Return to Normal View

Australian Plant Phenomics Facility

The University of Adelaide Australia



  • Pricing and availability - please contact us by e-mail or on +61 2 6218 3430.

PlantScan provides non-invasive analyses of plant structure (topology, surface orientation, number of leaves), morphology (leaf size, shape, colour, area, volume) and function by utilising cutting edge information technology including high resolution cameras and three-dimensional (3D) reconstruction software.

PlantScan Hardware
The High Resolution Plant Phenomics Centre, located at CSIRO Agriculture and Food in Canberra, associated with Neo Vista System Integrators, located in Sydney, have developed the next generation phenotyping platform – PlantScan - to capture information on plant structure and function on an industrial scale enabling the discovery of new traits and the selection of varieties for tomorrow's agriculture.

PlantScanCurrently, there is a disjuncture in scientists’ capacity to measure at the molecular and functional levels. Genomics is providing a huge amount of data retrieved using high-throughput technologies but our ability to measure life at the whole-plant level (Phenomics) has lagged behind. There is, therefore, a need to level the balance between the speed of genotyping and phenotyping. Thus, high-throughput phenotyping tools are becoming vital for research looking at improving crop performance and adapting crop varieties to climate change. By providing an integrated multi-sensing platform with a range of imaging sensors, PlantScan helps restore this balance by increasing researchers’ capacity to precisely and accurately quantify the biological processes involved in the development and functioning of plants and, this, with greater detail, frequency and objectivity than traditional methods.

PlantScan is composed of a double conveyor belt, manually loaded, with plants held in position on pot carriers. Individual plants are identified by bar code. A first transfer station diverts the plant to a split conveyor belt which accurately positions the plant for imaging using laser proximity sensors. A rotating motor, fitted with an incremental encoder with up to 65,536 lines per revolution and mounted on a scissor-lift platform ensures the plant is scanned from every angle. A top-imaging system complements the 3D information collected from the sides. After imaging, the plant is conveyed to a second transfer station before being ejected by an actuator onto a double gravity belt. Plants are then manually unloaded and transported back to their growing environments. All motion control and image acquisition was designed in LabView. The interface was built in a hybrid 32- and 64-bit architecture, which automatically detects the various sensors available to the instrument and provides the user with all configuration options available for specific sensors. PlantScan relies on its modular design for the flexibility to match to the requirements of many different researchers and biological systems. The user may choose the combination of modalities to be measured from:

  • Light Detection and Ranging Sensors (LiDAR) for 3D plant canopy architecture measurement with a resolution of 100 microns;
  • Two far-Infrared imaging cameras for canopy temperature measurements with a resolution of 0.045°K;
  • Multi-wavelength imaging for pigment or chemical composition spanning both visible and near-Infrared.

The software provides a range of feedback on the motion controls, e.g. defective proximity sensors, as well as a graphical interface displaying the location of the pot carriers in the system in real time. During image acquisition, the system is closed to ensure reproducibility of imaging conditions, and avoid any safety hazards from moving parts and the class 2A LiDAR lasers, although the lasers are eye-safe. The light spectrum is generated by fluorescent light run on a 75 kilo-Hertz electrical signal to avoid noise in the acquired images. The light is diffuse to approximate Lambertian conditions within the imaging chamber. The system uses standard industrial components from reliable manufacturers with worldwide service.

PlantScan 3D imagePlantScan is able to scan very small seedlings or plants, from a few centimeters to a couple of meters in height and up to a meter thick, with equal accuracy. Data are captured with their contextual information and collated into one multi-layer data file before being stored in a purpose-built database.


PlantScan Software
One the main challenges from this data acquisition is the automation of the data analysis. To address this, our team developed an automated software solution that aims at extracting accurate phenotypic measurements of the plants from the acquired data. The pipeline is composed of:

  • A plant surface mesh reconstruction that aims at reconstructing a full 3D model of the studied plant.
  • A morphological mesh segmentation that aims at identifying the different morphological parts of the plant (i.e. main stem, branches, and leaves) in order to allow further analysis of these different parts.
  • An accurate phenotypic data extraction that uses the segmented regions in order to extract the phenotypic data of interest from the plant meshes.
  • A longitudinal matching features that aims at tracking the different morphological parts of the plant over time, and thus allowing a longitudinal monitoring of the different phenotypic data.

For more information, please contact Warren Creamers.