Skip to content

Viroslav/CV_Chess_project

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

10 Commits
 
 
 
 
 
 

Repository files navigation

Chess Detector and Adviser (CDA)

The main idea is to help chess players who like to play on a real board, keep track of their moves, help with decisions, and if you are alone, simulate the game with a real person using a computer)

We used classical CV algorithms (like blur+binarization) to detect the board, also ProjectiveTransform and Approximation to get it to the right position. Then we cut it to the squares and train classificator (Neural Network with 5 Convolution layers). After we get the array of prediction and use it with python-chess library to detect the board statement, possible top move, position estimation etc.

Example of detection and best move-adviser:

IMG_3071 OUHRpZsGWoo-PhotoRoom (1)

PS. All main code can be found in CV_project.jpynb

About

This is chess recognition using CV.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published