A close-loop verification approach for pedestrian stability based on machine vision

In public places, it is significant to analyze the stability of the crowd which can support the crowd management and control, and protect the evacuees safely and effectively. The numerical analysis method of system stability based on Lyapunov theory suffers problems that it is difficult to avoid random errors in the initialization of pedestrian density and velocity, as well as cumulative errors due to time increasing, limiting its application. This study adopts a complementary model of theoretical numerical analysis and machine vision with a parallel convolutional neural network (CNN) model. It proposes an approach of stability analysis and closed-loop verification for crowd merging systems. Thereby, this research provides theoretical and methodological support for planning of the functional layout of crowd flow in public crowd-gathering places and the control measures for stable crowd flow.

Crowd Management in der Lehre

ZusammenfassungWer schon einmal dicht gedrängt vor der Konzertbühne stand kann sich die aussichtslose Lage, wenn die Stimmung kippt und Panik aufkommt, gut vorstellen. Damit eine öffentliche Veranstaltung reibungslos verläuft ist eine gründliche Planung, also ein qualitativ hochwertiges Crowd Management unabdingbar. Im Studiengang „BWL-, Messe-, Kongress- und Eventmanagement“ der dualen Hochschule Ravensburg Baden-Württemberg (DHBW) ist das Thema Crowd Management ein wichtiger Bestandteil. Für dieses Fachthema wurde im Rahmen des IBH-Labs Seamless Learning das Lernobjekt Cman_event erstellt, das im oben genannten Studiengang entwickelt und erprobt wurde. Die Entwicklung des didaktischen Konzepts basiert auf dem Design Based Research (DBR) Ansatz, wie es im IBH-Lab Seamless Learning entwickelt und definiert wurde. Im Sinne des grenzenlosen Lernens werden in diesem Lernobjekt die Übergänge in den Brüchen „Theorie – Praxis/Realität“ adressiert. Auf verschiedenen Ebenen und in verschiedenen Dimensionen wird neben der Vermittlung theoretischer Kenntnisse den Lernenden ermöglicht, bestimmte Situationen auf verschiedene Weisen zu betrachten, die einer besseren Vorstellung der realen Situation dienen. Dazu gehört das Modellieren in zwei und drei Raumdimensionen auf dem Tisch über Visualisierung durch computergestützte Simulationen bis hin zum Fühlen eines Gedränges am eigenen Leib.

Crowd Abnormality Detection Using Optical Flow and GLCM-Based Texture Features

Detection of abnormal crowd behavior is one of the important tasks in real-time video surveillance systems for public safety in public places such as subway, shopping malls, sport complexes and various other public gatherings. Due to high density crowded scenes, the detection of crowd behavior becomes a tedious task. Hence, crowd behavior analysis becomes a hot topic of research and requires an approach with higher rate of detection. In this work, the focus is on the crowd management and present an end-to-end model for crowd behavior analysis. A feature extraction-based model using contrast, entropy, homogeneity, and uniformity features to determine the threshold on normal and abnormal activity has been proposed in this paper. The crowd behavior analysis is measured in terms of receiver operating characteristic curve (ROC) & area under curve (AUC) for UMN dataset for the proposed model and compared with other crowd analysis methods in literature to prove its worthiness. YouTube video sequences also used for anomaly detection.

A Framework for Crowd Management during COVID-19 with Artificial Intelligence

COVID-19 requires crowded events to enforce restrictions, aimed to contain the spread of the virus. However, we have seen numerous events not observing these restrictions, thus becoming super spreader events. In order to contain the spread of a human to human communicable disease, a number of restrictions, including wearing face masks, maintaining social distancing, and adhering to regular cleaning and sanitization, are critical. These restrictions are absolutely essential for crowded events. Some crowded events can take place spontaneously, such as a political rally or a protest march or a funeral procession. Controlling spontaneous crowded events, like a protest march, political rally, celebration after a sporting event, or concert, can be quite difficult, especially during a crisis like the COVID-19 pandemic. In this article, we review some well-known crowded events that have taken place during the ongoing pandemic. Guided by our review, we provide a framework using machine learning to effectively organize crowded events during the ongoing and for future crises. We also provide details of metrics for the validation of some components in the proposed framework, and an extensive algorithm. Finally, we offer explanations of its various functions of the algorithm. The proposed framework can also be adapted in other crises.

Big Data Analytics Strategy Framework: A Case of Crowd Management During the Hajj Pilgrimage, Mecca, Saudi Arabia

The objective of the present study was an investigation of applications of big data analytics in Hajj and Umrah for pilgrims, who come to Saudi Arabia every year for tourism and observation of religious rites as per the sacred beliefs of Islam. It has now become a necessity to see more applications of big data analytics in these pilgrimages because of the growing number of people every year. Therefore, crowd control, crowd management and conflict management are essential for reduction of stress, troubles, fatalities, accidents, theft and possible deaths during Hajj and Umrah events. Developing a predictive data analytic model for Hajj and Umrah will improve the efficiency, gross domestic product (GDP), surveillance, revenue generation, opportunities and satisfaction for the pilgrimages. In this paper, review of big data tools was presented along with their use in the decision support system and how it can be used for surveillance and crowd management. A robust big data framework applicable for Hajj and Umrah events was also presented in this paper. This was meant to aid seamless adoption and implementation of big data applications across sectors and government parastatals involved in Hajj and Umrah. The presented framework was also included all the relevant use cases related to these pilgrimages.

Applying Crowd Risk Mitigation Technologies in Urban Sport Events

In recent years, the need for advanced precautions for mitigating the risks imposed by events, which involve high volumes of people in shared spaces, has multiplied. The occurrence of COVID-19 pandemic has further altered event practices, spaces, and event attendees’ mindsets in large-scale events. Proper crowd management not only seeks to prevent acts of violence and injury, but in today’s event environments; efforts should be consciously applied to reduce the spread of respiratory infections such as COVID-19. As the events industry continues to evolve and face new limitations, ways in which event organizers respond must evolve as well. Smartphone technologies are opening new ways for event organizers to communicate with and monitor attendees. This case study explores current crowd management strategies, analyzes the gaps in widely used models, and finally proposes event management technologies trending in the field.

Enabling and Emerging Sensing Technologies for Crowd Management in Public Transportation Systems: A Review

Management of crowd information in public transportation (PT) systems is crucial to foster sustainable mobility, by increasing the user’s comfort and satisfaction during normal operation, as well as to cope with emergency situations, such as pandemic crises, as recently experienced with COVID-19 limitations. This paper presents a taxonomy and review of sensing technologies based on Internet of Things (IoT) for real-time crowd analysis, which can be adopted in various segments of the PT system (buses/trams/trains, railway/subway stations, and bus stops). To discuss such technologies in a clear systematic perspective, we introduce a reference architecture for crowd management, which employs modern information and communication technologies (ICT) in order to: (i) monitor and predict crowding events; (ii) adapt in real-time PT system operations, by modifying service frequency, timetables, routes, and so on; (iii) inform in real-time the users of the crowding status of the PT system, by means of electronic displays installed inside vehicles or at bus stops/stations, and/or by mobile transport applications. It is envisioned that the innovative crowd management functionalities enabled by ICT/IoT sensing technologies can be incrementally implemented as an add-on to traditional intelligent transportation system (ITS) platforms, which are already in use by major PT companies operating in urban areas. Moreover, it is argued that, in this new framework, additional services can be delivered, such as, e.g., on-line ticketing, vehicle access control and reservation in severely crowded situations, and evolved crowd-aware route planning.

A Descriptive Analysis of the Impact of COVID-19 on Emergency Department Attendance and Visit Characteristics in Singapore

Background: The coronavirus disease 2019 (COVID-19) has impacted the utilisation of Emergency Department (ED) services worldwide. This study aims to describe the changes in attendance at a single ED and corresponding patient visit characteristics before and during the COVID-19 period. Methods: In a single-centre retrospective cohort study, we used descriptive statistics to compare ED attendance, patient demographics and visit characteristics during the COVID-19 period (1 January–28 June 2020) and its corresponding historical period in 2019 (2 January–30 June 2019). Results: The mean ED attendance decreased from 342 visits/day in the pre-COVID-19 period to 297 visits/day in the COVID-19 period. This was accompanied by a decline in presentations in nearly every ICD-10-CM diagnosis category except for respiratory-related diseases. Notably, we observed reductions in visits by critically ill patients and severe disease presentations during the COVID-19 period. We also noted a shift in the ED patient case-mix from ‘Non-fever’ cases to ‘Fever’ cases, likely giving rise to two distinct trough-to-peak visit patterns during the pre-Circuit Breaker and Circuit Breaker period. Conclusions: This descriptive study revealed distinct ED visit trends across different time periods. The COVID-19 pandemic caused a reduction in ED attendances amongst patients with low-acuity conditions and those with highest priority for emergency care. This raises concern about treatment-seeking delays and the possible impact on health outcomes. The downward trend in low-acuity presentations also presents learning opportunities for ED crowd management planning in a post-COVID-19 era.

Real-Time Monitoring of COVID-19 SOP in Public Gathering Using Deep Learning Technique

Crowd management has attracted serious attention under the prevailing pandemic conditions of COVID-19, emphasizing that sick persons do not become a source of virus transmission. World Health Organization (WHO) guidelines include maintaining a safe distance and wearing a mask in gatherings as part of standard operating procedures (SOP), considered thus far the most effective preventive measures to protect against COVID-19. Several methods and strategies have been used to construct various face detection and social distance detection models. In this paper, a deep learning model is presented to detect people without masks and those not keeping a safe distance to contain the virus. It also counts individuals who violate the SOP. The proposed model employs the Single Shot Multi-box Detector as a feature extractor, followed by Spatial Pyramid Pooling (SPP) to integrate the extracted features to improve the model's detecting capabilities. The MobilenetV2 architecture as a framework for the classifier makes the model highly light, fast, and computationally efficient, allowing it to be employed in embedded devices to do real-time mask and social distance detection, which is the sole objective of this research. This paper's technique yields an accuracy score of 99% and reduces the loss to 0.04%. Doi: 10.28991/esj-2021-SPER-14 Full Text: PDF