Dynamic video surveillance systems guided by domain ontologies

Abstract

Esta es la versión de autor de la comunicación de congreso publicada en: This is an author produced version of a paper published in: AbstractIn this paper we describe how the knowledge related to a specific domain and the available visual analysis tools can be used to create dynamic visual analysis systems for video surveillance. Firstly, the knowledge is described in terms of application domain (types of objects, events... that can appear in such domain) and system capabilities (algorithms, detection procedures...) by using an existing ontology. Secondly, the ontology is integrated into a framework to create the visual analysis systems for each domain by inspecting the relations between the entities defined in the domain and system knowledge. Additionally, when necessary, analysis tools could be added or removed on-line. Experiments/Application of the framework show that the proposed approach for creating dynamic visual analysis systems is suitable for analyzing different video surveillance domains without decreasing the overall performance in terms of computational time or detection accuracy.

A semantic-guided and self-configurable framework for video analysis

“…For composing the workow, we extend the approach proposed by [35] that selects and orders the Algorithm entities to perform a specic task. In particular, we propose to solve the problem of selecting Algorithm instances for performing a task when multiple choices are available by modeling this process as a constraint satisfaction problem (CSP) [1].…”

“…This paper presents the combination of the enhancements achieved starting from previous work in the design of video analysis frameworks [31], knowledge representa-tion [34] and dynamic workows composition [35]. Firstly, the basic framework structure [31] is extended by including the support to semantic-based analysis.…”

“…Thirdly, automatic workow composition and update are proposed for analyzing each domain based on these descriptions. For this composition, we extend the approach introduced in [35] in order to be able to select the most appropriate algorithm for performing a task when multiple choices are available by modeling this selection as a constraint satisfaction problem (CSP) [1]. Finally, we demonstrate the success of our approach for composing workows for the video surveillance domain.…”

Ontology-Based Analysis of CCTV Data

“…Using ontologies for understanding events promises to significantly improve automated video analysis [4]- [7] . Some research have proposed ontologies for specific domains of visual surveillance [8]- [10] . This paper presents an ontologybased analysis of CCTV data consisting of features extraction, parsing module for inter-level data transformation, higherlevel event representation, knowledge reasoning and queries.…”