Ontology-based context representation and reasoning for object tracking and scene interpretation in video

Abstract

a b s t r a c tComputer vision research has been traditionally focused on the development of quantitative techniques to calculate the properties and relations of the entities appearing in a video sequence. Most object tracking methods are based on statistical methods, which often result inadequate to process complex scenarios. Recently, new techniques based on the exploitation of contextual information have been proposed to overcome the problems that these classical approaches do not solve. The present paper is a contribution in this direction: we propose a Computer Vision framework aimed at the construction of a symbolic model of the scene by integrating tracking data and contextual information. The scene model, represented with formal ontologies, supports the execution of reasoning procedures in order to: (i) obtain a high-level interpretation of the scenario; (ii) provide feedback to the low-level tracking procedure to improve its accuracy and performance. The paper describes the layered architecture of the framework and the structure of the knowledge model, which have been designed in compliance with the JDL model for Information Fusion. We also explain how deductive and abductive reasoning is performed within the model to accomplish scene interpretation and tracking improvement. To show the advantages of our approach, we develop an example of the use of the framework in a video-surveillance application.

Context-based scene recognition from visual data in smart homes: an Information Fusion approach

“…More details about the structure of the knowledge model described in this paper can be found in a previous research work [27]. In that paper, we introduced an ontology-based framework for cognitive surveillance.…”

Context-based multi-level information fusion for harbor surveillance

“…In addition, they are not particularly effective to represent perdurants; i.e., entities that change in time. This requires the creation of artificial representational patterns [38] or the use of non-standard extensions to the standard languages [39]. As introduced in the previous section, we propose a combined architecture that extends the typical deductive reason-ing with probabilistic abductive reasoning.…”

Topological properties in ontology-based applications

“…RACER has been chosen because it includes support for different kind of inference rules such as deductive, abductive, spatial, temporal, etc. [2]. In addition, RACER manages the spatial knowledge using the RCC theory as a substrate.…”

“…The model, based on the JDL data processing model for Information Fusion systems, is stepped in several levels ranging from low-level track data to high-level scene situations [2]. These levels are:…”

“…Nevertheless, most Description Logics -the formalisms underpinning ontologies-are not expressive enough to capture the semantics of RCC and, when possible, it comes at a very high computational cost. This paper proposes Dynamic RCC, an approach to incorporate qualitative topological relations based on the RCC theory in a previous framework for the contextual interpretation of data acquired from a visual sensor network [2]. The cornerstone of the framework is an ontological model designed according to the Joint Directors of Laboratories (JDL) fusion model [3] that represents sensor and context information in several levels -from low-level tracking data to high-level situation knowledge.…”