No abstract

Tragically, mass gatherings such as music festivals, sports events or pilgrimage quite often end in terrible crowd disasters with many victims. In the past, research focused on developing physical models that model human behavior in order to simulate pedestrian flows and to identify potentially hazardous locations. However, no automatic systems for detection of dangerous motion behavior in crowds exist. In this paper, we present an automatic system for the detection and early warning of dangerous situations during mass events. It is based on optical flow computations and detects patterns of crowd motion that are characteristic for hazardous congestions. By applying an online change-point detection algorithm, the system is capable of identifying changes in pedestrian flow and thus alarms security personnel to take necessary actions

Recognizing human actions is of vital interest in video surveillance or ambient assisted living. We consider an action as a sequence of body poses which are themselves a linear combination of body parts. In an offline procedure, nonnegative tensor factorization is used to extract basis images that represent body parts. The weighting coefficients are obtained by filtering a frame with the set of basis images. Since the basis images are obtained from nonnegative tensor factorization, they are separable and filtering can be implemented efficiently. The weighting coefficients encode dynamics and are used for action recognition. In the proposed action recognition framework, neither explicit detection and tracking of humans nor background subtraction are needed. Furthermore, for recognizing location specific actions, we implicitely take scene objects into account.

Congestions in pedestrian traffic typically occur when the number of pedestrians exceeds the capacity of pedestrian facilities. In some cases, the pedestrian density reaches a critical level which may lead to a crowd stampede as happens rather frequently at mass gatherings, in stadiums or at train stations. In the past, research has focused on improving simulations of crowd motion in order to identify potentially dangerous locations and to direct pedestrian streams. Recently, works towards the automatic real-time detection of critical mass behavior based on optical flow computations have been proposed. In this paper, we verify these approaches by analyzing mircoscopic pedestrian behavior in congestions and conducting experiments on synthetic as well as on real datasets

Video surveillance has become a hot research topic due to the recently increased importance of safety and security issues. Usually, security personnel has to monitor a surveillance area and often they have to do this for 24 hours a day. Thus, it would be desirable to develop intelligent surveillance systems that support this task automatically. The system described in this contribution is thought of such an automatic surveillance system that has been developed to detect several dangerous situations in subway stations. The workflow and the most important steps from foreground segmentation, shadow detection, tracking and classification to event detection are described, discussed and evaluated in detail. The developed surveillance system yields satisfying results, as dangerous situations that need to be recognized are detected in most cases.