Scrabble is a commonly played word game in which players take turns forming words using a set of seven letter tiles and placing them onto a grid, following placement rules similar to a crossword puzzle. Various applications exist for helping players with forming words from their tiles, but few account for all of the possible positions on the board simply because manually entering the board state is tedious.
This project presents an image processing algorithm for extracting and recognizing characters from Scrabble game boards for use in general Scrabble backends. In the test images used to evaluate this algorithm, boards were digitized with 87% character accuracy on average.
There are no reviews yet.