DocumentCode :
447397
Title :
A Weighted Voting Model of Associative Memory: Experimental Analysis
Author :
Mu, Xiaoyan ; Watta, Paul ; Hassoun, Mohamad H.
Author_Institution :
Dept. of Electr. & Comput. Eng., Rose-Hulman Inst. of Technol., Terre Haute, IN
Volume :
2
fYear :
2005
fDate :
12-12 Oct. 2005
Firstpage :
1252
Lastpage :
1257
Abstract :
In a related paper (X. Mu et al., 2004), a weighted voting RAM-based associative memory model was proposed, and a theoretical analysis of its performance on binary and random memory sets was given. In this paper, we give an experimental analysis of the weighted voting memory using both binary-random memory sets and more practical memory sets consisting of gray scale face images. The results show that the weighted voting memory offers higher performance over the voting memory on both types of memory sets
Keywords :
content-addressable storage; face recognition; pattern classification; associative memory; binary memory set; face recognition; gray scale face image; random memory set; weighted voting model; Associative memory; Control systems; Databases; Distortion measurement; Image processing; Pattern matching; Random access memory; Read-write memory; Testing; Voting; Face Recognition; associative memory; classification; voting;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Systems, Man and Cybernetics, 2005 IEEE International Conference on
Conference_Location :
Waikoloa, HI
Print_ISBN :
0-7803-9298-1
Type :
conf
DOI :
10.1109/ICSMC.2005.1571318
Filename :
1571318
Link To Document :
بازگشت