Plant Phenotyping

Part of the ECE 542 Virtual Symposium (Spring 2020)

Plant phenotyping of a plant is describing the visual or observable characteristics such as height, biomass, leaf shape and so on. In this project we are using phenotyping to find the water requirement of the plant. Two major areas of focus are the tip detection and the skeletal identification of the plant. Visual data of the plants are used for phenotyping. The goal of the project is to use Deep Learning to automatically identify leaf tips and collars. Specifically, we use Tensorflow Object Detection API modified to Key Point Detection with the Inception Net Pre-trained architecture to detect leaf tips and collars with a reasonable accuracy.