http://hdl.handle.net/10016/7457 Mon, 20 May 2013 17:51:18 GMT 2013-05-20T17:51:18Z http://hdl.handle.net/10016/9323 Title: Analysis of distributed fusion alternatives in coordinated vision agents Author(s): Castanedo, Federico; García, Jesús; Patricio Guisado, Miguel Ángel; Molina, José M. Abstract: In this paper, we detail some technical alternatives when building a coherent distributed visual sensor network by using the multi-agent paradigm. We argue that the multi-agent paradigm fits well within the visual sensor network architecture and in this paper we specially focus on the problem of distributed data fusion. Three different data fusion coordination schemes are proposed and experimental results of passive fusion are presented and discussed. The main contributions of this paper are twofold, first we propose the use of multi-agent paradigm as the visual sensor architecture and present a real system results. Secondly, the use of feedback information in the visual sensors, called active fusion, is proposed. The experimental results prove that the multi-agent paradigm fits well within the visual sensor network and provide a novel mechanism to develop a real visual sensor network system. Description: 6 pages, 10 figures.-- Contributed to: 11th International Conference on Information Fusion (FUSION'2008, Cologne, Germany, Jun 30-Jul 3, 2008). Mon, 31 Dec 2007 23:00:00 GMT http://hdl.handle.net/10016/9323 2007-12-31T23:00:00Z http://hdl.handle.net/10016/9322 Title: Ontological representation of context knowledge for visual data fusion Author(s): Gómez Romero, Juan; Patricio Guisado, Miguel Ángel; García, Jesús; Molina, José M. Abstract: Context knowledge is essential to achieve successful information fusion, especially at high JDL levels. Context can be used to interpret the perceived situation, which is required for accurate assessment. Both types of knowledge, contextual and perceptual, can be represented with formal languages such as ontologies, which support the creation of readable representations and reasoning with them. In this paper, we present an ontology-based model compliant with JDL to represent knowledge in cognitive visual data fusion systems. We depict the use of the model with an example on surveillance. We show that such a model promotes system extensibility and facilitates the incorporation of humans in the fusion loop. Description: 8 pages, 4 figures.-- Contributed to: 12th International Conference on Information Fusion, 2009 (FUSION '09, Seattle, Washington, US, Jul 6-9, 2009). Wed, 31 Dec 2008 23:00:00 GMT http://hdl.handle.net/10016/9322 2008-12-31T23:00:00Z http://hdl.handle.net/10016/9321 Title: Improving the Segmentation Stage of a Pedestrian Tracking Video-based System by means of Evolution Strategies Author(s): Pérez Concha, Óscar; Patricio Guisado, Miguel Ángel; García, Jesús; Molina, José M. Abstract: Pedestrian tracking video-based systems present particular problems such as the multi fragmentation or low level of compactness of the resultant blobs due to the human shape or movements. This paper shows how to improve the segmentation stage of a video surveillance system by adding morphological post-processing operations so that the subsequent blocks increase their performance. The adjustment of the parameters that regulate the new morphological processes is tuned by means of Evolution Strategies. Finally, the paper proposes a group of metrics to assess the global performance of the surveillance system. After the evaluation over a high number of video sequences, the results show that the shape of the tracks match up more accurately with the parts of interests. Thus, the improvement of segmentation stage facilitates the subsequent stages so that global performance of the surveillance system increases. Description: 12 pages, 7 figures.-- Contributed to: Eighth European Workshop on Evolutionary Computation in Image Analysis and Signal Processing (EvoIASP 2006, Budapest, Hungary, Apr 10-12, 2006). Sat, 31 Dec 2005 23:00:00 GMT http://hdl.handle.net/10016/9321 2005-12-31T23:00:00Z