A skeleton parallel algorithm for image edge detection

被引：0

作者：

He, XJ ^{[1
]}

Hintz, T ^{[1
]}

Szewcow, U ^{[1
]}

机构：

[1] Univ Technol Sydney, Sch Comp Sci, Broadway 2007, Australia

来源：

INTERNATIONAL CONFERENCE ON PARALLEL AND DISTRIBUTED PROCESSING TECHNIQUES AND APPLICATIONS, VOLS I-IV, PROCEEDINGS | 1998年

关键词：

multi-scale theory; spiral architecture; edge detection; algorithmic skeleton; parallel processing;

D O I：

暂无

中图分类号：

TP3 [计算技术、计算机技术];

学科分类号：

0812 ;

摘要：

Image edge detection in computer vision and image processing is a process which detects one kind of significant feature in an image that appears as large delta values in intensities. In this paper, a parallel algorithmic skeleton for edge detection is proposed based on a Replicated Shared Object Model. This is a generic program at the design level at which neither data types nor the code of specific data-dependent procedures need to be detailed. Multi-Scale Image Theory plays art important role in this research. The Spiral Architecture is fundamental to this approach. Our work provides an efficient algorithm for edge detection and is robust to noise.

引用

页码：1665 / 1672

页数：8

共 11 条

[1]

ARAQUE F, 1996, P INT C PAR DISTR PR, P1027

[2] EDGE FOCUSING [J].

BERGHOLM, F .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1987, 9 (06) :726-741

[3] A COMPUTATIONAL APPROACH TO EDGE-DETECTION [J].

CANNY, J .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1986, 8 (06) :679-698

[4]

HE X, 1996, P PANS WORKSH VIS IN, P90

[5]

He XJ, 1998, LECT NOTES COMPUT SC, V1388, P252

[6]

LEE T, 1996, P INT C PAR DISTR PR, P199

[7]

LINDEBERG T, 1994, SCAPE SPACE THEORY C

[8]

LIPPMAN SB, 1995, CPLUSPLUS PRIMER

[9]

Sheridan P., 1996, THESIS U TECHNOLOGY

[10]

SHERIDAN P, UNPUB J IMAGE VISION

← 1 2 →