There are about 150 million farmers in India. Farmers are the backbone of our society. They are the ones who provide us with all the food that we eat. As a result, the entire population of the country depends upon farmers. Crop diseases are a major threat to food security, but their rapid identification remains difficult in many parts of the world due to the lack of necessary infrastructure. Improper care can lead to serious effects on plants and affects the quality, quantity or productivity of the yield. The combination of increasing technology and recent advances in computer vision made possible by deep learning which paved the way for disease diagnosis. Getting affected by a disease is common in plants due to various factors such as fertilizers, cultural practices followed, etc. Leaves of a plant can be used to determine the health status of a plant. Thus, we came up with an approach to detect plant disease by uploading an image of the leaf.

What it does

Plant disease detection:Detecting whether a leaf is healthy or not. Expert advice: The user can connect to an expert to seek guidance(Video call and chat features are provided). Forums: The user can post their queries related to plant health and can also reply to the other users queries. Fertilizer advice: Advice will be given to the user to improve their plant health based on soil nutrients(Nitrogen, Phosphorus, Potassium). E-learning: Displays some informative resources to the user related to plant-health, soil and fertilizer.

How we built it

We built our Project using Django framework. HTML, CSS and JS are used for designing webpage.

Plant disease detection: We built an API (using Flask Framework) which takes a leaf image as input from our website, deep learning model predicts the plant health. The prediction is sent back to the website. Expert advice: Here we used Django channels to establish a connection between farmer and an expert. Forums: User queries and replies are stored in database(used Django in-built database).

Built with 🛠️

Languages and Tools:

bootstrap css3 django figma flask git html5 javascript linux mysql pandas python pytorch sqlite tensorflow

Challenges we ran into

  1. While establishing the connection between two users, we faced many errors.
  2. Displaying the response from an API in our website.

Accomplishments that we're proud of

  1. Building up a deep learning model was really a challenging task on our part, as we all were new to machine learning.
  2. We learnt how to build an API from scratch.
  3. We feel building an end-to-end project in a short period of time is itself a great accomplishment.

What we learned

  1. We learnt how to use django channels.
  2. We learnt how to deal with django in built databases.
  3. Gained experience with deep learning modules such as pytorch.
  4. During this journey we gained knowledge in the field of agriculture.

What's next for Farmfully

  1. We can add regional languages as well, so that even a farmer who don't know english can use our website.
  2. E-commerce feature can be added, where farmer can buy the Fertilizers recommended by Farmfully.
  3. We can also add search engines for retrieving farmer-specific data.
Share this project: