/aws-textract-sudoku-solver

scan and solve sudoku with AWS textract and backtracking algorithm!

Primary LanguagePythonMIT LicenseMIT

aws-textract-sudoku-solver

tp

Requirements

$ pip install awscli

$ pip install opencv-python

After installation of awscli, login by

$ aws configure

Now create a bucket in AWS S3 by

$ aws s3api create-bucket --bucket <bucket-name> --region <region-name> --create-bucket-configuration LocationConstraint=<region-name>

Working

The file detect.py, saves the image of the board and uploads it to the S3 bucket in AWS. Now by using Textract API with boto3 this image is analyzed with FeatureTypes as TABLES. The digits are then fed to the sudoku solver which solves and prints the solution.

Usage

Run the file detect.py, and put your phone with the sudoku board image in front of your webcam

solved sample->

solved