Title :
A scalable parallel algorithm for the extraction of active contours
Author :
Wakatani, Akiyoshi
Author_Institution :
Fac. of Sci., Konan Univ., Kobe, Japan
Abstract :
We have developed a new parallel and scalable algorithm for active contour extractions. Our algorithm contains only communications between neighboring processors to achieve high performance parallel execution. Since the convergence ability of the contour is almost the same as the original (sequential) algorithm, the proposed method is a practical and effective algorithm to apply to any images
Keywords :
active vision; dynamic programming; feature extraction; parallel algorithms; performance evaluation; active contour extraction; convergence; dynamic programming; high performance parallel execution; image recognition; neighboring processor communication; scalable parallel algorithm; sequential algorithm; Active contours; Computational efficiency; Concurrent computing; Cost function; Data mining; Dynamic programming; Open loop systems; Optimal control; Parallel algorithms; Parallel processing;
Conference_Titel :
Parallel Computing in Electrical Engineering, 2000. PARELEC 2000. Proceedings. International Conference on
Conference_Location :
Trois-Rivieres, Que.
Print_ISBN :
0-7695-0759-X
DOI :
10.1109/PCEE.2000.873609
