目前, 虽然监控视频数据量迅速增加, 但是每年仍因交通违规行为造成大量的交通事故和人身伤亡. 海量监控视频的检索系统存在“找不到”和“难理解”的问题. 为了解决上述问题, 在交通领域中引入本体. 首先详细介绍视频语义内容模型, 并对模型中包含的对象、概念、事件、时间关系、空间关系等语义知识进行形式化描述; 然后以交通违规领域为例, 用软件protégé 4.2构建该领域的本体, 并用JCreator 编辑器导出交通违规领域本体的代码; 最后将该模型应用于南昌的有关项目中, 取得了良好的效果.
Although application of surveillance video in traffic monitoring grows tremendously, traffic violations still cause a large number of accidents and personal injuries every year. In mass surveillance video retrieval systems, events are difficult to find and understand. Introduction of ontology into surveillance video retrieval systems can improve effectiveness and efficiency. This paper presents a model of ontology construction. Formal descriptions of objects, concepts, temporal relation, spatial relation and events are given and included in the model. A method of constructing concept ontology is explored for traffic violations by identifying the concepts and concept hierarchy. This paper uses protégé 4.2 to conduct traffic violations ontology, and exports the ontology using JCreator that displays the Rdf/OWL code generated by the ontology.
[1] Luo X, Xu Z, Yu J, et al. Building association link network for semantic link on web resources [J]. IEEE Transactions on Automation Science and Engineering, 2011, 8(3): 482-494.
[2] Xu Z, Luo X, Wang L. Incremental building association link network [J]. Computer Systems Science and Engineering, 2011, 26(3): 153-162.
[3] Gollapalli M, Li X. A framework of ontology guided data linkage for evidence based knowledge extraction and information sharing [C]//Data Engineering Workshops (ICDEW), 2013 IEEE 29th International Conference on. 2013: 294-297.
[4] 夏骄雄, 徐俊, 吴耿锋. 基于本体核与直方图的聚类预处理方法[J]. 上海大学学报: 自然科学版, 2008, 14(1): 19-25.
[5] 刘志, 张兆杨. 语义对象分割技术综述[J]. 上海大学学报: 自然科学版, 2007, 13(4): 477-484.
[6] Francois A R J, Nevatia R, Hobbs J, et al. VERL: an ontology framework for representing and annotating video events [J]. MultiMedia, 2005, 12(4): 76-86.
[7] Lee M W, Hakeem A, Haering N, et al. SAVE: a framework for semantic annotation of visual events [C]//Computer Vision and Pattern Recognition Workshops, IEEE Computer Society Conference on. 2008: 1-8.
[8] Naphade M, Smith J R, Tesic J, et al. Large-scale concept ontology for multimedia [J]. MultiMedia, 2006, 13(3): 86-91.
[9] León Z, Sánchez L. An ontology for mobile video games [C]//Artificial Intelligence (MICAI), 2010 9th Mexican International Conference on. 2010: 154-159.
[10] Jeong J W, Hong H K, Lee D H. Ontology-based automatic video annotation technique in smart TV environment [J]. IEEE Transactions on Consumer Electronics, 2011, 57(4): 1830-1836.
[11] Chen M, Yang C. Private recommendation system based on user social preference model and online-video ontology in interactive digital TV [C]//Intelligent Human-Machine Systems and Cybernetics (IHMSC), 2012 4th International Conference on. 2012: 260-263.
[12] Saad S, De Beul D, Mahmoudi S, et al. An ontology for video human movement representation based on Benesh notation [C]//Multimedia Computing and Systems (ICMCS), 2012 International Conference on. 2012: 77-82.
[13] Ouyang J Q, Liu R R. Ontology reasoning scheme for constructing meaningful sports video summarization [J]. Image Processing, 2013, 7(4): 324-334.
