| 基于小波神经网的SAR图像识别算法 |
| |
| 引用本文： | 曹兰英,朱自谦,夏良正.基于小波神经网的SAR图像识别算法[J].测控技术,2005,24(7):14-16,23. |
| |
| 作者姓名： | 曹兰英 朱自谦 夏良正 |
| |
| 作者单位： | 中国雷华电子技术研究所,江苏,无锡,214063;东南大学,自动控制系,江苏,南京,210096;中国雷华电子技术研究所,江苏,无锡,214063;东南大学,自动控制系,江苏,南京,210096 |
| |
| 摘 要： | 针对SAR图像的自动目标识别问题,研究了基于小波分析和神经网络的识别算法.由非线性小波基作为网络中神经元的激励函数,隐层结点数由小波分解次数和处理目标类别数决定,输出层由目标的类别数决定,同时利用目标的方位角来限定被识别目标的范围.实验结果表明,该方法有效降低了训练和识别的难度,取得了优于BP网络的识别结果,具有广阔的应用前景.
|
| 关 键 词： | 合成孔径雷达 自动目标识别 小波神经网 方位角估计 |
| 文章编号： | 1000-8829(2005)07-0014-03 |
SAR Image Identification Based on Wavelet Neural Network |
| |
| CAO Lan-ying,ZHU Zi-qian,XIA Liang-zheng.SAR Image Identification Based on Wavelet Neural Network[J].Measurement & Control Technology,2005,24(7):14-16,23. |
| |
| Authors: | CAO Lan-ying ZHU Zi-qian XIA Liang-zheng |
| |
| Abstract: | A method based on wavelet analysis and neural network is presented,to the question of SAR image automatic target recognition.The wavelet function is used as the active function in the network.The number of hidden units is determined by the wavelet decomposition and the dimension of input signal.The output dimension is determined by the classes of the targets.The pose of the target is used for restrict the scope of the identified targets.Experiment results show that the property of the wavelet network is better than that of BP network in computational cost and identify rate. |
| |
| Keywords: | synthetic aperture radar(SAR) automatic target recognition(ATR) wavelet neural network(WNN) pose estimation |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
