首页 | 本学科首页   官方微博 | 高级检索  
     

Correction of Scanned Book Images
引用本文:钟晓刚,李新友,唐泽圣. Correction of Scanned Book Images[J]. 清华大学学报, 2003, 8(5): 624-627
作者姓名:钟晓刚  李新友  唐泽圣
作者单位:DepartmentofComputerScienceandTechnology,TsinghuaUniversity,Beijing100084,China
摘    要:Because of the spine, a book cannot be set flat on a flat bed scanner. As a result, the image suffers from shape distortion, variable brightness, and blurred characters. This paper describes a method based on book shape analysis to address these problems. The algorithm utilizes the relation between the distance from a line segment on the book to the flat bed and its projected length. The correction process uses edge extraction, iterative boundary point computations, linear image transformation, and histogram modification. Experimental results show that the shape distortion and brightness variance problems are perfectly resolved by the method.

关 键 词:图像处理 图像修正 平板扫描仪 扫描图像 书脊 图象失真 边缘抽取

Correction of Scanned Book Images
ZHONG Xiaogang,LI Xinyou,TANG Zesheng. Correction of Scanned Book Images[J]. Tsinghua Science and Technology, 2003, 8(5): 624-627
Authors:ZHONG Xiaogang  LI Xinyou  TANG Zesheng
Affiliation:ZHONG Xiaogang,LI Xinyou,TANG ZeshengDepartment of Computer Science and Technology,Tsinghua University,Beijing 100084,China
Abstract:Because of the spine, a book cannot be set flat on a flat bed scanner. As a result, the image suffers from shape distortion, variable brightness, and blurred characters. This paper describes a method based on book shape analysis to address these problems. The algorithm utilizes the relation between the distance from a line segment on the book to the flat bed and its projected length. The correction process uses edge extraction, iterative boundary point computations, linear image transformation, and histogram modification. Experimental results show that the shape distortion and brightness variance problems are perfectly resolved by the method.
Keywords:image distortion  edge extraction  book image  image processing  shape restoration
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号