Traffic Sign Image Classification

This is a streamlit app which helps you to classify images of different traffic signs. In total we have 58 classes in the Data Set.

Getting Started

This is an example of how you may give instructions on setting up your project locally. To get a local copy up and running follow these simple example steps.

Prerequisites

Download and install Python 3.10.0

Installation

1. First check python version in terminal, if it's 3.10.0 then go ahead.

   python --version 

2. Go to folder where you want to clone this repository and use below command to clone this repo to your local machine.

   git clone https://github.com/ipiyushvaghela/TrafficSignClassification.git

3. Create virtual Environment

   python -m venv venvForTrafficSign --system-site-packages

   To Activate our virtual environment we use

   venvForTrafficSign/Script/Activate.bat

4. Install packages using requirements.txt

   pip install -r requirements.txt

5. Now check if streamlit is installed properly or not.

   streamlit version

Usage

1. Run below command in terminal to run streamlit application.

   streamlit run app.py

Now we are all set to go. just upload the image of any given class and it will classify that image and also it gives other 5 possible classes from which image could belong to.

Info about other folders

1. Folder named BestPerformingModels contains download file from which you can download 1 model which has .h5 extension and put that model in BestPerformingModels and that used in app.py file.
2. Folder named ModelBuildingIPYNBs contains IPYNB files in which all the code is present, by using those IPYNBs our model is created, you can customize the code according to your need.

Contact

Piyush Vaghela - @ipiyushvaghela - ipiyushvaghela@gmail.com

Project Link - github.com/ipiyushvaghela/TrafficSignClassification