The Journal of the Institute of Image Electronics Engineers of Japan
Online ISSN : 1348-0316
Print ISSN : 0285-9831
ISSN-L : 0285-9831
Contributed Papers
Weak Perspective Projection Invariant Pattern Recognition without Tangent Information
Takayuki NAKATAYue BAOHidehiko AKASAKA
Author information
JOURNAL FREE ACCESS

2003 Volume 32 Issue 3 Pages 232-238

Details
Abstract
It is a fundamental and important problem to detect weak perspective projection invariant image shape from a picture. The method to detect image shapes using the tangent information of outlines had been reported. However, the algorithm using tangent information cannot be used for detecting straight line shapes. In this paper, a new detection method without the tangent information is proposed. This method extracts an axis unchanged in weak perspective projection transform from the image shape, and detects a weak perspective projection parameter by measuring the proportionality rate of the length from the axis to the outline of the image shape. Therefore, this method can be used not only for curved line shapes but also for straight line shapes. Several experimental results show the effectiveness of new method.
Content from these authors
© 2003 by the Institute of Image Electronics Engineers of Japan
Previous article Next article
feedback
Top