计算机工程与应用 ›› 2010, Vol. 46 ›› Issue (29): 158-160.DOI: 10.3778/j.issn.1002-8331.2010.29.045

• 图形、图像、模式识别 • 上一篇    下一篇

基于不变矩和BP神经网络的剪纸纹样识别

李国祥1,2,张显全1,秦芳远1   

  1. 1.广西师范大学 计算机科学系,广西 桂林 541004
    2.广西财经学院 计算机与信息管理系,南宁 530003
  • 收稿日期:2009-03-05 修回日期:2009-04-27 出版日期:2010-10-11 发布日期:2010-10-11
  • 通讯作者: 李国祥

Paper cut-out patterns recognition based on moment invariants and BP neural network

LI Guo-xiang1,2,ZHANG Xian-quan1,QIN Fang-yuan1   

  1. 1.Department of Computer Science,Guangxi Normal University,Guilin,Guangxi 541004,China
    2.Department of Computer and Information Management,Guangxi University of Finance and Economics,Nanning 530003,China
  • Received:2009-03-05 Revised:2009-04-27 Online:2010-10-11 Published:2010-10-11
  • Contact: LI Guo-xiang

摘要: 针对剪纸纹样艺术夸张变形的特点,将剪纸图像进行预处理,提取7个不变矩作为剪纸纹样的特征向量,采用LM算法优化BP神经网络,通过归一化后的不变矩对BP神经网络进行训练,应用训练后的神经网络作为分类器对剪纸纹样进行模式识别,实验证明该方法能够较好地识别有一定艺术变形的剪纸纹样。

Abstract: According to the characteristics of paper cut-out patterns exaggerative deformations,the seven moment invariants can be used as the eigenvector after the pretreatment of paper cut-out patterns image.The seven moment invariants normalized are input into BP neural network based on Levenberg-Marquardt algorithm and the trained network is used to be a classifier to realize the image pattern recognition.Experiments show that a relatively good recognition performance can be achieved to recognize paper cut-out patterns of exaggerative deformations.

中图分类号: