Rapid analysis of minirhizotron images

Observing plant roots is an important task in the comprehensive study of natural and agricultural environments. RootShape provides rapid and accurate root segmentation for collections of minirhizotron images with minimal guidance from expert users.

RootShape segmentation in action.
RootShape segmentation in action.

Machine learning optimized feature extraction

Topological Data Analysis techniques are used to extract critical features from minirhizotron images based on the the structure of the underlying data. These features are well suited for machine learning algorithms and can be coupled with the latest techniques in active learning for rapid, accurate analysis of minirhizotron image collections.

Expert informed automatic segmentation

With an intuitive active learning interface, the RootShape algorithms quickly refine the root segmentation process based on user feedback to accommodate sample specific features. Keeping the expert in the loop ensures consistent, accurate results across experiments.

Screenshot of the RootShape segmentation interface

Segmentation as a service

Root segmentation is the first step in many plant phenotyping and analysis experiments. Designed to integrate with existing analysis tools, RootShape provides consistent, high quality root segmentations that can be easily integrated into existing workflows. This low barrier to entry makes RootShape an easy option for root image segmentation and the first step in the phenotyping journey.

Learn more

RootShape is currently available to select customers for beta testing and refinement. Email to discuss to how to integrate RootShape into your workflow.