학술논문

Shape Orientation Pattern
Document Type
Periodical
Source
IEEE Signal Processing Letters IEEE Signal Process. Lett. Signal Processing Letters, IEEE. 16(8):711-714 Aug, 2009
Subject
Signal Processing and Analysis
Computing and Processing
Communication, Networking and Broadcast Technologies
Shape measurement
Computer vision
Image processing
Application software
Remote sensing
Robot sensing systems
Robot vision systems
Pattern recognition
Standards development
Symmetric matrices
Orientability
orientation
RST invariant descriptor
shape orientation pattern
Language
ISSN
1070-9908
1558-2361
Abstract
Due to the variety of shapes as well as the diversity of applications there does not exist any single method for computing the shape orientation of all shapes. In this paper, we introduce a new shape orientation measure, shape orientation pattern, to find the crisp and fuzzy shape orientation as well as shape orientability from any shape. We also illustrate that the new shape orientation measure can be used as a RST invariant shape descriptor.