Abstract

Image segmentation techniques typically require proper weighting of competing data fidelity and regularization terms. Conven- tionally, the associated parameters are set through tedious trial and error procedures and kept constant over the image. However, spatially vary- ing structural characteristics, such as ob ject curvature, combined with varying noise and imaging artifacts, significantly complicate the selec- tion process of segmentation parameters. In this work, we propose a novel approach for automating the parameter selection by employing a robust structural cue to prevent excessive regularization of trusted (i.e. low noise) high curvature image regions. Our approach autonomously adapts local regularization weights by combining local measures of im- age curvature and edge evidence that are gated by a signal reliability measure. We demonstrate the utility and favorable performance of our approach within two ma jor segmentation frameworks, graph cuts and active contours, and present quantitative and qualitative results on a variety of natural and medical images.