Computer Engineering and Applications ›› 2008, Vol. 44 ›› Issue (33): 167-169.DOI: 10.3778/j.issn.1002-8331.2008.33.051

• 图形、图像、模式识别 • Previous Articles     Next Articles

Image analysis by Hahn moments of two variables

WU Hai-yong1,2   

  1. 1.School of Physics and Electronic Engineering,Nanjing Xiaozhuang College,Nanjing 210017,China
    2.School of Computer and Science & Engineering,Southeast University,Nanjing 210096,China
  • Received:2008-05-15 Revised:2008-07-23 Online:2008-11-21 Published:2008-11-21
  • Contact: WU Hai-yong

基于两变量Hahn矩的图像分析

吴海勇1,2   

  1. 1.南京晓庄学院 物理与电子工程学院,南京 210017
    2.东南大学 计算机科学与工程学院,南京 210096
  • 通讯作者: 吴海勇

Abstract: This paper proposes new image moments whose kernel functions are discrete orthogonal Hahn polynomials of two variables.Meanwhile,a simple method is deduced for computing normalized discrete orthogonal Hahn polynomials of two variables.Reconstruction experiments on a binary image,a gray-scale image,as well as a noised image show that,as compared with Hahn moments of one variable,Hahn moments of two variables have lower reconstruction error.Consequently,they are expected to have better performance in extracting features of an image.

Key words: discrete orthogonal Hahn moments of two variables, image reconstruction, Hahn moments

摘要: 提出了一种新的、以两变量离散正交Hahn多项式为核函数的图像矩,推导了正则化后,两变量离散正交Hahn多项式的简单的计算方法。对二值图像、灰度图像以及噪声图像的重建实验表明:相对于同系数的单变量的Hahn矩,两变量Hahn矩的重建误差更小。因此,它们能够更好地提取图像的特征。

关键词: 两变量Hahn矩, 图像重建, Hahn矩