An Efficient Multiprocessor Architecture for Image Processing in VLSI

Image processing generally requires to manipulate a large number of data of small granularity. In this paper a multiprocessor architecture which allows an efficient image processing in VLSI is proposed. The proposed network is as efficient as hypercube using O(n1/2log n) less chip area when n is the number of nodes in the network. The effectiveness of the proposed design is verified by studying some important image processing algorithms developed for hypercube network. The proposed architecture is also simple and modular, and thus it is relatively easier to develop efficient fault tolerance designs.

[1]  Yousef Saad,et al.  Multigrid Algorithms on the Hypercube Multiprocessor , 1986, IEEE Transactions on Computers.

[2]  John P. Hayes,et al.  An Array Layout Methodology for VLSI Circuits , 1986, IEEE Transactions on Computers.

[3]  Sartaj Sahni,et al.  Image Template Matching on MIMD Hypercube Multicomputers , 1990, J. Parallel Distributed Comput..

[4]  Victor P. Nelson.  Fault-tolerant computing: fundamental concepts , 1990, Computer.

[5]  J. H. Kim,et al.  An Efficient Reconfiguration Scheme for WSI of Cube-Connected Cycles with Bounded Channel Width , 1990 .

[6]  Raymond J. Offen.  VLSI image processing , 1986 .

[7]  W.R. Moore,et al.  A review of fault-tolerant techniques for the enhancement of integrated circuit yield , 1986, Proceedings of the IEEE.

[8]  W. Kent Fuchs,et al.  Reconfigurable Cube-Connected Cycles Architectures , 1990, J. Parallel Distributed Comput..

[9]  Sartaj Sahni,et al.  Hypercube Algorithms for Image Transformations , 1989, ICPP.

[10]  Franco P. Preparata,et al.  The cube-connected-cycles: A versatile network for parallel computation , 1979, 20th Annual Symposium on Foundations of Computer Science (sfcs 1979).

[11]  Israel Koren,et al.  Yield enhancement designs for WSI cube connected cycles , 1989, [1989] Proceedings International Conference on Wafer Scale Integration.