Vision-based monitoring method using gray relational analysis

Hsu, Wei-Lieh; Hsiao, Ching-Chun; Chang, Yang-Lang; Chen, T.-L.

doi:10.1049/iet-cvi.2007.0045

Cited by 7 publications

(1 citation statement)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The background differencing method is first used to extract visual features and to establish a grid model to describe crowd distribution. Using the proposed model [6] , the measured area is divided into several unit areas, where each unit area represents a simple cell in the grid model. The status value of each cell is represented by the change of the active pixels in its corresponding unit area, which is described as follows: …”

Section: The Grid Modelmentioning

confidence: 99%

Abnormal crowd event detection based on outlier in time series

Hsu

Wang

Lin

2014

2014 International Conference on Machine Learning and Cybernetics

View full text Add to dashboard Cite

Crowd management research shows a lack of depth in the literature insofar as most major incidents can be prevented or minimized by a proper management strategy. Specifically, if abnormal crowd events can be detected early and the relevant governing agency can take appropriate actions towards mitigating the dangers, 978-1-4799-4215-2/14/$31.00 ©2014 IEEE innovational outliers are used to determine the cause of abnormal feature variation.

show abstract

Section: The Grid Modelmentioning

confidence: 99%

Abnormal crowd event detection based on outlier in time series

Hsu

Wang

Lin

2014

2014 International Conference on Machine Learning and Cybernetics

View full text Add to dashboard Cite

show abstract

Research on the Selection of Construction Supplier in the Construction Supply Chain Environment

Xie

2013

Proceedings of 2013 4th International Asia Conference on Industrial Engineering and Management Innovation (IEMI2013)

View full text Add to dashboard Cite

The detection of congestion in a crowd using discrete moments

Hsu

Tsaur

2013

2013 International Conference on Machine Learning and Cybernetics

View full text Add to dashboard Cite

The management and control of crowds is crucial to the maintenance of public safety. Since crowd congestion prevents the smooth flow of traffic, possibly creating crammed and potentially unsafe conditions, it is important to closely monitor crowd-congestion conditions, to provide timely data analysis and to evaluate the potential for the development of unsafe conditions. This study proposes a method that uses video-monitoring devices to closely monitor crowd conditions and creates a grid model to efficiently detect crowd congestion and to facilitate the analysis necessary for crowd management.It is expected that the higher the congestion level, the higher is the risk of the development of unsafe conditions, especially in the case of congestion near an exit. This study uses a discrete geometric moment to determine the distribution of the congestion. A linear estimation is used to assess the varying moment, to determine the variation of congestion levels and to evaluate the increase or decrease in congestion levels with time.Based on the variation of congestion levels and the movement of the congestion center, it is possible to predict the future direction of the congestion's movement and its potential impact. The data collected forms the basis for a process for appropriate crowd control or evacuation measures, in order to reduce the risk of crammed and unsafe conditions.

show abstract

Vision-based monitoring method using gray relational analysis

Cited by 7 publications

References 5 publications

Abnormal crowd event detection based on outlier in time series

Abnormal crowd event detection based on outlier in time series

Research on the Selection of Construction Supplier in the Construction Supply Chain Environment

The detection of congestion in a crowd using discrete moments

Contact Info

Product

Resources

About