Plant Disease Detector
- Goal: Use images of plants to detect any kind of diseases
- Target audience: Farmers, Plant enthusiasts
Current development deployment on IBM Cloud K8s: http://184.172.252.173:30120/
Abstract
Agriculture in the United States is a major driver of economic growth, accounting for 11% percent of total employment and $132.8 billion dollars to U.S. GDP in 2017 [1]. Increasing efficiency and reducing waste is a key research area in agricultural, particularly as climate change creates favorable conditions for pests and diseases to proliferate throughout a crop. Our project provides a tool to help farmers discover diseases within their crop by using IBM’s visual recognition API to identify diseases of interest to farm owners. The tool provides farmers a platform to train a machine learning model customized to their problem while also providing a dashboard for visualizing past predictions, if needed. The model is trained and validated on the PlantVillage-Dataset.
Architecture Diagram
Technology Stack
IBM Watson Visual Recognition
Amazon Cognito
Amazon S3
React
Node.js
Kubernetes
Docker
Design Thinking
The Web app allows family farms to use computer vision to measure crop health and analyze historical trends with ease.