Identification and Measurement of Individual Roots in Minirhizotron Images of Dense Root Systems

Gillert, Alexander; Peters, Bo; Lukas, Uwe Freiherr von; Kreyling, Jürgen

doi:10.1109/iccvw54120.2021.00153

Cited by 5 publications

(5 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A couple studies have shown improved phenotyping prediction through added HS information [33, 34]. Researchers may also use the data to analyze root growth, architecture, and turnover of dense root systems [35, 36]. Some images contain other potential objects of interest such as fungus, mold, and algae which may be studied at their various timesteps to determine possible interactive dynamics between root and rhizosphere.…”

Section: Applicationsmentioning

confidence: 99%

“…We provide images for studying root traits in both peanut and sweet corn roots, and we apply a subset of peanut images to the task of semantic segmentation for both types of image data. Our dataset contains more in-depth evaluation of plant roots across a broad range of soil conditions, that can be applied to studies on phenotyping prediction [33, 34], dense root systems [35, 36], interactive dynamics between root and rhizosphere [37], and drought resiliency [5, 6]. The data can also be applied to the tasks of data reconstruction and semantic segmentation.…”

Section: Introductionmentioning

confidence: 99%

“…Our dataset contains more in-depth evaluation of plant roots across a broad range of soil conditions that can be applied to studies on phenotyping prediction [33,34], dense root systems [35,36], interactive dynamics between root and rhizosphere [37], and drought resiliency [5,6]. Most recently, HyperPRI has been used for studying and comparing the crop plants' drought tolerance and recovery [38].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

HyperPRI: A Dataset of Hyperspectral Images for Underground Plant Root Study

Chang,

Chowdhry,

Song

et al. 2023

Preprint

View full text Add to dashboard Cite

Collecting and analyzing hyperspectral imagery (HSI) of plant roots over time can enhance our understanding of their function, responses to environmental factors, turnover, and relationship with the rhizosphere. Current belowground red-green-blue (RGB) root imaging studies infer such functions from physical properties like root length, volume, and surface area. HSI provides a more complete spectral perspective of plants by capturing a high-resolution spectral signature of plant parts, which have extended studies beyond physical properties to include physiological properties, chemical composition, and phytopathology. Understanding crop plants’ physical, physiological, and chemical properties enables researchers to determine high-yielding, drought-resilient genotypes that can withstand climate changes and sustain future population needs. However, most HSI plant studies use cameras positioned above ground, and thus, similar belowground advances are urgently needed. One reason for the sparsity of belowground HSI studies is that root features often have limited distinguishing reflectance intensities compared to surrounding soil, potentially rendering conventional image analysis methods ineffective. In the field of machine learning (ML), there are currently no publicly available datasets containing the heavy correlation, highly textured background, and thin features characteristic of belowground root systems. Here we present HyperPRI, a novel dataset containing RGB and HSI data for in situ, non-destructive, underground plant root analysis using ML tools. HyperPRI contains images of plant roots grown in rhizoboxes for two annual crop species – peanut (Arachis hypogaea) and sweet corn (Zea mays). Drought conditions are simulated once, and the boxes are imaged and weighed on select days across two months. Along with the images, we provide hand-labeled semantic masks and imaging environment metadata. Additionally, we present baselines for root segmentation on this dataset and draw comparisons between methods that focus on spatial, spectral, and spatial-spectral features to predict the pixel-wise labels. Results demonstrate that combining HyperPRI’s hyperspectral and spatial information improves semantic segmentation of target objects.

show abstract

Section: Applicationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

HyperPRI: A Dataset of Hyperspectral Images for Underground Plant Root Study

Chang,

Chowdhry,

Song

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Recently, convolutional neural networks (CNNs) show promising results to identify roots in a variety of settings ( Rahmanzadeh and Shojaedini, 2016 ; Vincent et al , 2017 ; Huo and Cheng, 2019 ; Alonso-Crespo et al , 2022 ; Bauer et al , 2022 ). In particular, field soil minirhizotron imagery has been analysed several times ( Wang et al , 2019 ; Gillert et al , 2021 ; Han et al , 2021 ; Bauer et al , 2022 ; Peters et al , 2022 , Preprint; Smith et al , 2022 ), but transferability between sites and out of agricultural soils is difficult to assess without widespread adoption in new settings. These have also never been applied to high frequency studies where variability between images (e.g.…”

Section: Introductionmentioning

confidence: 99%

High frequency root dynamics: sampling and interpretation using replicated robotic minirhizotrons

Nair

Strube

Hertel

et al. 2022

Journal of Experimental Botany

View full text Add to dashboard Cite

Automating dynamic fine root data collection in the field is a longstanding challenge with multiple applications for co-interpretation and synthesis for ecosystem understanding. High frequency root data are only achievable with paired automated sampling and processing. However, automatic minirhizotron (root camera) instruments are still rare and data is often not collected in natural soils nor analysed at high temporal resolution. Instruments must also be affordable for replication and robust under variable natural conditions. Here, we show a system built with off-the-shelf parts which samples at sub-daily resolution. We paired this with a neural network to analyse all images collected. We performed two mesocosm studies and two field trials alongside ancillary data collection (soil CO2 efflux, temperature and moisture content, and ‘PhenoCam’-derived above-ground dynamics). We produce robust and replicated daily time-series of root dynamics under all conditions. Temporal root changes were a stronger driver than absolute biomass on soil CO2 efflux in mesocosm. Proximal sensed aboveground dynamics and belowground dynamics from minirhizotron data were not synchronised. Root properties extracted were sensitive to soil moisture and occasionally to time of day (potentially relating to soil moisture). This may only affect high frequency imagery and should be considered in interpreting such data.

show abstract

“…Recently, convolutional neural networks (CNNs) showed promising results to identify roots in a variety of settings (Delory et al ., 2016; Rahmanzadeh and Shojaedini, 2016; Vincent et al ., 2016; Huo and Cheng, 2019; Bauer et al ., 2022). In particular, field soil minirhizotron imagery has been analysed several times with good results (Wang et al ., 2019; Smith et al ., 2020; Gillert et al ., 2021; Han et al ., 2021; Peters et al ., 2022; Bauer et al ., 2022), but transferability between sites and out of agricultural soils is difficult to assess without widespread adoption in new settings. These have also never been applied to high frequency studies where variability between images (e.g.…”

Section: Introductionmentioning

confidence: 99%

High Frequency Root Dynamics: Sampling and Interpretation Using Replicated Robotic Minirhizotrons

Nair

Strube

Hertel

et al. 2022

Preprint

View full text Add to dashboard Cite

Minirhizotrons (paired camera systems and buried observatories) are the best current method to make repeatable measurements of fine roots in the field. Automating the technique is also the only way to gather high resolution data necessary for comparison with phenology-relevant above-ground remote sensing, and, when appropriately validated, to assess with high temporal resolution belowground biomass, which can support carbon budgets estimates. Minirhizotron technology has been available for half a century but there are many challenges to automating the technique for global change experiments. Instruments must be cheap enough to replicate on field scales given their shallow field of view, and automated analysis must both be robust to changeable soil and root conditions because ultimately, image properties extracted from minirhizotrons must have biological meaning. Both digital photography and computer technology are rapidly evolving, with huge potential for generating belowground data from images using modern technological advantages. Here we demonstrate a homemade automatic minirhizotron scheme, built with off-the-shelf parts and sampling every two hours, which we paired with a neural network-based image analysis method in a proof-of-concept mesocosm study. We show that we are able to produce a robust daily timeseries of root cover dynamics. The method is applied at the same model across multiple instruments demonstrating good reproducibility of the measurements and a good pairing with an above-ground vegetation index and root biomass recovery through time. We found a sensitivity of the root cover we extracted to soil moisture conditions and time of day (potentially relating to soil moisture), which may only be an issue with high resolution automated imagery and not commonly reported as encountered when training neural networks on traditional, time-distinct minirhizotron studies. We discuss potential avenues for dealing with such issues in future field applications of such devices. If such issues are dealt with to a satisfactory manner in the field, automated timeseries of root biomass and traits from replicated instruments could add a new dimension to phenology understanding at ecosystem level by understanding the dynamics of root properties and traits.

show abstract

Identification and Measurement of Individual Roots in Minirhizotron Images of Dense Root Systems

Cited by 5 publications

References 16 publications

HyperPRI: A Dataset of Hyperspectral Images for Underground Plant Root Study

HyperPRI: A Dataset of Hyperspectral Images for Underground Plant Root Study

High frequency root dynamics: sampling and interpretation using replicated robotic minirhizotrons

High Frequency Root Dynamics: Sampling and Interpretation Using Replicated Robotic Minirhizotrons

Contact Info

Product

Resources

About