You need JavaScript to view this

Large window median filtering on Clip7

Abstract

Median filtering has been found to be a useful operation to perform on images in order to reduce random noise while preserving edges of objects. However, in some cases, as the resolution of the image increases, so too does the required window size of the filter. For parallel array processors, this leads to problems when dealing with the large amount of data involved. That is to say that there tend to be problems over slow access of data from pixels over a large neighbourhood, lack of available storage of this data during the operation and long computational times for finding the median. An algorithm for finding the median, designed for use on byte wide architecture parallel array processors is presented together with its implementation on Clip7, a scanning array of such processors. 6 references.
Authors:
Publication Date:
Jul 01, 1983
Product Type:
Journal Article
Reference Number:
EDB-85-179591
Resource Relation:
Journal Name: Patt. Recog. Lett.; (Netherlands); Journal Volume: 5-6
Subject:
99 GENERAL AND MISCELLANEOUS//MATHEMATICS, COMPUTING, AND INFORMATION SCIENCE; ARRAY PROCESSORS; PARALLEL PROCESSING; IMAGES; USES; PROGRAMMING; 990200* - Mathematics & Computers
OSTI ID:
5135969
Research Organizations:
Univ. College London, England
Country of Origin:
Netherlands
Language:
English
Other Identifying Numbers:
Journal ID: CODEN: PRLED
Submitting Site:
HEDB
Size:
Pages: 341-346
Announcement Date:
Jul 01, 1985

Citation Formats

Mathews, K N. Large window median filtering on Clip7. Netherlands: N. p., 1983. Web. doi:10.1016/0167-8655(83)90073-9.
Mathews, K N. Large window median filtering on Clip7. Netherlands. https://doi.org/10.1016/0167-8655(83)90073-9
Mathews, K N. 1983. "Large window median filtering on Clip7." Netherlands. https://doi.org/10.1016/0167-8655(83)90073-9.
@misc{etde_5135969,
title = {Large window median filtering on Clip7}
author = {Mathews, K N}
abstractNote = {Median filtering has been found to be a useful operation to perform on images in order to reduce random noise while preserving edges of objects. However, in some cases, as the resolution of the image increases, so too does the required window size of the filter. For parallel array processors, this leads to problems when dealing with the large amount of data involved. That is to say that there tend to be problems over slow access of data from pixels over a large neighbourhood, lack of available storage of this data during the operation and long computational times for finding the median. An algorithm for finding the median, designed for use on byte wide architecture parallel array processors is presented together with its implementation on Clip7, a scanning array of such processors. 6 references.}
doi = {10.1016/0167-8655(83)90073-9}
journal = []
volume = {5-6}
journal type = {AC}
place = {Netherlands}
year = {1983}
month = {Jul}
}