计算机工程与应用 ›› 2007, Vol. 43 ›› Issue (34): 220-222.

• 工程与应用 • 上一篇    下一篇

基于视觉的象棋棋盘识别

杜俊俐1,2,张景飞3,黄心汉1   

  1. 1.华中科技大学 自动控制系,武汉 430074
    2.中原工学院 计算机科学系,郑州 450007
    3.郑州大学 工程力学系,郑州 450001
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2007-12-01 发布日期:2007-12-01
  • 通讯作者: 杜俊俐

Chess-board recognition based on vision

DU Jun-li1,2,ZHANG Jing-fei3,HUANG Xin-han1   

  1. 1.Department of Automatic Control,Huazhong University of Science and Technology,Wuhan 430074,China
    2.Department of Computer Science,Zhongyuan Insititute of Technology,Zhengzhou 450007,China
    3.Department of Engineering Mechanics,Zhengzhou University,Zhengzhou 450001,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-12-01 Published:2007-12-01
  • Contact: DU Jun-li

摘要: 基于视觉的象棋棋盘识别是象棋机器人软件的重要组成部分,其核心工作是棋盘图像二值化和棋子识别。针对棋盘全局二值化存在的问题,提出了基于相邻像素灰度差阈值的棋盘图像二值化方法;针对棋子文字方向任意的现象,提出了3种基于统计特征的棋子文字识别方法。经实验验证,这些方法处理速度快、效果理想。

关键词: 棋盘识别, 灰度差阈值, 年轮统计, 傅立叶变换

Abstract: Chess-board recognition based on vision is an important part of a chess robot softwart system.It’s key problems are binarization of chess-board image and character recognition.To save the problem caused by full chess-board binarization way,the binarization method based on the difference threshold of neighbor pixels’ gray-level is given.To treat the random of a character’s direction,three methods of character recognition based on statistics are given.Experiments show that they are fast and high quality.

Key words: chess-board recognition, threshold of gray scales’ difference, annual ring statistic, FFT