计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (31): 61-63.

• 学术探讨 • 上一篇    下一篇

基于神经网络生成图像语义的算法研究

伍小芹1,2,温小斌2,康耀红2,张宏科1   

  1. 1.北京交通大学 电子工程学院,北京 100044
    2.海南大学 信息科学技术学院,海口 570228
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-11-01 发布日期:2007-11-01
  • 通讯作者: 伍小芹

New algorithm to obtain image semantic by neural network

WU Xiao-qin1,2,WEN Xiao-bin2,KANG Yao-hong2,ZHANG Hong-ke1   


  1. 1.Electric Information Engineering College,Beijing Jiaotong University,Beijing 100044,China
    2.Information Science Technology College,Hainan University,Haikou 570228,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-11-01 Published:2007-11-01
  • Contact: WU Xiao-qin

摘要: 提出一种利用神经网络获取图像语义的算法。通过构建一个RBF神经网络,在图像的颜色、纹理、形状等低层视觉特征和高层语义特征之间建立映射关系。利用遗传算法训练RBF网络,获得RBF网络的隐节点个数、中心、宽度和连接权值等参数值,训练成功后的神经网络能够自动获取图像的语义。实验结果表明,该算法具有较好的基于语义的检索效果,体现了人对图像内容的理解,符合人的思维习惯。

Abstract: The paper proposes a new algorithm to obtain image semantic by neural network.By means of a designed RBF neural network,the paper establishes the mapping relationship between the low-level features such as color,texture and shape,and the high-level semantic.A new training method using genetic algorithm is presented,which can get all the parameters(such as quantity,centers,widths and connection weights of RBF neural network).The successfully-trained neural network can obtain image semantic automatically.Experimental results indicate that the proposed image retrieval algorithm is effective in characterizing image semantic.