Deep learning presents many opportunities for image-based plant phenotyping. Here we consider the capability of deep convolutional neural networks to perform the leaf counting task. Deep learning techniques typically require large and diverse datasets to learn generalizable models without providing a priori an engineered algorithm for performing the task. This requirement is challenging, however, for applications in the plant phenotyping field, where available datasets are often small and the costs associated with generating new data are high. In this work we propose a new method for augmenting plant phenotyping datasets using rendered images of synthetic plants. We demonstrate that the use of high-quality 3D synthetic plants to augment a dataset can improve performance on the leaf counting task. We also show that the ability of the model to generate an arbitrary distribution of phenotypes mitigates the problem of dataset shift when training and testing on different datasets. Finally, we show that real and synthetic plants are significantly interchangeable when training a neural network on the leaf counting task.Electronic supplementary materialThe online version of this article (10.1186/s13007-018-0273-z) contains supplementary material, which is available to authorized users.

Phyllotaxis, the distribution of organs such as leaves and flowers on their support, is a key attribute of plant architecture. The geometric regularity of phyllotaxis has attracted multidisciplinary interest for centuries, resulting in an understanding of the patterns in the model plants Arabidopsis and tomato down to the molecular level. Nevertheless, the iconic example of phyllotaxis, the arrangement of individual florets into spirals in the heads of the daisy family of plants (Asteraceae), has not been fully explained. We integrate experimental data and computational models to explain phyllotaxis in Gerbera hybrida. We show that phyllotactic patterning in gerbera is governed by changes in the size of the morphogenetically active zone coordinated with the growth of the head. The dynamics of these changes divides the patterning process into three phases: the development of an approximately circular pattern with a Fibonacci number of primordia near the head rim, its gradual transition to a zigzag pattern, and the development of a spiral pattern that fills the head on the template of this zigzag pattern. Fibonacci spiral numbers arise due to the intercalary insertion and lateral displacement of incipient primordia in the first phase. Our results demonstrate the essential role of the growth and active zone dynamics in the patterning of flower heads.

HighlightThe key modes of auxin-driven patterning— the formation of convergence points and canals—depend on whether auxin efflux and influx feed back on polar auxin transport synergistically or antagonistically.

Association mapping studies have enabled researchers to identify candidate loci for many important environmental tolerance factors, including agronomically relevant tolerance traits in plants. However, traditional genome-by-environment studies such as these require a phenotyping pipeline which is capable of accurately measuring stress responses, typically in an automated high-throughput context using image processing. In this work, we present Latent Space Phenotyping (LSP), a novel phenotyping method which is able to automatically detect and quantify response-to-treatment directly from images. We demonstrate example applications using data from an interspecific cross of the model C4 grass Setaria, a diversity panel of sorghum (S. bicolor), and the founder panel for a nested association mapping population of canola (Brassica napus L.). Using two synthetically generated image datasets, we then show that LSP is able to successfully recover the simulated QTL in both simple and complex synthetic imagery. We propose LSP as an alternative to traditional image analysis methods for phenotyping, enabling the phenotyping of arbitrary and potentially complex response traits without the need for engineering-complicated image-processing pipelines.

The model is able to reproduce differences in vine and fruit growth arising from various experimental treatments. This implies it will be a valuable tool for refining our understanding of kiwifruit growth and for identifying strategies to improve production.