基于小波分解及KCN的雷达目标特征提取

讨论了利用一维距离像进行雷达目标识别的特征提取。从缩减表征距离像的特征维数和表征目标所需特征数出发，提出了一种基于小波分解及KCN的目标特征提取方法。为比较不同特征提取方法及所提取特征的性能，引入了Fisher距离来表明各类的类内紧致性和类间分离性。实验结果表明能够有效地进行特征提取。

Radar target feature extraction using wavelet transform and KCN

Feature extraction of radar target recognition using high resolution range profiles is discussed in this paper. A feature extraction scheme based on wavelet transform and Kohonen clustering network(KCN) is proposed from the points of reducing feature dimension and feature numbers to representing a target. To compare the performance of different feature extraction schems, the Fisher distance measure is introduced to reveal the in-class compactness and between-class separateness. Experiment result show that the presented scheme can extract the target feature effectively.