Handwritten Line Detection via an EM Algorithm

In this paper we present a handwritten line segmentation method devised to work on documents composed of several paragraphs with multiple line orientations. The method is based on a variation of the EM algorithm for the estimation of a set of regression lines between the connected components that compose the image. We evaluated our method on the ICDAR2009 handwriting segmentation contest dataset with promising results that overcome most of the presented methods. In addition, we prove the usability of the presented method by performing line segmentation on the George Washington database obtaining encouraging results.

[1]  Its'hak Dinstein,et al.  2009 10th International Conference on Document Analysis and Recognition Line segmentation for degraded handwritten historical documents , 2022 .

[2]  G. Louloudisa,et al.  Text line detection in handwritten documents , 2008 .

[3]  Georgios Louloudis,et al.  ICDAR 2009 Handwriting Segmentation Contest , 2009, ICDAR.

[4]  Andreas Keller,et al.  Lexicon-free handwritten word spotting using character HMMs , 2012, Pattern Recognit. Lett..

[5]  Georgi Gluhchev,et al.  Handwritten document image segmentation and analysis , 1993, Pattern Recognit. Lett..

[6]  Abdel Belaïd,et al.  Noname manuscript No. (will be inserted by the editor) A General Approach for Multi-oriented Text Line Extraction of Handwritten Documents , 2011 .

[7]  Venu Govindaraju,et al.  Line separation for complex document images using fuzzy runlength , 2004, First International Workshop on Document Image Analysis for Libraries, 2004. Proceedings..

[8]  Alicia Fornés,et al.  On Influence of Line Segmentation in Efficient Word Segmentation in Old Manuscripts , 2012, 2012 International Conference on Frontiers in Handwriting Recognition.

[9]  Laurence Likforman-Sulem,et al.  Text line segmentation of historical documents: a survey , 2007, International Journal of Document Analysis and Recognition (IJDAR).

[10]  George D. C. Cavalcanti,et al.  Text Line Segmentation Based on Morphology and Histogram Projection , 2009, 2009 10th International Conference on Document Analysis and Recognition.

[11]  Syed Saqib Bukhari,et al.  Script-Independent Handwritten Textlines Segmentation Using Active Contours , 2009, 2009 10th International Conference on Document Analysis and Recognition.

[12]  D. Rubin,et al.  Maximum likelihood from incomplete data via the EM - algorithm plus discussions on the paper , 1977 .

[13]  Jihad El-Sana,et al.  Text line segmentation for gray scale historical document images , 2011, HIP '11.

[14]  Paolo Nesi,et al.  Projection based segmentation of musical sheets , 1999, Proceedings of the Fifth International Conference on Document Analysis and Recognition. ICDAR '99 (Cat. No.PR00318).