Robust Skeletonization for Plant Root Structure Reconstruction from MRI

TL;DR

This paper introduces a two-stage method for reconstructing plant root structures from MRI scans, addressing challenges like noise and low resolution to improve accuracy.

Contribution

The paper presents a novel two-stage approach combining segmentation and skeletonization for more robust plant root reconstruction from MRI data.

Findings

Successfully applied to 22 MRI scans

Outperforms previous methods in connectivity accuracy

Comparable to human expert reconstructions

Abstract

Structural reconstruction of plant roots from MRI is challenging, because of low resolution and low signal-to-noise ratio of the 3D measurements which may lead to disconnectivities and wrongly connected roots. We propose a two-stage approach for this task. The first stage is based on semantic root vs. soil segmentation and finds lowest-cost paths from any root voxel to the shoot. The second stage takes the largest fully connected component generated in the first stage and uses 3D skeletonization to extract a graph structure. We evaluate our method on 22 MRI scans and compare to human expert reconstructions.