现代防御技术 ›› 2018, Vol. 46 ›› Issue (1): 120-124.doi: 10.3969/j.issn.1009-086x.2018.01.019

• 探测跟踪技术 • 上一篇    下一篇

D-S证据理论改进算法提高水下目标识别准确性

刘标1,2, 许腾1, 李光1,2   

  1. 1.海军指挥学院,江苏 南京 210000;
    2.海军士官学校,安徽 蚌埠 233012
  • 收稿日期:2017-05-28 修回日期:2017-07-03 出版日期:2018-02-28 发布日期:2020-11-25
  • 作者简介:刘标(1984-),男,安徽太和人。讲师,博士生,研究方向为作战效能评估。通信地址:210000 江苏省南京市玄武区半山园21号研3队 E-mail:feilonglb@126.com

Algorithm Improvement of D-S Evidence Theory in Submarine Target Recognition

LIU Biao1,2, XU Teng1, LI Guang1,2   

  1. 1. Naval Command College,Jiangsu Nanjing 210000,China;
    2. Naval Petty Officer Academy,Anhui Bengbu 233012,China
  • Received:2017-05-28 Revised:2017-07-03 Online:2018-02-28 Published:2020-11-25

摘要: D-S证据理论在目标识别中有着广泛的应用,是较好的数据融合方法之一,但在证据高度冲突时,会产生有悖常理的结果。针对此问题,引入传感器的可信度,提出了一种新的改进算法。该算法弥补了D-S证据理论所存在的不足,比其他改进算法融合效果好,此算法应用于水下目标识别,计算结果表明提高了水下目标识别的准确性和有效性。

关键词: 数据融合, 多传感器, 目标识别, 证据理论, 改进算法, 可信度

Abstract: D-S evidence theory is widely used in target recognition. It is an effective method of multi-sensor data fusion. However, it involves counter-intuitive behaviors when the high conflict information exists. In view of the problem, the reliability of sensor is introduced and the new algorithm is proposed. The algorithm makes up for the shortcomings of D-S evidence theory, and has better fusion effect than other improved algorithms. The algorithm is applied in submarine target recognition and the calculation result indicates that the algorithm is effective and correct.

Key words: data fusion, multi-sensor, target recognition, evidence theory, improvement algorithm, reliability

中图分类号: