Snooker is a very popular game in the United Kingdom, but is also gaining in popularity in other countries. Table size in snooker is almost two times the size of a pool billiards table. Determining the ball location with the naked eye is thus even more difficult.
The goal of this diploma thesis is to detect snooker balls and determine their position on the table with the use of computer vision. We developed an algorithm that detects over 99% of the balls and determines the ball location
to centimetre accuracy. With a few modifications, it would also be possible to use our algorithm for ball detection on any billards table.
|
