Inspiration

We want to provide high quality helps to the farmers to get informed decision about the farming strategy.

What it does

This is a Flutter based application. The application has 2-main parts: Crop-Recommendation System and Plant-Disease Prediction Crop-Recommendation System predicts which crop needs to be planted for high-quality yield by using the parameters like soil components(Nitrogen, Potassium, Phosphorus, and pH) and climate constraints (Rainfall, Humidity, and Temperature). The model will predict which crop to be sown on the bases of the provided parameters by the user. Plant-Disease Prediction predicts the plant disease. the user will provide a live photo and the DL model will instantly predict the disease.

How we built it

We build a Flutter based application which is integrated with 2-Deep learning model. First DL model is uses Deep Convolutional Neural Networks which predicts over 17 classes with an accuracy of 98%. Second DL model is based on Artificial Neural Networks which is having a total of 22 classes and predicts with an accuracy of 96%.

Challenges we ran into

We faced issues in integration the Deep Learning model to our Flutter based application. We also faced problems in increasing the accuracy of our Deep learning model.

Accomplishments that we're proud of

We integrated our Deep Learning model to our flutter application using FastAPI.

What we learned

This was our second time creating a Flutter application so we learnt a lot of features of flutter and learnt how integrate the DL model to our Flutter application.

What's next for Cropance

Hardware implementation for collecting the data using Adrino.

Built With

  • deep-learning
  • flutter
  • keras
  • tensorflow
Share this project:

Updates