Plastic-Model-Deployment
YouTube Link:
Problem Statement
Develop a reliable and efficient AI-based object detection model using drone images to detect plastic waste in rivers and demonstrate a feasible solution/system architecture for implementation, ultimately reducing the negative impact of plastic pollution on the environment and human health.
Objective
This project aims to:
- detect plastic waste in rivers and help in reducing water pollution
- to give the location of river and alsom the distance from the user's currnt location where plastic is present and also latitude and longitude of plastic
Proposed Solution
This project covers all aspects that need to be emphasised on to minimise the problem:
- plastic is detected
- location of river is given
- latitude and longitude of plastic is given -distance between plastic and the device/concerned authorities is given
Proposed TechStack
- Python
- Yolov8(for model training refer:ultralytics)
- Streamlit(for web app)
- FastAPI(for creating API endpoints)
- Docker(For dockerising the entire content for hosting)
- Cvat(For annotation purpose)
Steps to Start our WebApp:
- First CLone the repository
- Create a Virtual Environment and activate it
- pip install requirements.txt (basically fastapi ultralytics streamlit phonenumbers geopy PIL pathlib)
- Run python startscript.py to start the server and streamlit run Stream_lit.py to start the frontend
To start server only:
- docker build -t yolo .
- docker run -p 8000:8000 yolo
- now we can go on local host 8000 and test the API