The LemnaTec FieldAIxpert is an automatic sensor-to-plant field phenotyping system that is installed over plot testing sites. It moves on rails installed along a plot site (X-direction), and has a portal structure spanning over the plots. On this structure, the sensor head moves across the plot area (Y-direction), while an axis allows height adjustment of the sensor head (Z-direction). With accurate and repeatable positioning of the sensor head, the system enables time-course imaging and scanning of the plants in the plot area. It is weatherproof and designed to run measurements of the plants throughout the growing season form germination to harvest time. Inside the plot area, plants can be cultivated in the soil, or they can be placed in containers. Depending on the planting pattern and spacing, the system can record images at canopy level or at single plant level. With continuous and unsupervised operation, the system can monitor plant development throughout the growth period. Programmable imaging patterns enable application-oriented image recording and scanning.

The FieldAIxpert is delivered with a comprehensive software suite. The software enables the users to provide sample information and control cameras and lights. After imaging all data are stored and accessible for image and data processing. LemnaControl is the interface with which users operate the system. With LemnaControl, users carry out all settings that are required for image recording with the system. In particular, his comprises settings for the camera(s) and lights included in the system. All settings can be combined and scheduled to configure recordings. The available functions are tailored to the corresponding hardware. Each hardware system delivered by LemnaTec comes with its own control software adapted for the functions of that system. The LemnaGrid app uses an intuitive graphical programming environment, which allows for easy integration of different image analysis algorithms. LemnaGrid phenotyping software offers high flexibility for user programming. The image processing pipeline extracts desired properties/features from the original image and stores results in a dedicated storage system. LemnaGrid can include functions of state-of-the-art machine learning tools, next to standard image analysis to detect and extract features of interest. Guided by the LemnaGrid graphical user interface, users can build their own analytical pipeline with pre-defined devices.

LemnaExperiment serves to manage experiments and measurements, and to schedule and run analyses. Moreover, it enables viewing, visualizing, and exporting results. It provides capabilities to access and browse through recorded images and to display metadata for samples and measurements. As core function, LemnaExperiment selects recordings for analysis and assigns an image analysis pipeline –written in LemnaGrid or in Lemna3D – to a given set of recordings. After assignment, the analyses are run in the background by the Analysis Executor. Parameters of interest can be selected in LemnaExperiment so that they are displayed in the analysis output.