计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (7): 161-164.

• 图形、图像、模式识别 • 上一篇    下一篇

工程图纸字符串及标注信息提取

范 帆1,关佶红2   

  1. 1.复旦大学 计算机学院,上海 200433
    2.同济大学 计算机科学与技术系,上海 201804
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2012-03-01 发布日期:2012-03-01

Recognition of characters and annotations on engineering drawings

FAN Fan1, GUAN Jihong2   

  1. 1.School of Computer Science, Fudan University, Shanghai 200433, China
    2.Department of Computer Science and Technology, Tongji University, Shanghai 201804, China
  • Received:1900-01-01 Revised:1900-01-01 Online:2012-03-01 Published:2012-03-01

摘要: 工程图纸中字符串及标注信息的提取是工程图纸自动化处理极为重要的组成部分,是进行尺寸理解、图像理解等高层次理解的前提和基础。提出一种基于工程图纸知识的预分割字符串及标注信息提取方法,重点关注工程图纸中以表格形式存在的字符串以及图元标注信息的解析、定位、提取。通过前期处理保持字符串与字符串、图元与标注信息之间的逻辑联系,解析获得字符串的坐标信息,对字符串所在的区域进行水平化,去除杂质线段等操作,以达到最佳的识别效果。

关键词: 预分割字符串, 标注信息提取, 文字识别

Abstract: Recognition of characters and annotations on engineering drawings is very important to automated processing of engineer drawings. It is the precondition and foundation of size and image understanding. A new algorithm is presented for recognition of characters and annotations which is based on pre-cut characters and annotations mainly focusing on the recognition, location and extraction. By preserving the logical relations of characters and graphic annotations, what could have are coordinates of characters through complanation and denoising.

Key words: pre-cut characters, annotations extraction, Optical Character Recogntion(OCR)