Line Segmentation in Handwritten Assamese and Meetei Mayek Script Using Seam Carving Based Algorithm

  • Chandan Jyoti Kumar
  • Sanjib Kr. Kalita
Conference paper
Part of the Springer Proceedings in Physics book series (SPPHY, volume 166)


Line segmentation is a key stage in an Optical Character Recognition system. This paper primarily concerns the problem of text line extraction on color and grayscale manuscript pages of two major North-east Indian regional Scripts, Assamese and Meetei Mayek. Line segmentation of handwritten text in Assamese and Meetei Mayek scripts is an uphill task primarily because of the structural features of both the scripts and varied writing styles. Line segmentation of a document image is been achieved by using the Seam carving technique, in this paper. Researchers from various regions used this approach for content aware resizing of an image. However currently many researchers are implementing Seam Carving for line segmentation phase of OCR. Although it is a language independent technique, mostly experiments are done over Arabic, Greek, German and Chinese scripts. Two types of seams are generated, medial seams approximate the orientation of each text line, and separating seams separated one line of text from another. Experiments are performed extensively over various types of documents and detailed analysis of the evaluations reflects that the algorithm performs well for even documents with multiple scripts. In this paper, we present a comparative study of accuracy of this method over different types of data.


Document Image Text Line Line Segmentation Handwritten Text Handwritten Document 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Chaudhuri BB, Pal U, Mitra M (2002) Automatic recognition of Printed Oriya Script. Sadhana 27(1):23–34Google Scholar
  2. 2.
    Avidan S, Shamir A (2007) Seam carving for content-aware image resizing. ACM Trans Graph (TOG) 26(3):9 (Proceedings of ACM SIGGRAPH 2007, article no 10)Google Scholar
  3. 3.
    Solanki P, Bhatt M (2013) Printed Gujarati script OCR using hopfield neural network. Int J Comput Appl 69(13):0975–8887Google Scholar
  4. 4.
    Pal U, Wakabayashi T, Kimura F (2007) Handwritten Bangla compound character recognition using gradient feature. In: 10th international conference on information technologyGoogle Scholar
  5. 5.
    Roy A, Bhowmik TK, ParuiS K, Roy U (2005) A novel approach to skew detection and character segmentation for handwritten Bangla words. In: Proceedings of the digital imaging computing: techniques and applications, pp 125–132Google Scholar
  6. 6.
    Chaudhuri BB, Bera S (2009) Handwritten text line identification in indian scripts. In: Proceedings of 10th international conference on document analysis and recognitionGoogle Scholar
  7. 7.
    Saha S, Basu S, Nasipuri M, Basu DK (2010) A hough transform based technique for text segmentation. J Comput 2(2):134–141Google Scholar
  8. 8.
    Obaida MA et al (2011) Skew correction function of OCR: stroke-whitespace based algorithmic approach. Int J Comput Appl 28(8):7–12Google Scholar
  9. 9.
    Priyanka N, Pal S, Mandal R (2010) Line and word segmentation approach for printed documents. IJCA Spec Issue Recent Trends Image Process Pattern Recogn RTIPPR 1:30–36Google Scholar
  10. 10.
    Saabni R, El-Sana J (2011) Language-independent text lines extraction using seam carving. In: International conference on document analysis and recognition (ICDAR 2011), pp 563–568Google Scholar
  11. 11.
    Adiguzel H, Sahin E, Duygulu P (2012) A hybrid approach for line segmentation in handwritten documents. In: Proceedings of international conference on frontiers in handwriting recognition, pp 501–506Google Scholar
  12. 12.
    Palakollu S, Dhir R, Rani R (2011) Segmentation of handwritten Devanagri Script. Int J Comput Sci Inf Technol 2(3):1244–1247Google Scholar
  13. 13.
    Arvanitopoulos N, Stisstrunk S (2014) Seam carving for text line extraction on color and grayscale historical manuscripts. In: ICHFRGoogle Scholar
  14. 14.
    Koppula VK, Negi A (2011) Fringe map based text line segmentation of printed Telugu document images. In: International conference on document analysis and recognitionGoogle Scholar

Copyright information

© Springer India 2015

Authors and Affiliations

  1. 1.Department of Computer ScienceGauhati UniversityGauhatiIndia

Personalised recommendations