| 基于神经网络融合的目标识别技术研究 |
| |
| 引用本文： | 杨彦飞,董慧颖. 基于神经网络融合的目标识别技术研究[J]. 沈阳理工大学学报, 2006, 25(5): 53-57 |
| |
| 作者姓名： | 杨彦飞 董慧颖 |
| |
| 作者单位： | 沈阳理工大学,信息科学与工程学院,辽宁,沈阳,110168 |
| |
| 摘 要： | 采用人工神经网络和数据融合技术设计了一种图像识别分类器.采用单层感知机网络、BP网络、径向基网络对汽车目标的特征数据进行识别,最后分别运用多数投票、平均Bayes、专家委员会三种融合算法把对各网络识别结果进行融合,得出最终判别结果.仿真结果证明了融合分类器用于图像目标识别/分类的有效性和可行性.
|
| 关 键 词： | 神经网络 数据融合 目标识别 |
| 文章编号： | 1003-1251(2006)05-0053-05 |
| 收稿时间： | 2005-10-20 |
| 修稿时间： | 2005-10-20 |
A Study on Target Recognition Based on Neural Network Data Fusion |
| |
| YANG Yan-fei,DONG Hui-ying. A Study on Target Recognition Based on Neural Network Data Fusion[J]. Transactions of Shenyang Ligong University, 2006, 25(5): 53-57 |
| |
| Authors: | YANG Yan-fei DONG Hui-ying |
| |
| Affiliation: | Shenyang Ligong University, Shenyang 110168, China |
| |
| Abstract: | In this paper,an image recognition classifier based on neural network and data fusion is designed.Single layer perceptron neural network,BP neural network and RBF neural network are used to recognize vehicle target by features vector.Finally,the recognition result is fused by using the voting method,average Bayes classifier and committee of expert to get the final result.Experiment reveals the validity and feasibility of the neural network data fusion for applying to image recognition/ classifying. |
| |
| Keywords: | neural network data fusion target recognition |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
