摘要
在非相参雷达目标识别方法中，基于ＦＦＴ－ＭＥＬＬＩＮ多重变换的特征提取方法，在消除由于雷达回波视角和时延所引起的目标特征不稳定性方面有较好的应用效果。针对飞机目标的具体特性，运用小波分析方法提出了一种新的多重变换特征提取方法，经实际数据验证，在飞行目标的架次识别中取得了良好的识别率和实时性。最后用该方法对三类水声目标的分类实验结果分析，亦取得了良好的分类效果。
in the field of noncoherent radar target recognition, the multi-transform for feature extraction with FFT-MELLIN transform has a good effect on eliminating the feature vector's unstability due to the change of target's azimuth and time delay. In view of the concrete feature of targets, this paper presents a new real-time multi--transform method, based on wavelet analysis, and gets a satisfactory real time characteristics and rate of target recognition, proved by practially sampled data. In the end, experimental results show that the new method can also exert a better influence on classification of different typyes of hydroacoustic target.
出处
《空军预警学院学报》
1999年第1期11-16,共6页
Journal of Air Force Early Warning Academy
关键词
目标识别
特征提取
多重变换
小波分析
Target recognition
Feature extraction
Multi- transform
Wavelet analysis
