Smartphones' computational performance and sensor quality show a rapid development. The most recent devices are already capable for completing tasks that require high computational capacity. This makes them applicable even for complex image processing operations.
One example for this is the detection of the moves of a chess game, on photos taken by the camera of a mobile device. In my thesis I make an application for Android platform which is able to recognize a chessboard and the moves made on it.
I analyse other applications made for similar purposes and I review the theoretical background of their function.
I divide the problem into two parts. I make a line detector algorithm for the recognition of the chessboard grid. For the detection of the moves I use a clustering algorithm of the colours on the board.
I write the application in Java language, using the OpenCV computer vision library, and I implement it also in mobile environment. By testing the application I conclude that it is applicable for the real-time tracking of a chess game’s moves.