- This is my 7th semester mini-project which is basically about recognising 42 types traffic signs and i made web app and deployed over heroku platform. I have used CNN classification algorithm and GTSRB - German Traffic Sign Recognition Benchmark dataset from kaggle.
- Welcome page of web-app
- As we all know from now on all things and mechanisms are becoming automatic and the best example of that is Tesla's self driving car. we all know Elon musk and heard about Tesla's self- Driving car. So one out of thousands factor which helps car to drive itself is traffic sign detection and recognition so basically i thought about let's make one part of it so here i made it. Obviously Self driving car has much better model's accurcay and all but here is one intution.
- I have used most off CNN and different languages to build this mini project. To build the model and train the model I have used python language and CNN architecture. To build this website I have used HTML, CSS, JavaScript and Bootstrap framework. and here is time i spent on each Technology.
- models : different trained model
- static : CSS, icons and JS files ( This is structure of flask microservice )
- templates : HTML files
- Procfile : required to deploy web-app on heroku platform.
- app.py : flask python file for making apis
- requirements.txt : dependencies and libraries i have used to make this project.
- welcome_page.png : screenshot of home page of deployed web-app
- First run following command to download external library
- python3 -m pip install -r requirement.txt
- After completed go to Run part.
- Go to project directory then run following command
- python3 app.py
- go to link given in terminal during starting kernel
- https://towardsdatascience.com/a-comprehensive-guide-to-convolutional-neural-networks-the-eli5-way-3bd2b1164a53
- https://www.geeksforgeeks.org/deploy-python-flask-app-on-heroku/
- Any feedback will be appriciated.