Non-convex non-local flows for saliency detection
This work addresses saliency detection for medical image analysis, specifically glioblastoma segmentation, but appears incremental as it builds on existing non-local and p-Laplacian frameworks.
The authors tackled the problem of automatic saliency detection in digital images, particularly for glioblastoma segmentation in MRI-Flair images, by proposing a variational model with non-convex non-local flows, achieving monotonically better results in standard metrics.
We propose and numerically solve a new variational model for automatic saliency detection in digital images. Using a non-local framework we consider a family of edge preserving functions combined with a new quadratic saliency detection term. Such term defines a constrained bilateral obstacle problem for image classification driven by p-Laplacian operators, including the so-called hyper-Laplacian case (0 < p < 1). The related non-convex non-local reactive flows are then considered and applied for glioblastoma segmentation in magnetic resonance fluid-attenuated inversion recovery (MRI-Flair) images. A fast convolutional kernel based approximated solution is computed. The numerical experiments show how the non-convexity related to the hyperLaplacian operators provides monotonically better results in terms of the standard metrics.