三角剖分以及径向基函数神经网络在星图识别中的应用

张少迪; 王延杰; 孙宏海

doi:10.3788/OPE.20122002.0395

您当前的位置：

首页 >

文章列表页 >

三角剖分以及径向基函数神经网络在星图识别中的应用

信息科学 | 更新时间：2020-08-12

- 三角剖分以及径向基函数神经网络在星图识别中的应用
- Application of triangulation and RBF neural network to star pattern recognition
- 光学精密工程 2012年20卷第2期页码：395-402
- 作者机构：
  
  1. 中国科学院长春光学精密机械与物理研究所,吉林长春,中国,130033
  2. 中国科学院研究生院北京,100039
- 作者简介：
- 基金信息：
  
  国家863高技术研究发展计划资助项目(No.2006AA703405F)()
- DOI：10.3788/OPE.20122002.0395
  中图分类号： TP183;TP391.4
- 收稿日期：2011-04-18，
  
  修回日期：2011-06-21，
  
  网络出版日期：2012-02-25，
  
  纸质出版日期：2012-02-25
- 稿件说明：
移动端阅览
张少迪, 王延杰, 孙宏海. 三角剖分以及径向基函数神经网络在星图识别中的应用[J]. 光学精密工程, 2012,20(2): 395-402

ZHANG Shao-di, WANG Yan-jie, SUN Hong-hai. Application of triangulation and RBF neural network to star pattern recognition[J]. Editorial Office of Optics and Precision Engineering, 2012,20(2): 395-402
张少迪, 王延杰, 孙宏海. 三角剖分以及径向基函数神经网络在星图识别中的应用[J]. 光学精密工程, 2012,20(2): 395-402 DOI： 10.3788/OPE.20122002.0395.

ZHANG Shao-di, WANG Yan-jie, SUN Hong-hai. Application of triangulation and RBF neural network to star pattern recognition[J]. Editorial Office of Optics and Precision Engineering, 2012,20(2): 395-402 DOI： 10.3788/OPE.20122002.0395.

摘要

根据经典径向基函数(RBF)神经网络的优势

结合星图模式样本集的特点

设计了一种适合星图模式样本的网络训练算法。从提取星图模式入手

引入三角剖分理论

将可能出现在同一视场内的恒星以三角形的形式连接起来

提取连接的角距作为星图模式

建立了具有完备性、平移旋转不变性的星图模式样本集。然后

利用RBF神经网络做星图识别

研究顺序训练方法和批量训练方法

总结多种经典算法的优缺点

并设计了一种训练方法。通过实验证明了该种方法较其他经典算法更为适合学习星图模式样本。最后

给出RBF神经网络相关的训练数据

并通过模拟星图软件获得若干模拟星图作为观测样本

利用已经训练好的神经网络进行识别。试验结果表明

测试网络能够正确识别这些星图。

Abstract

A network training method for star pattern recognition was designed by combining a classific Radial Basic Function(RBF) neural network and star pattern samples. Firstly

the star pattern abstraction method was discussed and a triangulation based on star magnitudes was induced to connect the stars which probably appear in the same field of view.By taking extrated angular distances as the characteristic of star pattern

a star pattern sample set with completion

translation and rotation invariance was established. Then

RBF neural network was studied to recognize the star patterns. RBF network training method was classified as sequence learning and batch learning. Some typical algorithms that could represent the two methods were studied on their advantages and disadvantages

and a new training method was designed based on the specialty of above star pattern sample sets.Experiments indicate that the designed method is more appropriate than those typical algorithms. Several star images were simulated through software

which was regarded as the observatory data and entered into the trained RBF neural network to test. The experiment results show that the network can recognize all the star patterns successfully.

关键词

Keywords

references

钟红军. 流水线并行处理提高星敏感器数据更新率[J]. 光学精密工程,2009,17(9):2230-2235. ZHONG H J. Increasing update rate for star sensor by pipelining parallel processing method [J]. Opt. Precision Eng., 2009,17(9):2230-2235 . (in Chinese)[2] 张磊. 快速全天自主星图识别[J]. 光学精密工程,2009,17(4):909-915. ZHANG L. Fast all-sky autonomous star identification [J]. Opt. Precision Eng., 2009,17(4):909-915. (in Chinese)[3] ZHANG S D, WANG Y J, SUN H H. Design of high accuracy star image locating method used in star sensor technology . CMCE IEEE,2010:411-414.[4] 郑胜,吴伟仁,田金文,等. 一种基于三角形几何结构的星图识别算法[J]. 光学技术,2004,30(1):70-77. ZHENG SH, WU W R, TIAN J W,et al..A novel geometric structure-based autonomous star pattern identification algorithm[J]. Optical Technique, 2004,30(1):30-77.(in Chinese)[5] 刘朝山,黄欣,刘光斌. 凸多边形星图识别算法[J]. 光电工程,2004,31(9):7-9. LIU CH SH,HUANG X,LIU G B.Aconvex polygon star pattern identification algorithm for star sensor[J].Opto-Electronic Engineering,2004,31(9):7-9.(in Chinese)[6] 李葆华,张迎春. 用KMP算法进行星敏感器星图识别的方法[J]. 光电工程,2004,31(1):9-11. LI B H,ZHANG Y CH.A star map recognition method of star sensor with KMP algorithm[J]. Opto-Electronic Engineering,2004,31(1):9-11.(in Chinese)[7] YEE W W,KAH P S,LI-MINN A.Radial basis function neural network with incremental learning for face recognition[J]. IEEE Trans.,2011,2(2):302-309. [8] CHEN S. Local regularization assisted orthogonal least squares regression[J]. Neurocomputing, 2005,7: 559-585.[9] HUANG G B.A generalized growing and pruning RBF neural network for function approximation[J]. IEEE Trans.,2005,16(1):57-67.[10] HUANG G B. An efficient sequential learning algorithm for growing and pruning RBF(GAP-RBF)Networks [J]. IEEE, 2004, 34(6):2284-2292.[11] ZHANG R X. Improved GAP-RBF network for classification problems[J]. Science Direct,2006,70:3011-3018.[12] LIAO Y X.Sequential growing and pruning learning for recurrent neural networks using unscented or extended Kalman filter . Proceedings of the 27th Chinese Control Conference,2008:242-247.[13] BORTMAN M,ALADJEM M. A growing and pruning method for radial basis function networks[J].IEEE Trans.,2009,20(6):1039-1045.[14] LIANG N Y.A fast and accurate online sequential learning algorithm for feedforward networks[J].IEEE Trans.,2006,17(6):1411-1423.[15] 李斌,赖晓平. 改进的GGAP-RBF算法及其在函数逼近中的应用[J]. 模式识别与人工智能,2007,20(2):230-235. LI B, LAI X P. An improved GGAP-RBF algorithm and its application to function approximation[J]. PR&AI, 2007, 20(2):230-235. (in Chinese)[16] 周知. 三角剖分算法研究 . 哈尔滨:哈尔滨理工大学,2007. ZHOU ZH. Research on triangulation algorithm . Harbin: Harbin University of Science and Technology, 2007.(in Chinese)[17] 房建成,全伟,孟小红. 基于Delaunay三角剖分的全天自主星图识别算法[J]. 北京航空航天大学学报,2005,31(3):311-315. FANG J CH,QUAN W,MENG X H.All-sky autonomous star map identification algorithm based on Delaunay triangulation curring algorithm[J].Journal of Beijing University of Aeronautics and Astronautics, 2005,31(3):311-315.(in Chinese)

浏览量

409

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

多传感器的天文标定与空间配准

添加补偿码的快速径向伴星特征星图识别

大型散货堆体积的快速测量

三视场天文导航视场间星图识别的导航星选取

用于快速星跟踪的双向递推匹配识别