This is the official code repository for Code Monks . This simple, light weight mobile application built in Flutter was developed during Accelathon 2.0 the online the annual inter college hackathon and also awarded Certificate of Excellence at IFPPC conducted by Sikkim Manipal University.
- Utkarsh Agarwal
- Satyam Kumar
- Akarsh Simha
- Anusha Deshmukh
Kaggle Dataset Used for Plant disease detection
ML model for crop prediction written on Colab
APK (Just in case application doesn't run locally)
A Flutter App , but with a blend of Machine Learning in it and intergrated with a whole lot of features which makes it probably the best of the apps I have worked so far.
So basically I along with a bunch of seniors from my college have built this "KisanSeva" app - a one stop for Farmers of India who are facing a whole lot issues nowadays .
App has multiple features - 1- Secured Authentication (via OTP) 2- Multilingual for efficient use 2- Rent tools / farming essentials 3- Plant disease detection 4- Crop prediction 5- Smart connect to prevent third party person take advantage 6- Know the Weather 7- Feed 8- Toll free number/ expert assistance
The best thing about the app is its built in 10 days !!!
The most challenging tasks we faced for this project were tflite implementation,using pandas,numpy and scikit learn to predict crop which is yet not too accurate.
- Flutter
- Tensorflow
- Weather API
- Firebase
- Firestore
- Cloud Messaging
- Firebase Authentication
- Rest APIs
- Google Teachable Machine
- Pandas
- Numpy
- Real-time Database
.jpg "hover text")
In order to run the application on your local device make sure to have flutter environment setup on your local device
Clone the repo and open it in any text editor - VS Code for example
Connect your mobile in USB Debug mode
write in terminal "flutter run"
Agriculture :- Since farmers are facing a lot of issues and there is protest going on which has worsened the situation so we decided to come up with a smart solution to the problem by including features like Rent tools and Smart connect to the Mandi which will surely propel the growth of farmers.
To remove the issue of providing statical data in crop prediction feature we are thinking of integrating the app with an IOT model where perhaps a Node MCU model will collect all data like soil pH , Rainfall etc and send it to the real-time database and using the ML model built in the Goggle Collab we can predict the crop. Also in Feed feature we didn't get any API which would provide news solely related to Agro industry so we are working on that.
.jpg){kind=link}
