[1]朱斌,董剑,舒华忠,等.二值图像Legendre矩快速算法[J].东南大学学报(自然科学版),2003,33(1):90-93.[doi:10.3969/j.issn.1001-0505.2003.01.023]
 Zhu Bin,Dong Jian,Shu Huazhong,et al.Fast method for computing Legendre moments of binary images[J].Journal of Southeast University (Natural Science Edition),2003,33(1):90-93.[doi:10.3969/j.issn.1001-0505.2003.01.023]
点击复制

二值图像Legendre矩快速算法()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
33
期数:
2003年第1期
页码:
90-93
栏目:
计算机科学与工程
出版日期:
2003-01-20

文章信息/Info

Title:
Fast method for computing Legendre moments of binary images
作者:
朱斌 董剑 舒华忠 姜璐 罗立民
东南大学生物科学与医学工程系, 南京 210096
Author(s):
Zhu Bin Dong Jian Shu Huazhong Jiang Lu Luo Limin
Department of Biological Science and Medical Engineering, Southeast University, Nanjing 210096, China
关键词:
Legendre矩 离散格林公式 轮廓跟踪
Keywords:
Legendre moments discrete Green’s theorem contour tracking
分类号:
TP391.41
DOI:
10.3969/j.issn.1001-0505.2003.01.023
摘要:
提出一种有效的计算二值图像Legendre矩的方法.首先使用Yang-离散格林公式将二值图像矩计算中区域内求和转换为沿区域边界求和; 然后提取该图像的边界点; 再利用Shu提出的公式计算出边界点的Legendre多项式的叠加值.经过这3步后,二维Legendre矩计算转化为一维Legendre矩计算,从而有效地减少计算复杂度.介绍了用Hatamian滤波器计算一维Legendre矩的方法.最后给出实验结果证明方法的可行性.
Abstract:
An efficient algorithm for fast computing the Legendre moments of binary images is presented. First, Yang’s discrete Green’s theorem is used to transform the pixel-based calculation of Legendre moments into the contour-based calculation. Then, the boundary points are extracted. By using Shu’s theorems the sum of Legendre polynomial of the boundary points is calculated. By these three steps, the calculation of two-dimensional Lengendre moments is transformed into that of one-dimensional Legendre moments, and the complexity of calculation is efficiently reduced. The method of computing one dimensional Legendre moments by Hatamian filter is introduced. Experimental results show that this method is feasible.

参考文献/References:

[1] Hu M K.Visual pattern recognition by moment invariants[J]. IEEE Transaction on Information Theory,1962,8(1):179-187.
[2] Mukunman R,Ramakrishman K R.Fast computation of Lengendre and Zernike moments[J]. Pattern Recognition,2000,28(2):341-348.
[3] Liao S X,Pawlak M.On image analysis by moments[J].IEEE Trans Pattern Analysis and Machine Intelligence,1996,18(3):254-266.
[4] Zhou J D,Shu H Z.Two new algorithms for efficient computation of Legendre moment[J]. Pattern Recognition,2002,35:1143-1152.
[5] Yang L,Albregtsen F.Fast and exact computation of Cartesian geometric moments using discrete Green’s theorem[J].Pattern Recognition,1996,29:1061-1073.
[6] Shu Huazhong,Luo Limin,Bao Xudong,et al.An efficient method for computation of Legendre moment[J]. Graphical Models and Imaging Processing,2000,62:237-262.
[7] Hatamian M.A real-time two-dimensional moment generating algorithm and its single chip implementation[J].ASSP,1986,34:546-553.
[8] Li B C,Shen J.Fast computation of moment invariants[J]. Pattern Recognition,1991,24:1119-1128.
[9] Li B C.A new computation of geometric moments[J]. Pattern Recognition,1993,26:109-113.
[10] Shu H Z,Luo L M,Yu W X,et al.A new fast method for computing Lengendre moments[J]. Pattern Recognition,2000,33(2):341-348.
[11] Dai M,Baylou P,Nanjim M.An efficient algorithm for computation of shape moments from run-length codes or chain codes[J]. Pattern Recognition,1992,25:1119-1128.
[12] Zakaria M F,Vroomen L J,Zsombor-Murray P J A,et al.Fast algorithm for the computation of moment invariants[J]. Pattern Recognition,1987,20:639-643.

备注/Memo

备注/Memo:
作者简介: 朱斌(1977—),男,硕士生; 舒华忠(联系人),男,教授,博士生导师.
更新日期/Last Update: 2003-01-20