Breeding drought-tolerant crop varieties with higher water use efficiency could help maintain food supply to a growing population and save valuable water resources. Fast and accurate phenotyping is currently a bottleneck in the process towards attaining this goal, as available plant phenotyping platforms have an excessive cost for many research institutes or breeding companies. Here we describe a simple and low-cost, automatic platform for high-throughput measurement of plant water use and growth and present its utilisation to assess the drought tolerance of two soybean genotypes. The platform allows the evaluation of up to 120 plants growing in individual pots. A cart moving in only one direction carries the measuring and watering devices. Watering and measurement routines allow the simulation of multiple water regimes for each plant individually and indicate the timing of measurement of soil water content and image capture for growth estimation. Water use, growth and water use efficiency were measured in two experiments with different water scenarios. Differences in water use efficiency between genotypes were detected only in some treatments, emphasising the importance of phenotyping platforms to evaluate a genotype’s phenotype under a broad range of conditions in order to capture valuable differences, minimising the chance of artefacts and increasing precision of measurements.

Conventional field phenotyping for drought tolerance, the most important factor limiting yield at a global scale, is labor-intensive and time-consuming. Automated greenhouse platforms can increase the precision and throughput of plant phenotyping and contribute to a faster release of drought tolerant varieties. The aim of this work was to establish a framework of analysis to identify early traits which could be efficiently measured in a greenhouse automated phenotyping platform, for predicting the drought tolerance of field grown soybean genotypes. A group of genotypes was evaluated, which showed variation in their drought susceptibility index (DSI) for final biomass and leaf area. A large number of traits were measured before and after the onset of a water deficit treatment, which were analyzed under several criteria: the significance of the regression with the DSI, phenotyping cost, earliness, and repeatability. The most efficient trait was found to be transpiration efficiency measured at 13 days after emergence. This trait was further tested in a second experiment with different water deficit intensities, and validated using a different set of genotypes against field data from a trial network in a third experiment. The framework applied in this work for assessing traits under different criteria could be helpful for selecting those most efficient for automated phenotyping.

There has been latelly a significant progress in automating tasks for the agricultural sector. One of the advances is the development of robots, based on computer vision, applied to care and management of soy crops. In this task, digital image processing plays an important role, but must solve some important problems, like the ones associated to the variations in lighting conditions during image acquisition. Such variations influence directly on the brightness level of the images to be processed. In this paper we propose an algorithm to segment and measure automatically the leaf area of soy plants. This information is used by the specialists to evaluate and compare the growth of different soy genotypes. This algorithm, based on color filtering using the HSI model, detects green objects from the image background. The segmentation of leaves (foliage) was made applying Mathematical Morphology. The foliage area was calculated counting the pixels that belong to the segmented leaves. From several experiments, consisting in applying the algorithm to measure the foliage of about fifty plants of various genotypes of soy, at different growth stages, we obtained successful results, despite the high brightness variations and shadows in the processed images.