Go game strategy marking based on rudimentary binary machine vision

• Main Author: 蕭崇佑
• Other Authors: 張英德
• Format: Others
• Language: zh-TW
• Published: 2002
• Online Access: http://ndltd.ncl.edu.tw/handle/29980712651774417774

碩士 === 國立海洋大學 === 電機工程學系 === 90 === Connected Components Labeling Algorithm is a labeling motion in connection with some pixels that have the same regional property and connectivity, to distinguish different blocks by giving the unique label. We can find there is a similar relation between stones in the game of Go, many connected stones consist of a stone string, many stone strings consist of a stone clustering. To determine the final result, we calculate the side with the much grounds surrounded by stone clustering is the winner. Therefore, the main purpose of this thesis is to discuss how to combine connected components labeling algorithm with the basic rules of the game of Go, and creating the database of the patterns of the stones distribution, including connected, cut and connected to the boundary. It also applies to recognize the stone blocks and calculate the grounds surrounded by the game of Go. Then, we use some examples to put to the proof and discuss. In the end, we try to use the method of building database of the stone distribution to create basic corner evolution pattern, by way of pattern registering and displaying to reach the aim of learning the Go game. Keywords: Connected Components Labeling Algorithm, Binary, Go Game