交通运输系统工程与信息 ›› 2007, Vol. 7 ›› Issue (2): 32-338 .

• 智能交通系统与信息技术 • 上一篇    下一篇

基于SVM的数据层多源ITS数据融合方法初探

赵娜乐1,于雷1,2, 耿彦斌3,陈旭梅1   

  1. 1 北京交通大学 交通运输学院,北京100044;2 德克萨斯南方大学 美国休斯顿,77004;3交通部规划研究院交通仿真与决策支持研究中心,北京 100029
  • 收稿日期:2006-07-13 修回日期:1900-01-01 出版日期:2007-04-20 发布日期:2007-04-20

AN SUPPORT VECTOR MACHINE-BASED APPROACH TO DATA-LAYER MULTI-SOURCE ITS DATA FUSION

ZHAO Na-le1, YU Lei1, 2, GENG Yan-bin3 , CHEN Xu-mei1   

  1. 1 School of Traffic and Transportation, Beijing Jiaotong University, Beijing 100044;
    2 Texas Southern University, Houston, TX 77004, U.S.A;
    3 Decision Support Research Center, Transportation Planning and Research Institute, Ministry of Communications, Beijing 100044
  • Received:2006-07-13 Revised:1900-01-01 Online:2007-04-20 Published:2007-04-20

摘要: 在阐明ITS数据融合的意义及层次性的基础上,分析了数据层多源ITS数据融合及支持向量机的特点,根据支持向量机(SVM)的原理设计了利用支持向量机进行多源ITS数据融合的思路,并从支持向量机训练、训练结果评价以及支持向量机测试三个方面提出了该思路的实现步骤。在对日本阪神公路上堺入口的二源交通流数据进行支持向量机融合后,比较融合前后的数据,证明所提出的基于支持向量机技术的数据层多源ITS数据融合方法能够有效地进行数据质量控制,提高数据的精确度。

关键词: 支持向量机(SVM), 数据层, 多源ITS数据, 数据融合

Abstract: Through a characteristics analysis of multi-sources ITS data fusion on data layer and support vector machine, this paper proposes a multi-sources ITS data fusion approach using support vector machine on the basis of theory of support vector machine (SVM), and designs implementation processes of this approach from support vector machine training, training result evaluation, and support vector machine test. The comparison of data for before and after support vector machine fusion, when applying to two-source traffic flow data from BanShen highway ShangJie on-ramp in Japan, demonstrates that the proposed fusion approach can process the data quality control effectively, which improves the level of the data accuracy.

Key words: Support Vector Machine (SVM), Data Layer, Multi-source ITS Data, Data Fusion