Title :
Performance of parallel algorithms for a fingerprint image comparison system
Author :
Ammar, Hany H. ; Miao, Zhouhui
Author_Institution :
Dept. of Electr. & Comput. Eng., West Virginia Univ., Morgantown, WV, USA
Abstract :
This paper addresses the problem of analyzing the performance of parallel algorithms for the training procedure of a neural network based fingerprint image comparison (FIC) system. The target architecture is assumed to be a coarse-grain distributed memory parallel architecture. Two types of parallelism: node parallelism and training set parallelism (TSP) are investigated. These algorithms are implemented on a 32 node CM-5. Theoretical analysis and experimental results comparing the performance of these algorithms are presented
Keywords :
distributed memory systems; fingerprint identification; image matching; learning (artificial intelligence); neural nets; parallel algorithms; parallel architectures; software performance evaluation; CM-5; coarse-grain distributed memory parallel architecture; experimental results; fingerprint image comparison system; neural network; node parallelism; parallel algorithm performance; theoretical analysis; training procedure; training set parallelism; Algorithm design and analysis; Convolution; Filters; Fingerprint recognition; Image coding; Image matching; Neural networks; Parallel algorithms; Parallel processing; Pixel;
Conference_Titel :
Parallel and Distributed Processing, 1996., Eighth IEEE Symposium on
Conference_Location :
New Orleans, LA
Print_ISBN :
0-8186-7683-3
DOI :
10.1109/SPDP.1996.570362
