Human activity monitoring by local and global finite state machines

Abstract

There are a number of solutions to automate the monotonous task of looking at a monitor to find suspicious behaviors in video surveillance scenarios. Detecting strange objects and intruders, or tracking people and objects, is essential for surveillance and safety in crowded environments. The present work deals with the idea of jointly modeling simple and complex behaviors to report local and global human activities in natural scenes. Modeling human activities with state machines is still common in our days and is the approach offered in this paper. We incorporate knowledge about the problem domain into an expected structure of the activity model. Motion-based image features are linked explicitly to a symbolic notion of hierarchical activity through several layers of more abstract activity descriptions. Atomic actions are detected at a low level and fed to hand-crafted grammars to detect activity patterns of interest. Also, we work with shape and trajectory to indicate the events related to moving objects. In order to validate our proposal we have performed several tests with some CAVIAR test cases.

Ezagutzan Oinarritutako Giza Jardueren Eredu Dinamiko eta Pertsonalizatuak Ikasten

“…Giza jarduerak antzemateko gaitasunak eragin handia izan dezake hainbat aplikazio esparrutan, hala nola, ingurune adimentsuak [1], robotika soziala [2] edo bideo bidezko zaintza [3]. Ingurune adimentsuetan, giza jarduerak antzemateko gaitasunak ate asko ireki eta hainbat aukera berri ekar ditzake, hala nola pertsona baten portaeran aldaketak antzeman eta gaixotasunekin korrelazionatu, zailtasunak dituzten pertsonei laguntza eman jarduera zehatz batzuetan, bizimodu osasuntsu bat egiteko aholkuak eman eta beste hainbat aplikazio interesgarri.…”

Fusion of Static and Temporal Information for Threat Evaluation in Sensor Networks

“…In the literature, there are plenty of event modeling and reasoning systems, such as Finite State Machines [3], Bayesian Networks [2], and event composition with imperfect information [5,6], event modeling with decision support [8], etc. In general, these systems consider two branches to address the integration of heterogeneous information: most of them consider all information in a scenario as a whole, and with rules or fusion algorithms for each specific scenario.…”

“…In the literature, although there have been several event modeling and reasoning systems proposed [2,3,5,6,8], however, none of the models has properly addressed these two issues (detailed discussions will be in the related work section). In order to address these problems, in this paper, we first analyze general principles that should be obeyed by a fusion process that combines individual threat evaluations.…”

Selection of a Visible-Light vs. Thermal Infrared Sensor in Dynamic Environments Based on Confidence Measures

“…Visual monitoring, including people detection, tracking, recognition and activity interpretation [1], is a key component of intelligent video surveillance systems [2,3]. The contribution of a camera to the observation of a scene depends on its viewpoint and on the scene configuration.…”