I am currently creating an intelligent agent for a game solver. This system will be hosted in a computer different from the host of the game which it will play. Hence, I am looking for similar researches or projects that uses machine vision for a game solver. The game is a board game/involving tiles, by the way.
These are the things I think I need to do:
- Edge detection
- Pattern recognition
- Board/tile detection similar to that is being used to detecting chess boards
I really cannot find something that can be beneficial to me using Google search, even using advance searching in Google scholar does not give me relevant research.
It’s going to be a multi-step process, no single algorithm is going to do it all for you, and few researchers ever give out their full algorithms they used to solve problems. But all the pieces you’ll want to use to do your own work are well documented.
Basically, you’ll want to use edge detection and maybe also region growing to find the board, and then something like Hough circle detection to find the game pieces (assuming it’s checkers, if it’s minesweeper, you open up a whole other can of worms with OCR).
You won’t need anything related to patter recognition or machine learning for the vision part. You’ll be giving it a very clean image of a board with very strict rules about how it will look and act. So just write your code for that board and that problem.