Root systems play an essential role in ensuring plant productivity. Experiments conducted in controlled environments and simulation models suggest that root geometry and responses of root architecture to environmental factors should be studied as a priority. However, compared with aboveground plant organs, roots are not easily accessible by non-invasive analyses and field research is still based almost completely on manual, destructive methods. Contributing to reducing the gap between laboratory and field experiments, we present a novel phenotyping system (GROWSCREEN-Rhizo), which is capable of automatically imaging roots and shoots of plants grown in soil-filled rhizotrons (up to a volume of ~18 L) with a throughput of 60 rhizotrons per hour. Analysis of plants grown in this setup is restricted to a certain plant size (up to a shoot height of 80 cm and root-system depth of 90 cm). We performed validation experiments using six different species and for barley and maize, we studied the effect of moderate soil compaction, which is a relevant factor in the field. First, we found that the portion of root systems that is visible through the rhizotrons’ transparent plate is representative of the total root system. The percentage of visible roots decreases with increasing average root diameter of the plant species studied and depends, to some extent, on environmental conditions. Second, we could measure relatively minor changes in root-system architecture induced by a moderate increase in soil compaction. Taken together, these findings demonstrate the good potential of this methodology to characterise root geometry and temporal growth responses with relatively high spatial accuracy and resolution for both monocotyledonous and dicotyledonous species. Our prototype will allow the design of high-throughput screening methodologies simulating environmental scenarios that are relevant in the field and will support breeding efforts towards improved resource use efficiency and stability of crop yields.

Graphical AbstractFinely-grained annotated datasets for image-based plant phenotyping Massimo Minervini, Andreas Fischbach, Hanno Scharr, Sotirios A. TsaftarisIn this paper we present a collection of benchmark datasets for the development and evaluation of computer vision and machine learning algorithms in the context of plant phenotyping. We provide annotated imaging data and suggest suitable evaluation criteria for plant/leaf segmentation, detection, tracking as well as classification and regression problems. The Figure symbolically depicts the data available together with ground truth segmentations and further annotations and metadata. Research Highlights• First comprehensive annotated datasets for computer vision tasks in plant phenotyping.• Publicly available data and evaluation criteria for eight challenging tasks.• Tasks include fine-grained categorization of age, developmental stage, and cultivars.• Example test cases and results on plant and leaf-wise segmentation and leaf counting. ABSTRACTImage-based approaches to plant phenotyping are gaining momentum providing fertile ground for several interesting vision tasks where fine-grained categorization is necessary, such as leaf segmentation among a variety of cultivars, and cultivar (or mutant) identification. However, benchmark data focusing on typical imaging situations and vision tasks are still lacking, making it difficult to compare existing methodologies. This paper describes a collection of benchmark datasets of raw and annotated top-view color images of rosette plants. We briefly describe plant material, imaging setup and procedures for different experiments: one with various cultivars of Arabidopsis and one with tobacco undergoing different treatments. We proceed to define a set of computer vision and classification tasks and provide accompanying datasets and annotations based on our raw data. We describe the annotation process performed by experts and discuss appropriate evaluation criteria. We also offer exemplary use cases and results on some tasks obtained with parts of these data. We hope with the release of this rigorous dataset collection to invigorate the development of algorithms in the context of plant phenotyping but also provide new interesting datasets for the general computer vision community to experiment on. Data are publicly available at http://www.plant-phenotyping.org/datasets.

Stress caused by environmental factors evokes dynamic changes in plant phenotypes. In this study, we deciphered simultaneously the reaction of plant growth and chlorophyll fluorescence related parameters using a novel approach which combines existing imaging technologies (GROWSCREEN FLUORO). Three different abiotic stress situations were investigated demonstrating the benefit of this approach to distinguish between effects related to (1) growth, (2) chlorophyll-fluorescence, or (3) both of these aspects of the phenotype. In a drought stress experiment with more than 500 plants, poly(ADP-ribose) polymerase (PARP) deficient lines of Arabidopsis thaliana (L.) Heynh showed increased relative growth rates (RGR) compared with C24 wild-type plants. In chilling stress, growth of PARP and C24 lines decreased rapidly, followed by a decrease in Fv/Fm. Here, PARP-plants showed a more pronounced decrease of Fv/Fm than C24, which can be interpreted as a more efficient strategy for survival in mild chilling stress. Finally, the reaction of Nicotiana tabacum L. to altered spectral composition of the intercepted light was monitored as an example of a moderate stress situation that affects chlorophyll-fluorescence related, but not growth-related parameters. The examples investigated in this study show the capacity for improved plant phenotyping based on an automated and simultaneous evaluation of growth and photosynthesis at high throughput.

Image-based plant phenotyping is a growing application domain of computer vision in agriculture. A key task is the segmentation of all individual leaves in images. Here we focus on the most common rosette model plants Arabidopsis and young tobacco. Although leaves do share appearance and shape characteristics, the presence of occlusions and variability in leaf shape and pose, as well as imaging conditions, render this problem challenging. The aim of this paper is to compare several leaf segmentation solutions on a unique and first of its kind dataset containing images from typical phenotyping experiments. In particular, we report and discuss methods and findings of a collection of submissions for the first Leaf Segmentation Challenge (LSC) of the Computer Vision Problems in Plant Phenotyping (CVPPP) workshop in 2014. Four methods are presented: three segment leaves via processing the distance transform in an unsupervised fashion, and the other via optimal template selection and Chamfer matching. Overall, we find that although separating plant from background can be achieved with satisfactory accuracy (>90% Dice score), individual leaf segmentation and counting remain challenging when leaves overlap. Besides, accuracy is lower for younger leaves. We find also that variability in datasets does affect outcomes. Our findings motivate further investigations and development of specialized algorithms for this particular application, and that challenges of this form are ideally suited for advancing the state of the art. Data are publicly available (http://www.plantphenotyping.org/CVPPP2014-dataset) to support future challenges beyond segmentation within this application domain.

Abstract. Root phenotyping is a challenging task, mainly because of the hidden nature of this organ. Only recently, imaging technologies have become available that allow us to elucidate the dynamic establishment of root structure and function in the soil. In root tips, optical analysis of the relative elemental growth rates in root expansion zones of hydroponically-grown plants revealed that it is the maximum intensity of cellular growth processes rather than the length of the root growth zone that control the acclimation to dynamic changes in temperature. Acclimation of entire root systems was studied at high throughput in agar-filled Petri dishes. In the present study, optical analysis of root system architecture showed that low temperature induced smaller branching angles between primary and lateral roots, which caused a reduction in the volume that roots access at lower temperature. Simulation of temperature gradients similar to natural soil conditions led to differential responses in basal and apical parts of the root system, and significantly affected the entire root system. These results were supported by first data on the response of root structure and carbon transport to different root zone temperatures. These data were acquired by combined magnetic resonance imaging (MRI) and positron emission tomography (PET). They indicate acclimation of root structure and geometry to temperature and preferential accumulation of carbon near the root tip at low root zone temperatures. Overall, this study demonstrated the value of combining different phenotyping technologies that analyse processes at different spatial and temporal scales. Only such an integrated approach allows us to connect differences between genotypes obtained in artificial high throughput conditions with specific characteristics relevant for field performance. Thus, novel routes may be opened up for improved plant breeding as well as for mechanistic understanding of root structure and function.