Plant Pathology Challenge

Overview

Current plant disease diagnosis based on human scouting is time-consuming and expensive, and although computer-vision based models have the promise to increase efficiency, the great variance in symptoms due to age of infected tissues, genetic variations, and light conditions within trees decreases the accuracy of detection. We have collected high-quality RGB images of foliar diseases of apples and created an expert-annotated disease dataset. This dataset is used to set-up a Kaggle competition ‘Plant Pathology Challenge’ for FGVC7 workshop at CVPR2020.

Objectives of challenge are to train a model using images of training dataset to 1) Accurately classify a given image from testing dataset into different diseased category or a healthy leaf; 2) Accurately distinguish between many diseases, sometimes more than one on a single leaf; 3) Develop an algorithms for quantification of disease severity from the images taken under real-life variable conditions; 4) Deal with rare classes and novel symptoms; 5) Address depth perception—angle, light, shade, physiological age of the leaf; and 6) Incorporate expert knowledge in identification, annotation, quantification, and guiding computer vision to search for relevant features during learning. The three winners of the competition will be identified based on the demonstration of newly developed models to achieve one or more of the above objectives with high accuracy.

Competition Details

The competition page is here.

Competition Begins March 9 2020

Submission Deadline May 11 2020

Organizers

Awais Khan and Ranjita Thapa (Plant Pathology and Plant-Microbe Biology Section, Cornell AgriTech, Cornell University)

Serge Belongie (Department of Computer Science, Cornell Tech, Cornell University)

Acknowledgments

We acknowledge financial support from Cornell Initiative for Digital Agriculture (CIDA) and special thanks to Zach Guillian for help with data collection.