PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Nonlinear optimisation method for image segmentation and noise reduction using geometrical intrinsic properties
Sasan Mahmoodi and Bayan Sharif
Image and Vision Computing Volume 24, pp. 202-209, 2005.

Abstract

This paper considers the optimisation of a nonlinear functional for image segmentation and noise reduction. Equations optimising this functional are derived and employed to detect edges using geometrical intrinsic properties such as metric and Riemann curvature tensor of a smooth differentiable surface approximating the original image. Images are then smoothed using a Helmholtz type partial differential equation. The proposed approach is shown to be very efficient and robust in the presence of noise, and the reported results demonstrate better performance than the conventional derivative based edge detectors

EPrint Type:Article
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Machine Vision
Theory & Algorithms
ID Code:4243
Deposited By:Sasan Mahmoodi
Deposited On:20 December 2008