计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (30): 44-46.

• 学术探讨 • 上一篇    下一篇

基于ART-2神经网络及不变矩特征的图像模式识别

郝 雷,石圣羽,宗晓萍,淮小利   

  1. 河北大学 电子信息工程学院,河北 保定 071002
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-10-21 发布日期:2007-10-21
  • 通讯作者: 郝 雷

Application of ART-2 neural network and invariant moment in image pattern recognition

HAO Lei,SHI Sheng-yu,ZONG Xiao-ping,HUAI Xiao-li   

  1. College of Electronic and Information Engineering,Hebei University,Baoding,Hebei 071002,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-10-21 Published:2007-10-21
  • Contact: HAO Lei

摘要: 将图像进行预处理并提取图像的特征,计算出图像的不变矩,利用ART-2神经网络完成了对图像的模式识别。通过实验证明ART-2神经网络具有较高的识别率,并很好地解决了神经网络在模式识别中面对识别对象出现新模式时,网络的可塑性与稳定性的矛盾。

Abstract: This paper mainly uses image pre-processing and feature extraction to calculate the invariant moment of image,and ultimately realizes the image pattern recognition based on ART-2 neural network.Experimental results show that ART-2 neural network has high recogintion rate.It also solves the contradiction between network’s plasticity and stability,when new recogintion model appears.