Sudoku-Solver

Implementing Augmented Reality based Sudoku Solver, that uses OpenCV-Python to detect, recognize and solve the sudoku puzzle in the video frame. The answers for the sudoku puzzle is then displayed on the video frame.

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development and testing purposes:

1. Clone the project with link: https://github.com/Srikanth-Kb/Sudoku-Solver.git

Prerequisites

OpenCV-Python libraries (CV2)
Keras - Deep learning framework
Numpy library

Working principle

1. Detect the Sudoku Puzzle in a Video frame, including the corners of the enclosing square.
   
2. Remove noise, align the borders and obtain the coordinates of each of 81 squares.
   
3. Detect the presence of numbers in each square (if any).
   
4. Predict the digits in numbered squares.
   
5. Solve the sudoku for the predicted digits in corresponding squares.
   
6. Use inverse perspective transform, to display the solved sudoku in the same video frame.
   

Running the Code

Run the following file for webcam implementation of the system.

python3 webcam.py