Abstract:Summary Plant root system is plastic and dynamic,allowing plants to respond to their different environments in order to optimize acquisition of important soil resources.A number of root architecture parameters are known to be correlated with improved crop performance.Therefore,quantitative analysis of root architecture parameters,or the spatial configuration and distribution of root system in soil,will help to promote our understanding for root structure,function and their interactions with the soil in rhizosphere zone.The overall objective of the present work is to reconstruct 3D (three-dimensional) root architecture in situ non-destructively and to measure quantitatively geometric parameters of root system. <br> The maize roots were used as research object,and the cross-section images of maize roots were performed on a whole-body magnetic resonance(MR)imaging system with a static magnetic field of 3.0 Tesla.The MR method was found to be very effective for capturing root slice images dynamically without any contact or perturbation of plant root system or growth medium.Based on the root slice image data for maize root architectures obtained by this non-destructive approach,3D root architecture parameters were measured.Root slice images were generated using a fast spin echo sequence with effective echo time of 10.2 ms,repetition time of 1 800 ms,slice thickness of 2 mm,and imaging matrix of 5 12 × 5 12 pixels.Although MR imaging is closely related to the water content of material,volumetric water contents of rhizosphere in the range of 5% 20% have not significant impact on the quality of root slice images.The reconstruction process consisted of the following steps:image format conversion, filtering,roots target extraction,root data encapsulation,model reconstruction.Radiant DICOM viewer and Matlab r2009b were employed to preprocess the original slice images,including:1) gray level transformation of the original gray level images;2)contourlet transform filtering for noise reduction and image quality enhancement;3) segmentation through Otsu threshold segmentation algorithm.A procedure to obtain root architecture system of maize was developed by computer image graphics technology.The root model was reconstructed with improved volume rendering algorithm in the environment of Visualization Toolkit 5.4.In order to verify the reliability of the reconstructed model,the obtained root architecture models were converted to STL (standard template library) format,and then were transformed to a 3D solid object.Then they were imported into a computer aided-design software Pro/E to calculate their geometric parameters.Similarly,actual geometric parameters of the samples were measured manually by a vernier caliper and water displacement method. <br> By comparing the reconstruction model of root architecture with the physical object,it was found that the obtained models were well consistent with real samples,showing very good agreement in shape,volume and other morphological parameters,and the errors among them were less than 3%. <br> In sum, the presented methodology can avoid making great efforts in experimental measurements and consequently development of the root architecture models,and decrease the error generated from manual data extraction.This work is expected to be a useful contribution for modeling and simulation of root architecture system in situ non-destructively.Therefore,this approach provides a novel technique for the study of plant root growth and its adaptive changes to various environmental conditions.

Robust Skeletonization for Plant Root Structure Reconstruction from MRI

3D U-Net Segmentation Improves Root System Reconstruction from 3D MRI Images in Automated and Manual Virtual Reality Work Flows

3D U-Net for Segmentation of Plant Root MRI Images in Super-Resolution

A Shape-Based Method for Automatic and Rapid Segmentation of Roots in Soil from X-ray Computed Tomography Images: Rootine

3D skeletonization and its application in plant root image analysis

Three-dimensional visualization of maize roots based on magnetic resonance imaging

Method for modeling plant root morphology based on ball splines skeleton

Skeleton-Based Surface Reconstruction for Visualizing Plant Roots

VRoot: An XR-Based Application for Manual Root System Architecture Reconstruction

DynamicRoots: A Software Platform for the Reconstruction and Analysis of Growing Plant Roots

Resolving Form-Structure-Function Relationships in Plants with MRI for Biomimetic Transfer

3-D Reconstruction Framework for Tree Roots Based on Permittivity Inversion and Feature-Matching Interpolation

Joint reconstruction and segmentation in undersampled 3D knee MRI combining shape knowledge and deep learning

Robust Multi-scale Anatomical Landmark Detection in Incomplete 3D-CT Data

RhizoNet segments plant roots to assess biomass and growth for enabling self-driving labs

Ground-penetrating radar-based automatic reconstruction of three-dimensional coarse root system architecture

Seeking Common Ground While Reserving Differences: Multiple Anatomy Collaborative Framework for Undersampled MRI Reconstruction

Research on automatic 3D reconstruction of plant phenotype based on Multi-View images

Automatic Reconstruction of Tree Skeletal Structures from Point Clouds

Segmenting root systems in X-ray computed tomography images using level sets

Quantification of the three-dimensional root system architecture using an automated rotating imaging system