The problem AgroGrow solves

Inspiration-
According to the UN, Nearly 1/2 of all fruit & vegetables produced globally are wasted each year. With decreasing energy and resources across the planet, AgroGrow was made to promote better yield among farmers and better health among malnourished people and animals. Also, we know that it might be difficult for farmers to use websites, we have integrated the functionalities of this website in the form of a WhatsApp Bot as well in order to reach a greater target audience.
What it does
The Website-

1. Crop Recommendation - In the crop recommendation application, the user can provide the soil data(Nitrogen, Phosphorus, Potassium concentrations, pH value, Rainfall in mm, State and City) from their side and the application will recommend which crop the user should grow. We have used an open weather API to automatically check the temperature, moisture, and humidity for the location provided.
2. Fertilizer Recommendation - In the fertilizer recommendation application, the user can input the soil data and the type of crop they are growing, and the application will predict what the soil lacks or has an excess of and will recommend improvements accordingly.
3. Disease Prediction - In the disease prediction application, the user can upload the image of the crop with the disease, and the model will predict the disease, tell the cause of it and recommend its cure.
   Add Ons-
4. Soil Classification - We have also created this add-on streamlit based soil classification application, for you to classify soil with the help of an image of the soil.
5. Nursery Locator - We have added an option for you to locate a nursery near you to buy the needed crops and fertilizers.
6. Try the WhatsApp Bot Feature - We have also added the button to use our WhatsApp bot on the website. Clicking on the button would lead you to connect with out Twilio based Whatsapp Bot.

Twilio Based WhatsApp Bot also provides all the above functionalities.

Challenges we ran into

While we were working on this project the limited availability of data sets was a big problem we faced. Build different ML model for different diseases detection was another huge problem . Deploying was a problem too. Using these models in a WhatsApp Bot was a very big challenge for us.