|Table of Contents|

Skew Detection of Document Images Using Mathematical Morphology and Hough Transform

《南京理工大学学报》(自然科学版)[ISSN:1005-9830/CN:32-1397/N]

Issue:
2009年02期
Page:
178-182
Research Field:
Publishing date:

Info

Title:
Skew Detection of Document Images Using Mathematical Morphology and Hough Transform
Author(s):
WU Xin1ZHANG Zhi-wei2
1.School of Computer Science Science and Technology,NUST,Nanjing 210094,China;2.School of Engineering Science,University of Science and Technology of China,Hefei 230027,China
Keywords:
skew detection document image processing mathematical morphology Hough transform
PACS:
TP391.41
DOI:
-
Abstract:
To solve the problems that the existent methods detecting skew of documents have difficulties in precision,accuracy or speed,a method based on mathematical morphology and the Hough transform is presented.Morphology method is used to smooth the document image,eliminate the pixel noise and detect the edge of the text row,and the Hough Transform is applied to detect the skew angle.In order to improve the performance,fast morphology image transforms are implemented.The detection time for A4 paper is 1.78 s,and the biggest error is 0.2°.The test results show that this method is precise,accurate,fast and robust.

References:

[ 1] Yu B, Ja in A K. A robust and fast skew detection a-l gorithm for gene tic docum ents [ J]. Pattern Recogn-i tion, 1996, 29( 10) : 1599- 1629.
[ 2] 李政, 杨扬, 颉斌, 等. 一种基于H ough 变换的文档图像倾斜纠正方法[ J]. 计算机应用, 2005, 25 ( 3): 583- 585.
[ 3] 王姝华, 李佐, 蔡士杰. 基于最小二乘法的文档图像倾斜检测方法[ J]. 计算机应用与软件, 2001 ( 9): 43- 46.
[ 4] N ajm an L. U sing m athem atical mo rpho logy fo r do cum en t Skew estim ation [ A ]. In Pro c SPIE Docum ent Recognition and Retr ieva l XI[ C ]. San Jose: International Soc iety for Optical Eng ineer ing , 2004, 5296. 182- 191.
[ 5] Yan H. Skew co rrection o f docum ent im ages using inte rline cross-correlation[ J]. CVGIP: Graph ica lM odel and Im age Pro cessing, 1993, 55( 6): 538- 543.
[ 6] BoomGaard R V, Ba len R V. M ethods fo r fast m orpho log ica l im age transform s using bitm apped binary im ag es[ J]. CVGIP: Graphica lModel and Im age Processing, 1992, 54( 3): 252- 258.

Memo

Memo:
-
Last Update: 2012-11-19