期刊文献+

基于改进Hough变换的文本图像倾斜校正方法 被引量:33

Skew detection and correction method of document images based on improved Hough transform
在线阅读 下载PDF
导出
摘要 文本图像在扫描输入时产生的倾斜现象会对后续的页面分割及光学字符识别(OCR)处理产生很大的影响,而传统的标准Hough变换虽然具有对噪声不敏感,不依赖于直线连续性的优点,但由于计算量偏大,速度慢,在实用时有较大的局限性。提出一种基于改进的Hough变换的文本图像倾斜校正方法,通过在变分辨率图像中采用不同的文本方向提取算法,及选择合理投票门限等改进Hough变换的措施,减小了由图像区域及文字笔画粗细所产生的对倾角判定的不利影响,并使用基于偏移值的方法实现页面倾斜的快速校正。实验结果表明,该算法实现了大范围高精度的文本图像倾角的快速检测,具有较强的实用性。 The skew of document images originated from scanning will bring much trouble to page segmentation and optical character recognition (OCR). Although traditional standard Hough transform is insensitive to noise and discontinuous, it has limitation in application because of huge calculation. A skew detection and correction of document images based on improved Hough transform was presented. Text-line direction extraction in multi-resolution image and reasonable voting threshold selection were introduced to reduce the affect brought by image skew. Experimental results show that this method is effective and accurate in skew detection and correction of document images.
出处 《计算机应用》 CSCD 北大核心 2007年第7期1813-1816,共4页 journal of Computer Applications
关键词 文本图像 HOUGH变换 数学形态学 游程平滑 变分辨率图像 document image Hough transform morphology run-length smoothing multi-resolution image
作者简介 周冠玮(1982-),男,江西南昌人,硕士研究生,主要研究方向:图像处理;zgw_text@yahoo.com.cn 平西建(1953-),男,河南新乡人,教授,博士生导师,主要研究方向:图像处理、信息隐藏; 程娟(1979-),女,河南郑州人,博士研究生,主要研究方向:图像处理。
  • 相关文献

参考文献10

  • 1ISHITANI Y.Document skew detection based on local region complexity[C]// 1993 IEEE International Conference on Document Analysis and Recognition.Tsukuba,Japan:IEEE Computer Society,1993:49-52.
  • 2O'GORMAN L.The document spectrum for page layout analysis[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,1993,15(11):1162-1173.
  • 3PSTL W.Detection of linear oblique structure and skew scan in digitized documents[C]// Proceedings of the 8th International Conference on Pattern Recognition.Paris,France:[s.n.],1986:687-689.
  • 4CHEN M,DING X Q.A robust skew detection algorithm for grayscale document image[C]// 1999 IEEE International Conference on Document Analysis and Recognition.Bangalore,India:[s.n.]1999:617-620.
  • 5BALLARD D H.Generalizing the hough transform to detect arbitrary shapes[J].Pattern Recognition,1981,13(4):111-122.
  • 6瞿洋,杨利平.Hough变换OCR图象倾斜矫正方法[J].中国图象图形学报(A辑),2001,6(2):178-181. 被引量:62
  • 7章毓晋.图像处理和分析[M].北京:清华大学出版社,1999..
  • 8潘武模,焦扬,王庆人.Hough变换在中文名片图像倾斜校正中的应用[J].中文信息学报,2001,15(3):50-56. 被引量:18
  • 9李政,杨扬,颉斌,王宏.一种基于Hough变换的文档图像倾斜纠正方法[J].计算机应用,2005,25(3):583-585. 被引量:20
  • 10王姝华,李佐,蔡士杰.基于直线连续性的页面倾斜检测与校正[J].计算机辅助设计与图形学学报,2001,13(8):736-741. 被引量:18

二级参考文献21

  • 1徐建华.图象处理与分析[M].北京:科学出版社,1992..
  • 2CIARDIELLO G, SCAFUR G, DEGRANDI M, et al. An Experimental System for Office Document Handling and Text Recognition[A]. Proceedings of Ninth International Conference on Pattern Recognition[C],1998.739-743.
  • 3BAIRD HS. The Skew angle of printed documents[A].SPSE 40th Annual Conference and Symposium on hybrid Imaging System[C], 1987. 739-743.
  • 4ILLINGWORTH J, KITTLER J. The adaptive Hough Transform[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1987,9(5): 691-698.
  • 5LI H, LAVIN MA, Le MASTER RJ. Fast Hough Transform[R]. Research Report RC11080(#49754),IBM T.J.Watson Research Center,1985.
  • 6RISSE T. Hough Transform for Line Recognition: Complexity of Evidence Accumulation and Cluster Detection[J]. Computer Vision,Graphics and Image Processing, 1989,46(3): 327-345.
  • 7OKUN O, YU Y, ROBUST M. Text Detection from Binarized Document Images[A]. Proceedings of the 16th International Conference on Pattern Recognition(ICPR2002)[C], 2002.61-64.
  • 8Chiu Lyu,Proc 3rd ICDAR,1995年,1149页
  • 9Hashizume A,Pattern Recognit Lett,1986年,4卷,125页
  • 10Li H,Research Report RC110 80 (#49754),IBMT.J.Watson Research Center,1985年

共引文献437

同被引文献223

引证文献33

二级引证文献143

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部