DocumentCode
3230666
Title
A parallel computing array algorithm for GLC matrix calculations
Author
Ng, Chong Teck
Author_Institution
American Res. Corp. of Virginia, Radford, VA, USA
fYear
1993
fDate
7-9 Mar 1993
Firstpage
468
Lastpage
471
Abstract
The gray-level co-occurrence (GLC) method, a texture analysis algorithm in which GLC matrices are computed on subregions of an image, is considered. The large number of calculations required to find the matrices for an image of any practical size precludes use of the GLC method in real-time systems. The author defines the GLC matrix and shows that the computation of all GLC matrices for an image has time complexity O (N 4) for an image of size N × N when conventional methods are used. A parallel computing array for use in calculating GLC matrices is presented, and an associated GLC matrix calculation is explained. The algorithm is shown to have time complexity O (N 2) for large N
Keywords
computational complexity; image segmentation; image texture; matrix algebra; parallel algorithms; real-time systems; GLC matrix calculation; parallel computing array algorithm; real-time systems; texture analysis; time complexity; Algorithm design and analysis; Image analysis; Image texture analysis; Parallel processing; Real time systems; Shape; Software measurement; Streaming media; Tiles; Time measurement;
fLanguage
English
Publisher
ieee
Conference_Titel
System Theory, 1993. Proceedings SSST '93., Twenty-Fifth Southeastern Symposium on
Conference_Location
Tuscaloosa, AL
ISSN
0094-2898
Print_ISBN
0-8186-3560-6
Type
conf
DOI
10.1109/SSST.1993.522824
Filename
522824
Link To Document