DETECTION OF WEAK PERSPECTIVE INVARIANT 2D SHAPES WITHOUT TANGENT INFORMATION

スポンサーリンク

概要

• 論文の詳細を見る

• It is a fundamental and important problem that detecting Weak Perspective invariant shapes from images. The method of detecting a shape using the geometric relation of the tangent information of its outlines is already reported. But, the algorithm using tangent information is theoretically impossible for detecting shapes constituted in straight lines, and is difficult to sue for incomplete outlines. In this paper, a new detec- tion method without the tangent information is proposed. This method extracts an unchanging axis from the image in orthographic projection included in Weak Perspective transform, and detects an affine parameter by measuring the proportionality rate of the length from the axis to the partial outline of the image. Therefore, this method can be applied not only to curved line but to straight line or to the one combined with the both, and can use for incomplete outlines. The effectiveness had been shown in several experimental results

• 社団法人電子情報通信学会の論文
• 2003-01-15

著者

• Bao Y

  Yokohama National University

• Bao Yue

  Yokohama National University

• NAKATA Takayuki

  Kanazawa University,

• Nakata T

  Kanazawa University

• Nakata Takayuki

  Kanazawa University

• AKASAKA Hidehiko

  Yokohama National University

関連論文

• DETECTION OF FREE-FORM OBJECT WITH THREE DIMENSIONAL VOTING
• DETECTION OF WEAK PERSPECTIVE INVARIANT 2D SHAPES WITHOUT TANGENT INFORMATION

スポンサーリンク