So I have started working on an image processing project to learn a bit about this topic. The project is quite simple. I am building a chess bot which will make moves based on feed from a camera.
On one side of the board, there will be a human and on the other side my bot. After making move, the player will tap a key which will trigger the bot. Firstly a image of the board will be taken. On this image, a program will detect the grid and which cells are occupied and by which colored piece. After that heuristics will be used to display the next move of bot.
Today I finally got chess pieces and here is the result of initial program that I have written in python. The proect is live at Github repository [ChessBot].
The initial image :
After converting the image to GrayScale and sharpening :
After detecting the corners and making border :
After creating grid on board :
There is a lot of scope of improvement as this is just an initial program. Moving ahead...
COMMENTS :
This blog is like Notes-to-Self and is not intended to teach