Intermediate and low-level image processing with M3

M. Fikret Ercan; Yu Fai Fung; M. Suleyman Demokan

doi:10.1117/12.279566

Intermediate and low-level image processing with M3

M. Fikret Ercan, Yu Fai Fung, M. Suleyman Demokan

Research output: Journal article publication › Conference article › Academic research › peer-review

1 Citation (Scopus)

Abstract

We have developed a new architecture for computer vision which is a Multi-level, MIMD machine (M3). It is a homogenous system where the control unit, processing arrays and image store are all built using the same type of digital signal processor (DSP). DSPs provide satisfactory performance at low - and intermediate-level image processing and make the construction of a homogeneous system possible. The processing array of the system consists of three linear arrays. Each linear array is dedicated to one image processing type. The system has a modular design and it can be expanded easily. The overall system design utilizes distributed memory MIMD paradigm. The system is benchmarked with some low- and intermediate-level image processing algorithms such as convolution, edge finding, Hough transform etc. and results are presented.

Original language	English
Pages (from-to)	466-473
Number of pages	8
Journal	Proceedings of SPIE - The International Society for Optical Engineering
Volume	3164
DOIs	https://doi.org/10.1117/12.279566
Publication status	Published - 1997
Event	Applications of Digital Image Processing XX - San Diego, CA, United States Duration: 30 Jul 1997 → 1 Aug 1997

Keywords

Computer vision architectures
Image processing performance
Parallel processing

ASJC Scopus subject areas

Electronic, Optical and Magnetic Materials
Condensed Matter Physics
Computer Science Applications
Applied Mathematics
Electrical and Electronic Engineering

More information

10.1117/12.279566

Cite this

@article{45cf1353f2984f09a7d91c494af9dbe1,

title = "Intermediate and low-level image processing with M3",

abstract = "We have developed a new architecture for computer vision which is a Multi-level, MIMD machine (M3). It is a homogenous system where the control unit, processing arrays and image store are all built using the same type of digital signal processor (DSP). DSPs provide satisfactory performance at low - and intermediate-level image processing and make the construction of a homogeneous system possible. The processing array of the system consists of three linear arrays. Each linear array is dedicated to one image processing type. The system has a modular design and it can be expanded easily. The overall system design utilizes distributed memory MIMD paradigm. The system is benchmarked with some low- and intermediate-level image processing algorithms such as convolution, edge finding, Hough transform etc. and results are presented.",

keywords = "Computer vision architectures, Image processing performance, Parallel processing",

author = "\{Fikret Ercan\}, M. and Fung, \{Yu Fai\} and \{Suleyman Demokan\}, M.",

year = "1997",

doi = "10.1117/12.279566",

language = "English",

volume = "3164",

pages = "466--473",

journal = "Proceedings of SPIE - The International Society for Optical Engineering",

issn = "0277-786X",

publisher = "SPIE",

note = "Applications of Digital Image Processing XX ; Conference date: 30-07-1997 Through 01-08-1997",

}

TY - JOUR

T1 - Intermediate and low-level image processing with M3

AU - Fikret Ercan, M.

AU - Fung, Yu Fai

AU - Suleyman Demokan, M.

PY - 1997

Y1 - 1997

N2 - We have developed a new architecture for computer vision which is a Multi-level, MIMD machine (M3). It is a homogenous system where the control unit, processing arrays and image store are all built using the same type of digital signal processor (DSP). DSPs provide satisfactory performance at low - and intermediate-level image processing and make the construction of a homogeneous system possible. The processing array of the system consists of three linear arrays. Each linear array is dedicated to one image processing type. The system has a modular design and it can be expanded easily. The overall system design utilizes distributed memory MIMD paradigm. The system is benchmarked with some low- and intermediate-level image processing algorithms such as convolution, edge finding, Hough transform etc. and results are presented.

AB - We have developed a new architecture for computer vision which is a Multi-level, MIMD machine (M3). It is a homogenous system where the control unit, processing arrays and image store are all built using the same type of digital signal processor (DSP). DSPs provide satisfactory performance at low - and intermediate-level image processing and make the construction of a homogeneous system possible. The processing array of the system consists of three linear arrays. Each linear array is dedicated to one image processing type. The system has a modular design and it can be expanded easily. The overall system design utilizes distributed memory MIMD paradigm. The system is benchmarked with some low- and intermediate-level image processing algorithms such as convolution, edge finding, Hough transform etc. and results are presented.

KW - Computer vision architectures

KW - Image processing performance

KW - Parallel processing

UR - https://www.scopus.com/pages/publications/0031289539

U2 - 10.1117/12.279566

DO - 10.1117/12.279566

M3 - Conference article

AN - SCOPUS:0031289539

SN - 0277-786X

VL - 3164

SP - 466

EP - 473

JO - Proceedings of SPIE - The International Society for Optical Engineering

JF - Proceedings of SPIE - The International Society for Optical Engineering

T2 - Applications of Digital Image Processing XX

Y2 - 30 July 1997 through 1 August 1997

ER -

Intermediate and low-level image processing with M3

Abstract

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this