DocumentCode :
3242057
Title :
A Fast Algorithm for Image Euclidean Distance
Author :
Sun, Bing ; Feng, Jufu
Author_Institution :
Dept. of Machine Intell., Peking Univ., Beijing
fYear :
2008
fDate :
22-24 Oct. 2008
Firstpage :
1
Lastpage :
5
Abstract :
Determining, or selecting a distance measure over the input feature space is a fundamental problem in pattern recognition. A notable metric, called the image euclidean distance (IMED) was proposed by Wang et al. [5], which is demonstrated consistent performance improvements in many real-world problems. In this paper, we present a fast implementation of IMED, which is referred as the convolution standardizing transform (CST). It can reduce the space complexity from O(n1 2n2 2 ) to O(1) , and the time complexity from O(n1 2n2 2 ) to O(n1n2), for n1 X n2 images. Both theoretical analysis and experimental results show the efficiency of our algorithm.
Keywords :
convolution; image processing; convolution standardizing transform; convolution template; fast algorithm; image Euclidean distance; Algorithm design and analysis; Clustering algorithms; Convolution; Euclidean distance; Laboratories; Machine intelligence; Machine learning; Machine learning algorithms; Pattern recognition; Sun;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Pattern Recognition, 2008. CCPR '08. Chinese Conference on
Conference_Location :
Beijing
Print_ISBN :
978-1-4244-2316-3
Type :
conf
DOI :
10.1109/CCPR.2008.32
Filename :
4662985
Link To Document :
بازگشت