\nobject oriented database is used to store all specific dataextracted by segmenting, tracking, and annotation of the\u00a0salient objects. The presented technique combines the visual\u00a0and semantic descriptors with spatial information for a good\u00a0indexing of data. The object oriented approach used it was\u00a0imposed by the complex structure of physical objects you\u00a0can find in the videos as well as the addition of semantic\u00a0meanings, which must relate to the objects detected. The\u00a0annotation based on domain ontologies and the pseudonatural\u00a0query language close to human understanding are a\u00a0proper contribution of this article, which has led to fulfilling\u00a0the final goal, the extraction of semantics-based syntax and\u00a0retrieving the frames with relevant information within a\u00a0pseudo-natural query language close to human understanding.\u00a0The results obtained for the sport domain validate the\u00a0research carried out and represents a model for applying\u00a0this method on videos from multiple domain of activity.<\/td>\n
![\"images\"](\"http:\/\/elth.ucv.ro\/fisiere\/anale\/wp-content\/uploads\/2015\/03\/images1.jpg\")
<\/a><\/td>\n<\/tr>\n<\/tbody>\n<\/table>\n","protected":false},"excerpt":{"rendered":"Abstract This paper presents a method for video information\u00a0retrieval considering the frames extracted from the\u00a0video sources through the segmentation of each key frame,\u00a0the tracking of the salient objects and the domain ontologies.\u00a0Based on disambiguation algorithms are constructed domain\u00a0ontologies for annotation of the video contents. An object oriented database is used to store all specific dataextracted … Continue reading Object Oriented Video Retrieval System based on Domain Ontologies<\/span>