Adaptive Regularization for Image Segmentation using Local….pdfVIP

Adaptive Regularization for Image Segmentation using Local Image Curvature Cues 1 1 2 Josna Rao , Rafeef Abugharbieh and Ghassan Hamarneh 1Biomedical Image Signal Computing Lab, University of British Columbia, Canada 2Medical Image Analysis Lab, Simon Fraser University, Canada {josnar,rafeef}@ece.ubc.ca, hamarneh@cs.sfu.ca Abstract. Image segmentation techniques typically require proper weight- ing of competing data fidelity and regularization terms. Conventionally, the associated parameters are set through tedious trial and error pro- cedures and kept constant over the image. However, spatially varying structural characteristics, such as object curvature, combined with vary- ing noise and imaging artifacts, significantly complicate the selection process of segmentation parameters. In this work, we propose a novel approach for automating the parameter selection by employing a ro- bust 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 major segmentation frameworks, graph cuts and active contours, and present quantitative and qualitative results on a variety of natural and medical images. 1 Introduction Regularization plays a crucial role in improving the robustness and applicability of image segmentation techniques. Through the