A phenotype is the set of expressed traits of an organism. In plant breeding and research thousands of measurements are performed for selection of superior traits and underlying genetic analysis. Phenotyping of plants helps to understand how genetic information is expressed in different environments or stresses.

Phenotyping varies from modern high-throughput image analysis to hand measurements with a ruler. Advanced phenotyping systems are, despite positive developments, still expensive. Besides, in most cases computer vision / image analysis experts are needed to extract relevant information from the data.

At Plantfellow we have developed the AutoPheno platform. This is an automated and cost-efficient method for phenotyping plants using 3D imaging.

Why 3D imaging

By collecting a 3D image of a plant it becomes possible to study plant organs. Different plant traits such as height, leaf area and stem thickness can be measured and tracked over time. This information allows more complex analysis such as plant biomass calculations.

3D phenotyping is non-destructive and enables monitoring of plant traits over time. Its use ranges from trait discovery and QTL-analysis up to in-field phenotyping and (pre) selection for crosssings. Especially within biochemical research it tends to be crucial to track biomass development over time. Multiple measurements per day enable researchers to understand or proof the effectivity of a product.

Challenges

The challenge with 3D imaging of plants is occlusion: the overlapping of leaves or other plant material. Growth of the plant and development of more vegetative plant material increase this effect. Good project preparation does somewhat allow this challenge to be overcome. For instance by applying the right camera angle compared to the crop and selection of appropriate 3D imaging sensors.