Pattern Recognition of Hand-Written Numerals by Round Field

スポンサーリンク

概要

• 論文の詳細を見る

• A method to extract geometric features of hand-written numerals by round field is proposed. In this method a circular region is introduced into the field of figure in which features such as an end-point, a bend-point, a cross-point and others are considered. By this method, the extraction of geometric features is performed without the normalization of the thickness and the size of characters. Furthermore, by varying the size of the round field, different feature points can be extracted from the same character for different parameters. Finally, experiments show that this method is very effective for structural recognition of characters.

• 一般社団法人情報処理学会の論文

著者

• Agui Takeshi

  Imaging Science And Engineering Laboratory Tokyo Institute Of Technology

• NAKAJIMA MASAYUKI

  Imaging Science and Engineering Laboratory, Tokyo Institute of Technology

• Nakajima Masayuki

  Imaging Science And Engineering Laboratory Tokyo Institute Of Technology

関連論文

• Generating Breadth-First Expression for Gray Scale Quadtree
• Structural Evolution of Neural Networks Having Arbitrary Connections by a Genetic Method
• A New Method for Smooth Interpolation without Twist Constraints
• Interpolation of CT Slices for Laser Stereolithography
• Orientable Closed Surface Construction from Volume Data
• Pattern Recognition of Hand-Written Numerals by Round Field

スポンサーリンク