| 改进时空DS证据理论在车辆识别中的应用 |
| |
| 引用本文： | 张宗志,李一兵,林云. 改进时空DS证据理论在车辆识别中的应用[J]. 佳木斯工学院学报, 2010, 0(1): 54-57 |
| |
| 作者姓名： | 张宗志 李一兵 林云 |
| |
| 作者单位： | 哈尔滨工程大学信息与通信工程学院,黑龙江哈尔滨150001 |
| |
| 基金项目： | 黑龙江省青年学术骨干支持计划（1151G075）;哈尔滨市科技创新人才研究专项基金项目（2006RFQXG025）. |
| |
| 摘 要： | 将证据理论作为数据融合方法，将多传感器多测量周期的数据进行融合，讨论了集中式，分布式无反馈，分布式有反馈三种融合方法，将其应用于车辆类型识别方面，并通过实验对这三种方法进行了比较，得出在传感器提供数据较准确时，分布式有反馈融合方法的分辨效果最明显，而在传感器提供数据不是非常准确（甚至错误）时，分布式无反馈方法得到正确结果的准确率最高．
|
| 关 键 词： | DS证据理论 时空信息融合 车辆识别 |
The Application of Improving Space- time DS Evidence Theory in Distinguishing Vehicle |
| |
| ZHANG Zong - zhi,LI Yi - bing,LIN Yun. The Application of Improving Space- time DS Evidence Theory in Distinguishing Vehicle[J]. , 2010, 0(1): 54-57 |
| |
| Authors: | ZHANG Zong - zhi LI Yi - bing LIN Yun |
| |
| Affiliation: | (Information and Communication Engineering College, Harbin Engineering University, Harbin 150001, China) |
| |
| Abstract: | This paper takes evidence theory as data fusion method to fuse the data with multi- sensors and different measuring periods. It discusses three kinds of methods： concentrated fusion, distributed fusion without feedback and distributed fusion with feedback. In the application of vehicle type distinguishing, through comparing of the three methods by experiments, the paper got the conclusion that the distributed fusion with feedback performs best in the obviousness of fusion result. While in the aspect of fusion result accuracy rate, the distributed fusion without feedback is optimum. |
| |
| Keywords: | DS evidence theory space - time information fusion distinction of vehicle |
| 本文献已被 维普 等数据库收录！ |
