by existing solutions. We provide recommendations for a bigger role of technology in crowd management.

Characterizing and Optimizing Cellular Network Performance During Crowded Events

During crowded events, cellular networks face voice and data traffic volumes that are often orders of magnitude higher than what they face during routine days. Despite the use of portable base stations for temporarily increasing communication capacity and free Wi-Fi access points for offloading Internet traffic from cellular base stations, crowded events still present significant challenges for cellular network operators looking to reduce dropped call events and improve Internet speeds. For an effective cellular network design, management, and optimization, it is crucial to understand how cellular network performance degrades during crowded events, what causes this degradation, and how practical mitigation schemes would perform in real-life crowded events. This paper makes a first step toward this end by characterizing the operational performance of a tier-1 cellular network in the U.S. during two high-profile crowded events in 2012. We illustrate how the changes in population distribution, user behavior, and application workload during crowded events result in significant voice and data performance degradation, including more than two orders of magnitude increase in connection failures. Our findings suggest two mechanisms that can improve performance without resorting to costly infrastructure changes: radio resource allocation tuning and opportunistic connection sharing. Using trace-driven simulations, we show that more aggressive release of radio resources via 1-2 s shorter radio resource control timeouts as compared with routine days helps to achieve better tradeoff between wasted radio resources, energy consumption, and delay during crowded events, and opportunistic connection sharing can reduce connection failures by 95% when employed by a small number of devices in each cell sector.

Access control mechanism using flexible multi‐layer structure for mesh‐based P2P live streaming systems

SummaryWhen thousands of new peers seek to join the peer‐to‐peer (P2P) system within a very short time (i.e., the so‐called flash crowd event), most of the peers suffer a long startup delay as a result of peer over‐competition. Accordingly, recent studies have proposed a slot‐based user access control (UAC) mechanism, which periodically admits a certain number of new peers to the system, and a user batch join (UBJ) mechanism, which preconstructs the new peers into a fixed‐size tree structure before peer join process. However, in the UAC mechanism, it is difficult to determine the optimal time slot length; while in the UBJ mechanism, it is difficult to determine the optimal tree size. Accordingly, the present study proposes a structured access control (SAC) mechanism, in which the new‐arriving peers are preconstructed to a flexible multilayer structure to overcome the limitation of a fixed‐size tree structure, and the number of peers in each layer of the structure is determined based on a consideration of the remaining upload bandwidth of the P2P system. Based on the assumption of a heterogeneous upload bandwidth capability of the peers, an analytical model is derived to describe the growth behavior of a P2P system with access control. It is shown that the simulation results for the growth rate of a P2P system based on the SAC mechanism are in good agreement with those obtained from the analytical model. In addition, the simulation results show that the SAC mechanism outperforms the UAC and UBJ mechanisms in terms of a more rapid system growth and a shorter average startup delay. Copyright © 2016 John Wiley & Sons, Ltd.

Storming the Bastille: Indirect Tests of How the Vainqueurs Assembled*

How do people assemble in large groups? Do they come as individuals, do they move as an entire group from one location to another, or do they come in small groups of two or three? In this paper, we examine this issue in the context of the most prototypical of crowd events, the storming of the Bastille in 1789. We use data about the Vainqueurs (stormers) de la Bastille drawn from the French National Archives to indirectly test four possible mechanisms by which the Vainqueurs came to the Bastille: as individuals, through social networks, by proximity to gathering places, or through formal associations. Spatial analyses of their addresses, coupled with an eyewitness account of how people joined the Vainqueurs, suggest that social networks and proximity to gathering places best explain who stormed the Bastille, although there is evidence that some people came alone and some came through formal associations.

Crowd Turbulence with ABM and Verlet Integration on GPU Cards

Managing crowds is a key problem in a world with a growing population. Being able to predict and manage possible disasters directly affects the safety of crowd events. Current simulations focus mostly on navigation, but crowds have their own special characteristics and phenomena. Under specific conditions, a mass can turn into a crowd turbulence which may lead to a possible disaster. Understanding the internal phenomena is an important issue in order to model behavior. In the particular case of crowd turbulence, agents are moved by the crowd by a series of pushes, an involuntary movement that can be hard to reproduce. We propose a simple model to implement this complex problem based on intentional and involuntary interactions among the agents. The implementation is a hybrid model between the Verlet integration method and Agent Based Modeling. We implemented the proposed model using C and OpenCL and we evaluated its performance on a Nvidia GPU.

Empirical investigation on safety constraints of merging pedestrian crowd through macroscopic and microscopic analysis

A recent crowd stampede during a New Year's Eve celebration in Shanghai, China resulted in 36 fatalities and over 49 serious injuries. Many of such tragic crowd accidents around the world resulted from complex multi-direction crowd movement such as merging behavior. Although there are a few studies on merging crowd behavior, none of them have conducted a systematic analysis considering the impact of both merging angle and flow direction towards the safety of pedestrian crowd movement. In this study, a series of controlled laboratory experiments were conducted to examine the safety constraints of merging pedestrian crowd movements considering merging angle (60°, 90° and 180°) and flow direction under slow running and blocked vision condition. Then, macroscopic and microscopic properties of crowd dynamics are obtained and visualized through the analysis of pedestrian crowd trajectory data derived from video footage. It was found that merging angle had a significant influence on the fluctuations of pedestrian flows, which is important in a critical situation such as emergency evacuation. As the merging angle increased, mean velocity and mean flow at the measuring region in the exit corridors decreased, while mean density increased. A similar trend was observed for the number of weaving and overtaking conflicts, which resulted in the increase of mean headway. Further, flow direction had a significant impact on the outflow of the individuals while blocked vision had an influence on pedestrian crowd interactions and merging process. Finally, this paper discusses safety assessments on crowd merging behaviors along with some recommendations for future research. Findings from this study can assist in the development and validation of pedestrian crowd simulation models as well as organization and control of crowd events.

From Mindless Masses to Small Groups: Conceptualizing Collective Behavior in Crowd Modeling.

Computer simulations are increasingly used to monitor and predict behavior at large crowd events, such as mass gatherings, festivals and evacuations. We critically examine the crowd modeling literature and call for future simulations of crowd behavior to be based more closely on findings from current social psychological research. A systematic review was conducted on the crowd modeling literature (N = 140 articles) to identify the assumptions about crowd behavior that modelers use in their simulations. Articles were coded according to the way in which crowd structure was modeled. It was found that 2 broad types are used: mass approaches and small group approaches. However, neither the mass nor the small group approaches can accurately simulate the large collective behavior that has been found in extensive empirical research on crowd events. We argue that to model crowd behavior realistically, simulations must use methods which allow crowd members to identify with each other, as suggested by self-categorization theory.

An Examination of Evolving Crowd Management Strategies at Pilgrimage Sites: A Case Study of ‘Hajj’ in Saudi Arabia

The growing popularity of large crowded events has complicated their management, leading to the exploration of new crowd management strategies targeted at streamlining operations to address the risks from stampedes, infrastructure collapse, health, and terrorist threats. The emphasis has been to alleviate financial losses, keep event organizers safe from liability, and most importantly keep the attendees safe. Effective communication among and between officials and guests has been identified as a key element in this process. However, few scholars have investigated these strategies among heterogeneous crowds that congregate at religious sites. To overcome this gap, the authors use a case study approach focusing on initiatives adopted in managing approximately 3 million pilgrims from over 140 countries, congregating in Mecca, Saudi Arabia during the Hajj pilgrimage season. Findings suggest that the recently popularized use of Radio Frequency Identification (RFID) wrist bands for counting pilgrims at ports of entry and mosque gates alone is inadequate. A more networked approach that integrates the voices of Hajj pilgrims in risk communication is needed. Setting up Community Response Grids (CRGs) to use social media channels among pilgrim visitors need to be explored.

Crowd Event Detection on Optical Flow Manifolds.

Analyzing crowd events in a video is key to understanding the behavioral characteristics of people (humans). Detecting crowd events in videos is challenging because of articulated human movements and occlusions. The aim of this paper is to detect the events in a probabilistic framework for automatically interpreting the visual crowd behavior. In this paper, crowd event detection and classification in optical flow manifolds (OFMs) are addressed. A new algorithm to detect walking and running events has been proposed, which uses optical flow vector lengths in OFMs. Furthermore, a new algorithm to detect merging and splitting events has been proposed, which uses Riemannian connections in the optical flow bundle (OFB). The longest vector from the OFB provides a key feature for distinguishing walking and running events. Using a Riemannian connection, the optical flow vectors are parallel transported to localize the crowd groups. The geodesic lengths among the groups provide a criterion for merging and splitting events. Dispersion and evacuation events are jointly modeled from the walking/running and merging/splitting events. Our results show that the proposed approach delivers a comparable model to detect crowd events. Using the performance evaluation of tracking and surveillance 2009 dataset, the proposed method is shown to produce the best results in merging, splitting, and dispersion events, and comparable results in walking, running, and evacuation events when compared with other methods.

Cops and Chaos: A Historical Examination of the Police Role in Riot Control

Although professional police forces in the western world were developed owing in large part to the need for agencies capable of handling violent crowds and riots, the ability of police agencies to control riots remains questionable. The present work provides an overview of the nature of crowd violence, a brief description of the historical evolution of crowd violence in the United States, the role law enforcement agencies played in incidents of crowd violence, and broad theoretically informed policy recommendations for the handling of crowds and crowd violence. Building upon this review of the history of rioting in the United States and extant empirical research pertinent to crowds, group dynamics, and rioting, it is suggested that constant police professionalism and just treatment of the citizenry by police officers, thorough planning and preparation for crowd events, engagement of crowds, and rapid identification and removal of individual troublemakers may help reduce the potential for crowd violence.