基于径向基函数网络的雷达目标一维像识别技术研究

黄德双; 保铮

基于径向基函数网络的雷达目标一维像识别技术研究

黄德双,
保铮

计量
- 文章访问数: 2448
- HTML全文浏览量: 118
- PDF下载量: 430
- 被引次数: 0
出版历程
- 收稿日期: 1993-06-14
- 修回日期: 1991-02-14
- 刊出日期: 1995-01-19

THE STUDY ON RECOGNITION TECHNIQUE OF RADAR TARGET S ONE DIMENSIONAL IMAGE BASED ON RADIAL BASIS FUNCTION NETWORK

摘要

摘要: 本文研究了利用递推最小二乘(RLS)方法训练的径向基函数网络(RBFN)用于雷达目标一维像的识别与分类问题。证明了RBFN与Parzen窗函数概率密度估计的等价性,指出RBFN隐层单元传输函数可以推广到一般的Psrzen概率核函数或势函数形式。还就高斯、三角、双指数三种核函数讨论了径向基函数网络形状参数、递推最小二乘算法的遗忘因子对识别结果的影响以及与网络训练时间的关系。
- 模式识别; 核函数; 形状参数; 遗忘因子; 一维像; 径向基函数; 递推最小二乘法
Abstract: This paper studies the problem applying Radial Basis Function Network (RBFN) which is trained by the traditional Recursive Least Square Algorithm (RLSA) to the recognition of one dimensional image of radar targets. The equivalence between RBFN and the estimation of Parzen window probabilistic density is proved, it is pointed out that the I/O functions in RBFN hidden units can be extended to general Parzen window probabilistic kernel function or potential function, too. This paper discusses the effects of the shape parameter in RBFN and the forgotten factor , in RLSA on the results of the recognition of three kinds of kernel function such as Gaussian, Triangle, Double-exponential kernel functions, at the same time, and discusses also the relationship between and the training time in RBFN.

HTML全文

参考文献(1)

Rumelhaart D B, Hintonn G B, Williams R J. Nature, 1986, 32(3): 533-536.[2][2]Vogl T P. Biological C ybernics, 1988, 59(4): 257-263.[3]Dahl B D. Accelerated Learning the Generalized Delta Rule. IJCNN, Vol.Ⅱ,Washington DCt 1989, 523-530.[4][4 Hush D R. Salas J M. Improving the Learning Rate of Back-Propagation Wigradient Reuse Algorithm, lJCNN, Vol.Ⅰ, San Diego: 1988, 441-447.[5]Parker D B. Optimal algorithms for adaptive networks, Second order back propagation, second order direct propagation, and second order Hebbian learning, IJCNN, Vo1.Ⅱ, San Diego: Nov. 1987, 593-600.[6]Sadjadi M R A, Lion R J. IEEE Trans. on SP, 1992, SP-40(2): 446-450.[7]Chan L W, Fallside F. Computer Speech and Language, 1987, 21(2): 234-240.[8]Kollias S, Anastassiou D. Adaptive Training of Multilayer Neural Networks Using a Least Squares Estimation Technique. HCNN, Vol.1, Washington DC: 1989, 383-390.[9]Kollias S, Anastassiou D. IEEE Trans. on CAS, 1989, CAS-36(8): 1092-1101.[10][10][11]Scalero R, Tepedelenlioglu N. IEEE Trans. on SP, 1992, SP-40(1): 202-210.[12]黄德双.雷达目标一维象识别技术的研究:[博士论文].西安电子科技大学,1992.12.[13]西安电子科技大学电子工程研究所目标识别研究组.雷达目标特性测试,识别与分类研究报告,1992.12.[14]Renals S, Rohwer R. Phoneme Classification Experiments Using Radial Basis Functions, IJCNN, Vo1.1, Washington DC: 1989, 461-467.[15]Vrchovnik G, Carter C R, Haykin S. Radial Basis Function Classification of Impulse Radar Waveforms, IJCNN, Vo1.1, San Diego: 1990, 45-50.[16]Ruck D W, Rogers S K, Kahrisky, M, et al. IEEE Trans. on NN, 1990, NN-1(4): 296-298.[17]Parzen E. Ann[J].Math. Statist.1962, 33(6):1065-1076[18]Duin R P W. IEEE Trans. on Compt., 1976, C-25(8): 1175-1179.

施引文献

资源附件(0)

访问统计