|Table of Contents|

Image Feature Extraction and Recognition Based on Algebraic Method(PDF)

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

Issue:
1998年01期
Page:
5-9
Research Field:
Publishing date:

Info

Title:
Image Feature Extraction and Recognition Based on Algebraic Method
Author(s):
Huang Xiuwu Guo Yuefei Yang Jingyu
School of Information, NUST, Nanjing 210094
Keywords:
human facial recognition feature ex t ract ion pat tern reco gnit ion algebraic feature
PACS:
TP391.4,O235
DOI:
-
Abstract:
T his paper impro ves the similarity of matr ices proposed by refer ence [ 3] , replaced hig h-or der mat rix by low -order matr ix , to reduce calculat ion. First w e pro ve that the similarity of orig inal image mat rices is kept . T hen the new method is used to solv e the recog nit ion o f human facial images. Ex perimental result s hav e show n that compar ed w ith reference[ 3] , the method is effect ive, and of fers hig h reco gnizability.

References:

1 M ov ger a S D, Datta L. T ow ar d a fundamental theo ry o f optimal feature selection: Part Ⅰ.IEEE Patter n Anal and Ma ch Intel, 1981, 6( 5) : 601~616
2  Hong Z Q. Alg ebr aic fea ture ex tr actio n of imag e fo r recog nition. Patt ern Reco gnit ion,1991, 24( 3) : 211~219
3 程永清, 庄永明, 杨靖宇. 基于矩阵相似度的图像特征抽取和识别. 计算机研究与发展,1992, 11: 42~48

Memo

Memo:
-
Last Update: 2013-03-29