NURBS curve shape modification and fairness evaluation for computer aided aesthetic design

  • Authors:
  • Tetsuzo Kuragano;Akira Yamaguchi

  • Affiliations:
  • Graduate School of Information Science, Meisei University, Nagabuchi, Ome-City, Tokyo, Japan;Graduate School of Information Science, Meisei University, Nagabuchi, Ome-City, Tokyo, Japan

  • Venue:
  • ISCGAV'07 Proceedings of the 7th WSEAS International Conference on Signal Processing, Computational Geometry & Artificial Vision
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

A curve with a monotone radius of curvature distribution is considered as a fair curve in the area of Computer Aided Aesthetic Design (CAAD). But no official standards have been established. Therefore, a criterion for a fair curve is proposed. A quintic NURBS curve, the first derivative of a quintic NURBS curve, curvature vector, curvature, and radius of curvature are expressed. The concept of radius of curvature specification to modify the shape of a NURBS curve is illustrated. The difference between the NURBS curve radius of curvature and the specified radius of curvature is minimized by introducing the least-squares method to modify the shape of the NURBS curve. Algebraic functions such as linear, quadratic, cubic, quartic, quintic, and six degrees are applied to the radius of curvature distribution of the designed curve as the specified radius of curvature. The radius of curvature distributions given by these six algebraic functions are considered monotone, because the independent variable of these algebraic functions is monotone to the corresponding dependent variable of these functions. Similarity is evaluated using the radius of curvature distribution according to six algebraic functions as references and the radius of curvature distribution of the designed curves as matches by using correlation matching. Curve shape similarity evaluation is tried using an example. Considering that a curve with a monotone variation of radius of curvature distribution is fair, the similarity of the designed curve to a fair curve is evaluated. This measured similarity expresses fairness to the fair curve. Using this technique, the fairness of a curve is evaluated by using the similarity of the radius of curvature distribution.