Automatic Spatially-Adaptive Balancing of Energy Terms for Image Segmentation

TL;DR

This paper introduces a novel method for image segmentation that adaptively balances energy terms spatially, improving accuracy by incorporating image reliability into a graph-based framework, validated on synthetic and medical images.

Contribution

It presents a new technique for spatially-adaptive balancing of energy terms in image segmentation, addressing the limitations of fixed weights across the image domain.

Findings

Achieved 47% reduction in mean error on synthetic data

Validated improved segmentation accuracy on MRI and natural images

Demonstrated statistical significance of results (p-value << 0.05)

Abstract

Image segmentation techniques are predominately based on parameter-laden optimization. The objective function typically involves weights for balancing competing image fidelity and segmentation regularization cost terms. Setting these weights suitably has been a painstaking, empirical process. Even if such ideal weights are found for a novel image, most current approaches fix the weight across the whole image domain, ignoring the spatially-varying properties of object shape and image appearance. We propose a novel technique that autonomously balances these terms in a spatially-adaptive manner through the incorporation of image reliability in a graph-based segmentation framework. We validate on synthetic data achieving a reduction in mean error of 47% (p-value << 0.05) when compared to the best fixed parameter segmentation. We also present results on medical images (including segmentations of the corpus callosum and brain tissue in MRI data) and on natural images.