Sudoku solver using openCV and Tensorflow

Steps to run the program

1. Clone the repo
2. Create a virtual environment using python -m venv venv
3. Install all the dependencies using pip install -r requirements.txt
4. Run the sudokuMain.py file

The program uses openCV to detect and extract the numbers on the board. It then uses tensorflow to identify the digits. Then backtracking algorithm is used to solve the board.

Original Image

Preprocessing Image

Identifying Contours

Selecting Board

Predicting Numbers

Final Image