| 基于小波网络的主被动毫米波数据特征层融合 |
| |
| 引用本文： | 沈坤.基于小波网络的主被动毫米波数据特征层融合[J].科学技术与工程,2011,11(11):2492-2496. |
| |
| 作者姓名： | 沈坤 |
| |
| 作者单位： | 西安机电信息技术研究所,西安,710065 |
| |
| 摘 要： | 针对毫米波主被动数据决策层融合预处理代价高、丢失目标信息的缺点,提出了一种基于小波网络的毫米波主被动数据特征层融合方法。该方法从主被动数据中提取特征值,将特征值作为小波神经网络的输入,在小波网络中实现主被动数据的特征层融合,对目标进行识别。实验和计算机仿真表明,对与毫米波主被动数据融合,基于小波神经网络的特征层融合的识别率比基于D-S证据理论的决策层融合的识别率高。
|
| 关 键 词： | 毫米波 主被动复合 小波神经网络 D-S证据理论 融合 |
| 收稿时间： | 1/15/2011 5:41:55 PM |
| 修稿时间： | 1/15/2011 5:41:55 PM |
Active and Passive Millimeter Wave Data Feature Level Fusion Based on WNN |
| |
| shenkun.Active and Passive Millimeter Wave Data Feature Level Fusion Based on WNN[J].Science Technology and Engineering,2011,11(11):2492-2496. |
| |
| Authors: | shenkun |
| |
| Institution: | SHEN Kun,HUANG Zheng,LU Zhao-gan(Xi`an Institute of Electromechanical Information Technology,Xi`an 710065,P.R.China) |
| |
| Abstract: | Based Active and passive millimeter-wave decision fusion have expensive preprocessing and miss target information.In order to solve this an active and passive millimeter-wave feature level fusion method is proposed based on WNN.Eiqenvalue is extracted from active and passive data and input to the WNN.Achieve the feature level fusion of active and passive data,identify the target in the WNN. Experiments and computer simulations show that,the recognition rate of feature level fusion based on WNN is higher than the recognition rate of decision level fusion based on D-S.. |
| |
| Keywords: | millimeter-wave compound active and passive wavelet neural network D-S evidence theory information fusion |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《科学技术与工程》浏览原始摘要信息 |
| 点击此处可从《科学技术与工程》下载免费的PDF全文 |
|
