Interactive webapp for classification of images of fruits and vegetables into fresh or rotten made using Convolutional neural networks in python ,tensorflow and keras .Rembg (@rembg,https://github.com/danielgatis/rembg) used for removing the image background .Hosted using streamlit
Link - https://share.streamlit.io/shreyasvaidya/fresh-or-rotten/main/app.py
Working(in case some dependency breaks again)
Code for training saved_model.h5 can be found at https://github.com/Shreyasvaidya/Fresh-rotten_Classifier
Developed as a part of the Engineering Design project website "Annadanam"(@ED-Annadanam)
Team members(Who also worked on other parts of the website)
1.Samarth Sudhirkumar Bhalerao
2.Vikash Kumar
3.Uppala Giridhar
4.Basanti Meena
5.Saurabh kumar Meena
6.Vaidehi Bala(was a part of the team in the first semester,was an integral contributor in ideation)
Credits
Instructors
1.Dr.Manish Narwaria(narwaria@iitj.ac.in)
2.Dr.Sucharita Dey(sdey@iitj.ac.in)
Dataset taken from https://www.kaggle.com/datasets/raghavrpotdar/fresh-and-stale-images-of-fruits-and-vegetables
https://youtube.com/playlist?list=PLQVvvaa0QuDfhTox0AjmQ6tvTgMBZBEXN used For learning how to train a deep learning model
https://www.pluralsight.com/guides/deploying-image-classification-on-the-web-with-streamlit-and-heroku For hosting using streamlit and remotely deploying using heroku
Seniors who provided technical advice
Rohan Singh (https://github.com/rohansingh9001) on where to start and which resources to use
Soham Sonawane (https://github.com/killbotXD) for hosting