• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.9
• /
• pp.2170-2190
• /
• 2012

There are nowadays strong demands for intelligent surveillance systems, which can infer or understand more complex behavior. The application of crowd density estimation methods could lead to a better understanding of crowd behavior, improved design of the built environment, and increased pedestrian safety. In this paper, we propose a new crowd density estimation method, which aims at estimating not only a moving crowd, but also a stationary crowd, using images captured from surveillance cameras situated in various public locations. The crowd density of the moving people is measured, based on the moving area during a specified time period. The moving area is defined as the area where the magnitude of the accumulated optical flow exceeds a predefined threshold. In contrast, the stationary crowd density is estimated from the coarseness of textures, under the assumption that each person can be regarded as a textural unit. A multilayer neural network is designed, to classify crowd density levels into 5 classes. Finally, the proposed method is experimented with PETS 2009 and the platform of Gangnam subway station image sequences.

• Journal of the Society of Disaster Information
• /
• v.7 no.1
• /
• pp.12-20
• /
• 2011

This study was performed for identifying the characteristics of mass gathering and human injury in mass gathering based on the literature survey and analysis of mass gathering and crowd packing. The size and density of crowd influenced the characteristics of mass gathering according to type of mass gathering. The variables and causes of human injuries of mass gathering have positive or negative influences based on the weather, attendance, duration, location, mobility, event type, crowd mood, alcohol, drug, crowd density and age. Based on the physical mechanism of crowd packing, the degree of crowd packing was influenced by crowd pressure, crowd density and lasting time. But the magnitude of pressure for pedestrian injury criteria remains for further research.

• Journal of the Korean Society of Safety
• /
• v.38 no.2
• /
• 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.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.12 no.2
• /
• pp.103-110
• /
• 2012

The population growth along with the urbanization, has caused more problems in many public areas, such as subway airport terminals, hospital, etc. Many surveillance systems have been installed in the public areas, but not all of those can be monitored in real-time, because the operators that observe the monitors are very small compared with the number of the monitors. For example, the observer can miss some crucial accidents or detect after considerable delays. Thus, intelligent surveillance system for preventing the accidents are needed, such as Intelligent Surveillance Systems. in this paper, we propose a new crowd density estimation method which aims at estimating moving crowd using images from surveillance cameras situated in outdoor locations. The moving crowd is estimated from the area where using optical flow. The edge information is also used as feature to measure the crowd density, so we improve the accuracy of estimation of crowd density. A multilayer neural network is designed to classify crowd density into 5 classes. Finally the proposed method is experimented with PETS 2009 images.

• International journal of advanced smart convergence
• /
• v.12 no.4
• /
• pp.147-153
• /
• 2023

In recent times, an absence of effective crowd management has led to numerous stampede incidents in crowded places. A crucial component for enhancing on-site crowd management effectiveness is the utilization of crowd counting technology. Current approaches to analyzing congested scenes have evolved beyond simple crowd counting, which outputs the number of people in the targeted image to a density map. This development aligns with the demands of real-life applications, as the same number of people can exhibit vastly different crowd distributions. Therefore, solely counting the number of crowds is no longer sufficient. CSRNet stands out as one representative method within this advanced category of approaches. In this paper, we propose a crowd counting network which is adaptive to the change in the density of people in the scene, addressing the performance degradation issue observed in the existing CSRNet(Congested Scene Recognition Network) when there are changes in density. To overcome the weakness of the CSRNet, we introduce a system that takes input from the image's information and adjusts the output of CSRNet based on the features extracted from the image. This aims to improve the algorithm's adaptability to changes in density, supplementing the shortcomings identified in the original CSRNet.

• Journal of Korea Game Society
• /
• v.11 no.6
• /
• pp.193-199
• /
• 2011

Producing high density crowd simulation is time-consuming task as increasing the number of individuals in the crowds. In this paper, we propose a new control technique that can create realistic high density crowd simulation by using Smoothed Particle Hydrodynamics (SPH) method from fluid simulation field. Equations in SPH method are modified for evacuation, distance maintenance, and group maintenance forces for individual behaviors in the crowds. Experimental results showed that the proposed system could enable natural high density crowd simulation efficiently.

• Journal of the Korea Society of Computer and Information
• /
• v.17 no.7
• /
• pp.45-52
• /
• 2012

In this paper, an crowd density in elevator estimation method based on multi-class Adaboost classifier is proposed. The SOM (Self-Organizing Map) based conventional methods have shown insufficient performance in practical scenarios and have weakness for low reproducibility. The proposed method estimates the crowd density using multi-class Adaboost classifier with texture features, namely, GLDM(Grey-Level Dependency Matrix) or GGDM(Grey-Gradient Dependency Matrix). In order to classify into multi-label, weak classifier which have better performance is generated by modifying a weight update equation of general Adaboost algorithm. The crowd density is classified into four categories depending on the number of persons in the crowd, which can be 0 person, 1-2 people, 3-4 people, and 5 or more people. The experimental results under indoor environment show the proposed method improves detection rate by about 20% compared to that of the conventional method.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.18 no.2
• /
• pp.15-19
• /
• 1990

This survey was carried to determine the crowd In the city parks on the basis of the crowding and using denSity. The using density was vary different by season and day, and the density was much higher in this experiment than in the case of the foreign countries. This survey shows a high correlation between the using density and crowd as the crowd level was more influenced by the increasing number of park - users in the case of low using density than the high using density. The possible using space per individual was around 10㎥ in the parks, which means a strong endurance of the surveying group to the massing space, in the saturated crowding value that the crowd was not significantly affected by the increasing number of users.

• International Journal of Computer Science & Network Security
• /
• v.24 no.4
• /
• 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.

• Journal of the Korea Society of Computer and Information
• /
• v.22 no.11
• /
• pp.49-55
• /
• 2017

Recently, for safety of people, there are proposed so many technologies which detect density of people at the specific place or space. The representative technology for crowd density estimation was using image analysis method from CCTV images. However, this method had a weakness which could not be used and which's accuracy was lower at the dark or smog space. Therefore, in this paper, to solve this problem, we proposed a user density estimation system at closed space using high frequency and smart device. The system send inaudible high frequencies to smart devices and it count the smart devices which detect the high frequencies on the space. We tested real-time user density with the proposed system and ten smart devices to evaluate performance. According to the testing results, we confirmed that the proposed system's accuracy was 95% and it was very useful. Thus, because the proposed system could estimate about user density at specific space exactly, it could be useful technology for safety of people and measurement of space use state at indoor space.