期刊文献+

基于形态学变权神经网络的数据精炼 被引量:1

Data Abstraction Based on Morphological Adjusted-Weight Neural Network
下载PDF
导出
摘要 本文研究了模式样本的选取操作要领和典型样本数据的精炼方法 .为改善神经网络普遍存在着误差精度与收敛速率对初始权值颇为敏感甚至过于依赖的缺陷 ,采用感受野的Gabor函数模型来优化网络权值的初始量 .这不仅能为学习过程提供一个良好的开端 ,重要的是将赋予网络模型以具备可塑性优化的基础和适应环境变化的潜力 .算法运用主成份数学分析法对结构元分量的贡献率进行了统计度量与计算 ,以提取出对滤波运算最具影响力的主要分量 ,从而可显著降低数据处理量 ,使运行速度和学习效率更高 . The selective operation of pattern sample and abstracting method of typical sample data are discussed in this paper. In order to improve obvious shortcoming existing in neural networks in which error precision and convergent rate will be sensitive to initial weight values, even more depending on them, the Gabor function model of visual perception field is applied to optimize initial weight value of neural networks. In this way a good start in learning process can be provided. It is more important to obtain plastic superiority adaptive to complicated alterable environment for the neural network model, and to implement optimal computing principle in which operation load of structuring element weights in morphological filter can be distributed according to their contributive rate.
作者 景晓军 余农
出处 《电子学报》 EI CAS CSCD 北大核心 2005年第3期397-401,共5页 Acta Electronica Sinica
基金 信息产业部项目 (No .41 30 30 60 1 )
关键词 计算机视觉 图像处理 数学形态学 数据精炼 状态优化 Algorithms Feature extraction Image processing Mathematical models Mathematical morphology Neural networks Optimization Principal component analysis
  • 相关文献

参考文献9

  • 1齐翔林,汪云九,朱舜山.初级视觉信息的Gabor小波表达研究[J].自然科学进展(国家重点实验室通讯),1996,6(5):608-612. 被引量:4
  • 2Serra J.Image Analysis and Mathematical Morphology [M].London:Academic Press,1988.
  • 3Dougherty E R,et al.Digital image processing methods[J].Marcel Dekker,New York,1994.110-138.
  • 4Won Y G,et al.Morphological shared-weight networks with applications to automatic target recognition[J].IEEE Trans Neural Networks,1997,8(5):1195-1203.
  • 5Loce R P,Corp X,Dougherty.Optimal restoration using the morphological hit-or-miss transform[J].SPIE,1992,1769(1):94-105.
  • 6Davidson J L.Simulated annealing and morphology neural networks[J].SPIE,1992,1769(1):119-127.
  • 7Ehrhardt R.Morphological filter design with genetic algorithms[J].SPIE,1994,2300(1):2-12.
  • 8Kohonen T.Self-organization and associative memory[M].Springer-Vering,Third edition,1989.
  • 9汪云九,崔翯,齐翔林.BP学习网络中权值的感受野型初始化研究——Ⅰ.对收敛速度的影响[J].自然科学进展(国家重点实验室通讯),1996,6(3):346-350. 被引量:7

二级参考文献5

共引文献8

同被引文献23

  • 1G X Ritter, P Sussner. An introduction to morphological neural networks[ A ]. International Conference on Pattem Recognition [ C]. Vienna, 1EEE Xplom, 1996.709 - 717.
  • 2P Sussner. Morphological perceptron learning[ A]. IEEE Inter- national Symposium on Intelligent ConlrolProceed~ngs [ C ]. Gaithersburg, MD, 1EEE Xplore, 1998.477 - 482.
  • 3M Grana, B Raducanu. Some applications of morphological neural networks [ A ]. Preextings of the International Joint Conference on Neural Networks [ C ]. Washington, DC, IEEE Xplore, 2001.4: 2518 - 2523.
  • 4G X Ritter, G Urcid. Lattice algebra approach to single-neuron computation[ J]. IEEE Transactions on Neural Networks, 2003, 14(2) :282 - 295.
  • 5P Sussner, M E Valle. Gray-scale morphological associative memories[ J]. 1F.EE Transactions on Neural Networks, 2006, 17(3) :559 - 570.
  • 6Villaverde, M Grana, A d' Anjou. Morphological neural net- works and vision based mobile robot navigation[ A ]. Lecture Notes in Computer Science(including subseries Lecture Notes in Artificial InteUigence and Lecture Notes in Bioinformatics [ C ]. Berlin, Springer Berlin Heidelberg,2006. 878 - 887.
  • 7C W Hart. Range facial recognition with the aid of eigenface and morphological neural networks [ A ]. Lecture Notes in Computer Science (including subseries Lecture Notes in Artifi- cial Intelligence and Lecture Notes in Bioinformatics ) [ C ]. Berlin, Springer Berlin Heidelberg, 2008.217 - 224.
  • 8A M da Silva, P Sussner. A brief review and comparison of feedforward morphological neural networks with applications to classification[ A ]. Lecture Notes in Computer Science ( Includ- ing Subseries Lecture Notes in Artificial Intelligence and Lec- ture Notes in Bioinformatics) [ C ]. Berlin, Springer Berlin Hei- delberg. 2008.783 - 792.
  • 9P Sussner, E L Esmi. An introduction to morphological per- ceptrons with competitive learning[ A]. Proceedings of the In- ternational Joint Conference on Neural Networks[ C]. Atlanta, GA, IEEE Xplore, 2009. 3024 - 3031.
  • 10P Sussner, E L Esmi. Constructive morphological neural net- works: Some theoretical aspects and experimental results in classification[ A ]. Studies in Computational Intelligence [ C ]. Berlin, Springer Berlin Heidelberg, 2009. 258/2009:123 - 144.

引证文献1

二级引证文献3

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部