Plants

The Parametric Plant Project is a new exciting initiative at the University of Saskatchewan for creating a digital plant phenotype — a digital representation of the physical characteristics of plants and crops. This project part of the $37M Canada First Research Excellence Fund grant to the Global Institute for Food Security (GIFS) at the University of Saskatchewan. The project includes an academic and industrial research consortium in Digital Plant Imaging & Modeling. Research activities are directed towards high-resolution, statistical, and mechanistic model of plant development. I am leading the Image Acquisition & Analysis sub-theme of the CFREF grant.

Time Lapse Photography and Plant Image Recognition

As technology advances, there is an increase in the use of drone imaging to aid farmers in caring for their crops. Significant research has been done on the utility of areal images, but most of the known methods do not examine on-site fixed cameras. We believe that time-lapse cameras taking images continuously can not only provide us data with a higher resolution than drone pictures, but also can show how the plants evolve through time. Data extracted from this approach, combined with drone images, can give more accurate and applicable results.

Participants

Approved
Danny Huang
Approved
Ian Stavness

Publications

2018

Ubbens, Jordan; Cieslak, Mikolaj; Prusinkiewicz, Przemyslaw; Stavness, Ian

The use of plant models in deep learning: an application to leaf counting in rosette plants Journal Article

Plant Methods, 14 (1), pp. 6, 2018.

Links | BibTeX

2017

Aich, Shubhra; Stavness, Ian

Leaf Counting With Deep Convolutional and Deconvolutional Networks Inproceedings

ICCV Workshop on Computer Vision Problems in Plant Phenotyping, 2017.

Links | BibTeX

Ubbens, J R; Stavness, I

Deep plant phenomics: A deep learning platform for complex plant phenotyping tasks Journal Article

Frontiers in Plant Science, 8 , 2017.

Links | BibTeX