Title | ||
---|---|---|
Recursive Gabor transform for images of very large size and design of massively parallel architecture. |
Abstract | ||
---|---|---|
The conventional Gahor Transforms. Completeand Generalized cannot be appliedfor image matricesreceived as different partitions. In practice, very largeimages ore received in the form of submatrixpartitions. Then. taking transform on differentpartitions (of the same image) at different times is ofprime importance. The situation can be easily handledif the transformation algorithm is made recursive. Inthis paper we present a Recursive Gabor Transform(RGT) amenable for pipelining partitions of differentimages interleaved. Apriori knowledge of image sizeis not required for the RGT unlike the conventionalGabor Transform. A dedicated massively parallelarchitecture based on Pacube VLSI arrays ispresented wjhich achieves a high degree of pipeliningwith respect to different image matrices. Thepipelining rate is proportional to the delay of aPacube macrocell in mode I operation. |
Year | DOI | Venue |
---|---|---|
1997 | 10.1109/HIPC.1997.634538 | HiPC |
Keywords | Field | DocType |
inthis paper,different partition,conventionalgabor transform,apriori knowledge,appliedfor image,large size,different time,different image matrix,pacube vlsi,recursive gabor transform,completeand generalized,parallel architecture,computer science,digital signal processing,matrix decomposition,very large scale integration,image processing,gabor transform,vlsi | Pipeline (computing),Matrix (mathematics),Computer science,Parallel computing,A priori and a posteriori,Image processing,Very-large-scale integration,Image resolution,Gabor transform,Recursion | Conference |
ISBN | Citations | PageRank |
0-8186-8067-9 | 0 | 0.34 |
References | Authors | |
2 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
N. Venkateswaran | 1 | 6 | 4.48 |
M. Venkatachalam | 2 | 56 | 4.13 |
S. Swaminathan | 3 | 42 | 6.67 |
C. Srinivasan | 4 | 2 | 1.62 |