机械图识别中基于RAG的待识字符的提取与处理

北京邮电大学学报 ›› 2000, Vol. 23 ›› Issue (2): 19-23.

机械图识别中基于RAG的待识字符的提取与处理

邹华¹,方卫宁²

1. 北京邮电大学程控交换技术与通信网国家重点实验室,北京 100876;
2. 北方交通大学机电工程学院,北京 100044

收稿日期:1999-09-20 出版日期:2000-03-10
作者简介:邹华(1969-), 女, 四川成都人, 北京邮电大学讲师, 硕士.

The RAG Based Separation and Processing of Characters in the Mechanical Drawing Intelligent Recognition

ZOU Hua¹, FANG Wei-ning²

1. National laboratory of Switching Technology and Telecommunication Networks, Beijing Universit of Posts and Telecommunications, Beijing 100876, China;
2. Department of Electromechanical Engineering, Northern Jiaotong University, Beijing 100044, China

Received:1999-09-20 Online:2000-03-10

摘要/Abstract

摘要： 提出了机械工程图智能识别中一种基于RAG的字符分离方法, 它在有效将图形和文字分开的同时, 能较好地解决数字中的小数点、度数以及虚线和点划线中短线的保护问题. 还对分离字符中待识字符的处理进行了探讨.

关键词: 图像识别, 图文分离, 矩形相邻图

Abstract: A RAG based method for character separation in the intelligent recognition of mechanical drawing is provided in this paper. By using this method, not only the graphic and character can be separated effectively, but the decimal point, degree mark, dash line, and short line in the center line can be kept in the graphic part as well. Furthermore, the processing method for the characters to be recognized is discussed in this paper also.

Key words: image recognition, graphic and character separation, rectangle adjacency graph

中图分类号:

TP391.41

邹华,方卫宁. 机械图识别中基于RAG的待识字符的提取与处理[J]. 北京邮电大学学报, 2000, 23(2): 19-23.

ZOU Hua, FANG Wei-ning. The RAG Based Separation and Processing of Characters in the Mechanical Drawing Intelligent Recognition[J]. JOURNAL OF BEIJING UNIVERSITY OF POSTS AND TELECOM, 2000, 23(2): 19-23.

[1]	邹国锋, 傅桂霞, 王科俊, 高明亮, 申晋. 自适应深度卷积神经网络模型构建方法[J]. 北京邮电大学学报, 2017, 40(4): 98-103.
[2]	荣婧潘玉利. 基于图像的水泥路面裂缝识别方法及应用[J]. 北京邮电大学学报, 2012, 35(6): 121-124.
[3]	王波涛, 孙景鳌, 蔡安妮. 结构矩及其快速算法[J]. 北京邮电大学学报, 2000, 23(3): 10-14.