Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/17615
標題: 線繪圖影像自動化系統之研究
An automatic system for processing and interpreting line drawing images
作者: 黃瓊瑤
關鍵字: vectorization;向量化;line drawing images;image process;automation;線繪圖影像;影像處理;自動化
出版社: 應用數學系
摘要: 
我們設計並研製一個可以自動處理和解析線繪圖影像的系統,紙上的線繪圖經由掃描器輸入之後,本系統可將文字和圖形分離,辨認直線、曲線,尋找出所有線圍成的區塊。
本系統可辨別影像中文字的部分和圖形的部分,但並不作文字本身的辨識,所有不同大小、字型,方向的文字都會被分離。剩下的圖形部分則被轉換成向量化的物件-包括直線及曲線。最後,再分析向量化的線段而找到圖形上的各個最封閉區塊。圖形經向量化之後,本系統即可提供新增、刪除及修改等功能,且圖形上的各物件可被指定適當的屬性以加強圖形之解說能力。
由於向量化之後的圖形均涵蓋於原來的影像圖形之內,因此準確性高,轉換成向量表示法可大量節省儲存空間,最重要者,則將手繪圖形轉為電腦可操作之圖形,使往後對該圖形之修改變為更容易、更省時,並以經濟的向量化電子檔方式永久保存。

We have designed and implemented an automatic system for processing and interpreting paper-based line drawing images. The system is composed of four key components: Preprocessing, Line Segmentation, Curve Segmentation, and Block Recognition. Processing an image includes removing noise and separating text from graphics. Line Segmentation includes thinning and converting graphics into line segments. Curve segmentation converts “short” line segments into arcs. Block recognition identifies polygons on the images. Problems and difficulties occurring at each stage were analyzed . Appropriate solutions were proposed and incorporated into our system design. By exercising many test cases on our system, we demonstrated that our approach is accurate and very reliable. Possible applications of our system include automation cadastral maps and engineering drawings.
URI: http://hdl.handle.net/11455/17615
Appears in Collections:應用數學系所

Show full item record
 
TAIR Related Article

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.