High Sample rate architectures for median filters

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Citations (Scopus)

Abstract

This paper presents high sample rate array architectures and sorting network-based architectures for computing both recursive and non-recursive median niters. The proposed array architectures have a sampling rate that is higher than that of existing architectures. This is achieved by pipelining the computations in each processor. While the non-recursive niters are pipelined by placing latches in the feed-forward paths, the recursive niters are restructured to create additional delays in the feedback paths, and then pipelined using the delays as latches. The proposed sorting network-based architectures are highly pipelined, and consist of fewer compare-swap units than existing architectures. The reduction in the number of compare-swap units is achieved by processing multiple outputs at the same time, and also by using Batcher's odd-even merge sort. The latency of these networks is reduced by building them with sorting units which sort 2 elements (sort-S) as well as 3 elements (sort-3) in 1 time unit.

Original languageEnglish (US)
Title of host publicationWorkshop on VLSI Signal Processing 1992
EditorsWojtek Przytula, Kung Yao, Rajeev Jain, Jan Rabaey
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages490-499
Number of pages10
ISBN (Electronic)0780308115, 9780780308114
DOIs
StatePublished - Jan 1 1992
Event6th IEEE Workshop on VLSI Signal Processing - Los Angeles, United States
Duration: Oct 28 1992Oct 30 1992

Publication series

NameWorkshop on VLSI Signal Processing 1992

Conference

Conference6th IEEE Workshop on VLSI Signal Processing
CountryUnited States
CityLos Angeles
Period10/28/9210/30/92

Fingerprint

Median Filter
Median filters
Sorting
Sort
Sorting Networks
Unit
Swap
Sampling
Feedback
Pipelining
Path
Processing
Feedforward
Latency
Architecture
Odd
Computing
Output

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Applied Mathematics

Cite this

Chakrabarti, C. (1992). High Sample rate architectures for median filters. In W. Przytula, K. Yao, R. Jain, & J. Rabaey (Eds.), Workshop on VLSI Signal Processing 1992 (pp. 490-499). [641080] (Workshop on VLSI Signal Processing 1992). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/VLSISP.1992.641080

High Sample rate architectures for median filters. / Chakrabarti, Chaitali.

Workshop on VLSI Signal Processing 1992. ed. / Wojtek Przytula; Kung Yao; Rajeev Jain; Jan Rabaey. Institute of Electrical and Electronics Engineers Inc., 1992. p. 490-499 641080 (Workshop on VLSI Signal Processing 1992).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Chakrabarti, C 1992, High Sample rate architectures for median filters. in W Przytula, K Yao, R Jain & J Rabaey (eds), Workshop on VLSI Signal Processing 1992., 641080, Workshop on VLSI Signal Processing 1992, Institute of Electrical and Electronics Engineers Inc., pp. 490-499, 6th IEEE Workshop on VLSI Signal Processing, Los Angeles, United States, 10/28/92. https://doi.org/10.1109/VLSISP.1992.641080
Chakrabarti C. High Sample rate architectures for median filters. In Przytula W, Yao K, Jain R, Rabaey J, editors, Workshop on VLSI Signal Processing 1992. Institute of Electrical and Electronics Engineers Inc. 1992. p. 490-499. 641080. (Workshop on VLSI Signal Processing 1992). https://doi.org/10.1109/VLSISP.1992.641080
Chakrabarti, Chaitali. / High Sample rate architectures for median filters. Workshop on VLSI Signal Processing 1992. editor / Wojtek Przytula ; Kung Yao ; Rajeev Jain ; Jan Rabaey. Institute of Electrical and Electronics Engineers Inc., 1992. pp. 490-499 (Workshop on VLSI Signal Processing 1992).
@inproceedings{8eed9a16754c4b91a8e0059d8f369013,
title = "High Sample rate architectures for median filters",
abstract = "This paper presents high sample rate array architectures and sorting network-based architectures for computing both recursive and non-recursive median niters. The proposed array architectures have a sampling rate that is higher than that of existing architectures. This is achieved by pipelining the computations in each processor. While the non-recursive niters are pipelined by placing latches in the feed-forward paths, the recursive niters are restructured to create additional delays in the feedback paths, and then pipelined using the delays as latches. The proposed sorting network-based architectures are highly pipelined, and consist of fewer compare-swap units than existing architectures. The reduction in the number of compare-swap units is achieved by processing multiple outputs at the same time, and also by using Batcher's odd-even merge sort. The latency of these networks is reduced by building them with sorting units which sort 2 elements (sort-S) as well as 3 elements (sort-3) in 1 time unit.",
author = "Chaitali Chakrabarti",
year = "1992",
month = "1",
day = "1",
doi = "10.1109/VLSISP.1992.641080",
language = "English (US)",
series = "Workshop on VLSI Signal Processing 1992",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "490--499",
editor = "Wojtek Przytula and Kung Yao and Rajeev Jain and Jan Rabaey",
booktitle = "Workshop on VLSI Signal Processing 1992",

}

TY - GEN

T1 - High Sample rate architectures for median filters

AU - Chakrabarti, Chaitali

PY - 1992/1/1

Y1 - 1992/1/1

N2 - This paper presents high sample rate array architectures and sorting network-based architectures for computing both recursive and non-recursive median niters. The proposed array architectures have a sampling rate that is higher than that of existing architectures. This is achieved by pipelining the computations in each processor. While the non-recursive niters are pipelined by placing latches in the feed-forward paths, the recursive niters are restructured to create additional delays in the feedback paths, and then pipelined using the delays as latches. The proposed sorting network-based architectures are highly pipelined, and consist of fewer compare-swap units than existing architectures. The reduction in the number of compare-swap units is achieved by processing multiple outputs at the same time, and also by using Batcher's odd-even merge sort. The latency of these networks is reduced by building them with sorting units which sort 2 elements (sort-S) as well as 3 elements (sort-3) in 1 time unit.

AB - This paper presents high sample rate array architectures and sorting network-based architectures for computing both recursive and non-recursive median niters. The proposed array architectures have a sampling rate that is higher than that of existing architectures. This is achieved by pipelining the computations in each processor. While the non-recursive niters are pipelined by placing latches in the feed-forward paths, the recursive niters are restructured to create additional delays in the feedback paths, and then pipelined using the delays as latches. The proposed sorting network-based architectures are highly pipelined, and consist of fewer compare-swap units than existing architectures. The reduction in the number of compare-swap units is achieved by processing multiple outputs at the same time, and also by using Batcher's odd-even merge sort. The latency of these networks is reduced by building them with sorting units which sort 2 elements (sort-S) as well as 3 elements (sort-3) in 1 time unit.

UR - http://www.scopus.com/inward/record.url?scp=85068125485&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85068125485&partnerID=8YFLogxK

U2 - 10.1109/VLSISP.1992.641080

DO - 10.1109/VLSISP.1992.641080

M3 - Conference contribution

AN - SCOPUS:85068125485

T3 - Workshop on VLSI Signal Processing 1992

SP - 490

EP - 499

BT - Workshop on VLSI Signal Processing 1992

A2 - Przytula, Wojtek

A2 - Yao, Kung

A2 - Jain, Rajeev

A2 - Rabaey, Jan

PB - Institute of Electrical and Electronics Engineers Inc.

ER -