基于线性内插神经网络的雷达目标一维距离像识别
1-D RANGE PROFILE IDENTIFICATION OF RADAR TARGETS BASED ON LINEAR INTERPOLATION NEURAL NETWORK
-
摘要: 本文提出一种新颖的神经网络模型线性内插神经网络(Linear InterpolationNeural Network,LINN)用于雷达目标一维距离像识别。它可避开提取不变特征的难点,利用目标一维距离像特征随姿态变化的信息来提高目标识别性能。实验结果表明,采用LINN很好地解决了在大的姿态角范围内识别目标时所存在的计算量与识别率的矛盾,提高了雷达对任意姿态目标的识别性能。Abstract: A novel neural network model---Linear Interpolation Neural Network(LINN) has been presented, which is used for radar target identification. And the 1-D range profiles of targets are used as identification feature. It is well known that the 1-D range profile reflects the precise geometric structure feature of a target, but it varies with the pose of the target. The LINN utilizes just the variation information of the 1-D range profile with the pose to improve the identification performance of targets in any posture.
-
Stewart C, Lu Yichuan, Larson V. Vector quantization and learning vector quantization for radar target classification. SPIE 1993, Vo1.1960, 115-124.[2]Lu Yichuan, Chang Kuochu. A Neural network approach for high resolution target classification. SPIE 1995. Vol.2484. 558-566.[3]Neiberg L, Casasent D P. Feature Space Trajectory (FST) classifier neural network. SPIE 1994, Vol.2353,276-292.[4]Kouba E T, Rogers S K, Ruck D W, Bauer Jr. K W. Recurrent neural networks for radar target identification. SPIE 1993. Vo1.1965. 256-265.[5][5J Jouny I, Garber F D, Ahalt S C. Classfication of radar targets using synthetic neural networks.[6]IEEE Trans. on AES, 1993, 29(2): 336-343.
计量
- 文章访问数: 2288
- HTML全文浏览量: 154
- PDF下载量: 441
- 被引次数: 0