Project Plant AI

Just the other day I was reading some articles and I saw an online news article on a recent suicide case of Indian farmers with the headline “Farmer couple commits suicide after killing minor”, this made my heart skip a beat. And I thought to address the problems of yield losses due to diseases with what I love, tech. Many Indian farmers or crop growers (since most of our research was based out of India) are not yet aware of the right kind of fertilizers, manure, or biocides, and also modern farming practices​ and crop diseases over the last year have led up to 67% yield losses among Indian farmers which is humongous portion!

Plant AI is a web application that helps to easily diagnose diseases in plants from plant images using Machine Learning available on the web. We provide an interface on the website where you can upload images of your plant leaves. The idea this project proposes to use smartphones which most farmers already have access to, to diagnose diseases in their crops through plant images early and in an accurate manner​. The app would also provide the farmers with easy and actionable ways to solve the disease which would be quite helpful for most farmers since most plant diseases occur due to farmers being aware of the right techniques​.

This project uses TensorFlow and Azure ML to train the ML models. Grappler (distributed with TensorFlow) and AzureML to optimize the model to run on-device on mobile phones. React and TensorFlow.js to build out the website and perform on-device inferences and Azure Static Web Apps and Azure Blob Storage in some aspects to host the web app. Finally, we also provide an API since though on-device inferences are the best option in most such cases, having the ML offloaded to the server-side could be quite useful in some scenarios and serve our API with Azure Kubernetes Services.

Rishit Dagli

Rishit Dagli

TED-X, Ted-Ed speaker

Rishabh Singh

Rishabh Singh

Frontend & UI/UX Designer

Rucha Yagnik

Rucha Yagnik

Full-stack Developer

Other sessions from: Sessions October 2021

Implementation of Linear Regression using tfjs

Implementation of Linear Regression using tfjs

- Introduction to Tensorflow.js - Architecture of TFJS - Project ...

Kartikey Rawat Kartikey Rawat
#NoMoreMatildas

#NoMoreMatildas

Artificial Intelligence (AI) is the pinnacle of digitization; it is a tech...

Carla Florida Carla Florida
Introduction to Siamese Network

Introduction to Siamese Network

This session focuses on building a Siamese Network using TensorFlow functio...

Rohan Raj Rohan Raj
Can a Machine Learning System Outperform Humans in Emotion Recognition?

Can a Machine Learning System Outperform Humans in Emotion Recognition?

In 1961, when Carl Sagan learned that a NASA spacecraft to Venus would...

Valentina Zhang Valentina Zhang
Introduction to Azure Health Bot

Introduction to Azure Health Bot

Azure Health Bot enables users to build a Health Bot by using built-in or...

Shreyan J D Fernandes Shreyan J D Fernandes
Yesasvini Sai Meghana Chikkam Yesasvini Sai Meghana Chikkam
Learning Deep Learning

Learning Deep Learning

Recently, Microsoft has released two Learning Paths on MS Learn covering ...

Anaam Sayeda Anaam Sayeda