A research team has developed a powerful unsupervised deep learning network that can accurately separate wood and leaf components in 3D point clouds of trees—without the need for labor-intensive data labeling.

A research team has developed an innovative three-dimensional (3D) tree modeling method that dramatically improves accuracy in estimating tree structure and volume.

A research team has developed FreezeNet, a lightweight deep learning model that uses smartphone-captured images to accurately assess freeze injury in wheat seedlings.

A research team has combined magnetic resonance imaging (MRI) and positron emission tomography (PET) to non-invasively track how the sugar beet disease known as syndrome “basses richesses” (SBR) damages taproot structure and disrupts sugar distribution.

A research team has demonstrated that spectroscopy combined with partial least squares regression (PLSR) can accurately estimate plant leaf traits, but models built at one site often fail elsewhere.