This Team is Archived and no longer visible on the main VIP Teams page.

Digital Phenomics: Machine-Vision-based Novel Feature Extraction of Characterization of Wheat/Rice Panicles

The project involves the development of an imaging system which will be able to provide 2D images of wheat/rice plants with panicles.



The integration of phenomics and genomics could revolutionize the field of plant breeding. The integration of these two fields however, depends on the development and deployment of high-throughput phenotyping systems which have heavily leveraged machine-vision technologies. In this project, we’d like the team to come up with a system which could facilitate reliable identification of the panicles of wheat (and rice if possible) plants, and to provide the capability in counting the number of spikes and kernels. Furthermore, when/if a panicle is identified in an image, the algorithm should be able to facilitate in classifying the variety of the plant.  Please see the Controlled-Environment Phenotyping Facility info for more background on the project:


The project could involve the development of an imaging system which able to provide and a) 2D images of wheat/rice plants with panicles; and b) High resolution 3D images of individual wheat/rice panicles. Once the images are established, machine vision algorithms need to be developed in order to identify panicles in 2D images and to quantify the morphological and geometrical characteristics of the identified panicles. Further, using established characteristics, help to classify the variety of the imaged plants; from the 3D images, algorithms need to be developed to facilitate counting the number of spikes and kernels of each panicle. It also would be of interests to establish 3D morphological and geometrical traits of the panicles. 

Relevant Technologies:

Computer vision (2D/3D imaging; spike/kernel counting); image analysis(2D/3D morphological trait establishement); Machine learning (classification etc.)


Programming skills; knowledge in image analysis and/or machine vision is a plus

Meeting Time:

Fall 2021: TBA