User menu

Accès à distance ? S'identifier sur le proxy UCLouvain

Benchmarking of Document Image Analysis Tasks for Palm Leaf Manuscripts from Southeast Asia

Primary tabs

download
  • Open access
  • PDF
  • 7.79 M
Kesiman, Made Windu Antara [Laboratoire Informatique Image Interaction (L3i), Université de La Rochelle, France] Valy, Dona [UCL] Burie, Jean-Christophe [Laboratoire Informatique Image Interaction (L3i), Université de La Rochelle, France] Paulus, Erick [Department of Computer Science, Universitas Padjadjaran, Bandung , Indonesia] Suryani, Mira [Department of Computer Science, Universitas Padjadjaran, Bandung , Indonesia] Verleysen, Michel [UCL] Ogier, Jean-Marc [Laboratoire Informatique Image Interaction (L3i), Université de La Rochelle, France]

This paper presents a comprehensive test of the principal tasks in document image analysis (DIA), starting with binarization, text line segmentation, and isolated character/glyph recognition, and continuing on to word recognition and transliteration for a new and challenging collection of palm leaf manuscripts from Southeast Asia. This research presents and is performed on a complete dataset collection of Southeast Asian palm leaf manuscripts. It contains three different scripts: Khmer script from Cambodia, and Balinese script and Sundanese script from Indonesia. The binarization task is evaluated on many methods up to the latest in some binarization competitions. The seam carving method is evaluated for the text line segmentation task, compared to a recently new text line segmentation method for palm leaf manuscripts. For the isolated character/glyph recognition task, the evaluation is reported from the handcrafted feature extraction method, the neural network with unsupervised learning feature, and the Convolutional Neural Network (CNN) based method. Finally, the Recurrent Neural Network-Long Short-Term Memory (RNN-LSTM) based method is used to analyze the word recognition and transliteration task for the palm leaf manuscripts. The results from all experiments provide the latest findings and a quantitative benchmark for palm leaf manuscripts analysis for researchers in the DIA community.

Document type Article de périodique (Journal article) – Article de recherche
Access type Accès libre
Publication date 2018
Language Anglais
Journal information "Journal of Imaging" - Vol. 4, no.43, p. 27 (2018)
Peer reviewed yes
issn 2313-433X
Publication status Publié
Affiliation UCL - SST/ICTM/ELEN - Pôle en ingénierie électrique
Keywords Document image analysis; Binarization; Character recognition; Text line segmentation; Word recognition; Transliteration; Palm leaf manuscript; Dataset; Benchmark; Experimental test
Links
  1. tranScriptoriumhttp://transcriptorium.eu/
  2. READ Project—Recognition and Enrichment of Archival Documentshttps://read.transkribus.eu/
  3. IAM Historical Document Database (IAM-HistDB)—Computer Vision and Artificial Intelligencehttp://www.fki.inf.unibe.ch/databases/iam-historical-document-database
  4. Ancient Lives: Archivehttps://www.ancientlives.org/
  5. Document Image Analysis—CVISION Technologieshttp://www.cvisiontech.com/library/pdf/pdf-document/document-image-analysis.html
  6. Ramteke R. J., Invariant Moments Based Feature Extraction for Handwritten Devanagari Vowels Recognition, 10.5120/392-585
  7. Singh Siddharth Kartar, Dhir Renu, Rani Rajneesh, Handwritten Gurmukhi Numeral Recognition using Different Feature Sets, 10.5120/3361-4640
  8. Sharma Dharamveer, Jhajj Puneet, Recognition of Isolated Handwritten Characters in Gurmukhi Script, 10.5120/850-1188
  9. Aggarwal Ashutosh, Singh Karamjeet, Singh Kamalpreet, Use of Gradient Technique for Extracting Features from Handwritten Gurmukhi Characters and Numerals, 10.1016/j.procs.2015.02.116
  10. Ashlin Deepa, Int. J. Adv. Trends Comput. Sci. Eng., 3, 481 (2014)
  11. Kasturi Rangachar, O’Gorman Lawrence, Govindaraju Venu, Document image analysis: A primer, 10.1007/bf02703309
  12. Paper History, Case Paphttp://www.casepaper.com/company/paper-history/
  13. Doermann, 1055 (2014)
  14. Chamchong, 55 (2010)
  15. (2016)
  16. Balinese Alphabet, Language and Pronunciationhttp://www.omniglot.com/writing/balinese.htm
  17. Khmer Manuscript—Recherchehttp://khmermanuscripts.efeo.fr/
  18. Kesiman Made Windu Antara, Valy Dona, Burie Jean-Christophe, Paulus Erick, Sunarya I. Made Gede, Hadi Setiawan, Sok Kim Heng, Ogier Jean-Marc, Southeast Asian palm leaf manuscript images: a review of handwritten text line segmentation methods and new challenges, 10.1117/1.jei.26.1.011011
  19. Bezerra (2017)
  20. Arica N., Yarman-Vural F.T., Optical character recognition for cursive handwriting, 10.1109/tpami.2002.1008386
  21. O’Gorman, 107 (1997)
  22. Gatos B., Ntirogiannis K., Pratikakis I., DIBCO 2009: document image binarization contest, 10.1007/s10032-010-0115-7
  23. Howe Nicholas R., Document binarization with automatic parameter tuning, 10.1007/s10032-012-0192-x
  24. ICFHR2016 Competition on the Analysis of Handwritten Text in Images of Balinese Palm Leaf Manuscriptshttp://amadi.univ-lr.fr/ICFHR2016_Contest/
  25. Gupta Maya R., Jacobson Nathaniel P., Garcia Eric K., OCR binarization and image pre-processing for searching historical documents, 10.1016/j.patcog.2006.04.043
  26. Feng Meng-Ling, Tan Yap-Peng, Contrast adaptive binarization of low quality document images, 10.1587/elex.1.501
  27. Global image threshold using Otsu’s method—MATLAB graythresh—MathWorks Francehttps://fr.mathworks.com/help/images/ref/graythresh.html?requestedDomain=true
  28. Khurshid Khurram, Siddiqi Imran, Faure Claudie, Vincent Nicole, Comparison of Niblack inspired binarization methods for ancient documents, 10.1117/12.805827
  29. Sauvola J., Pietikäinen M., Adaptive document image binarization, 10.1016/s0031-3203(99)00055-2
  30. Rapid Feature Extraction for Optical Character Recognitionhttp://arxiv.org/abs/1206.0238
  31. Kumar Satish, Neighborhood Pixels Weights-A New Feature Extractor, 10.7763/ijcte.2010.v2.119
  32. Bokser M., Omnidocument technologies, 10.1109/5.156470
  33. Shishtla Praneeth, Ganesh V. Surya, Subramaniam Sethuramalingam, Varma Vasudeva, A language-independent transliteration schema using character aligned models at NEWS 2009, 10.3115/1699705.1699715
  34. Ocropy: Python-Based Tools for Document Analysis and OCR, 2018https://github.com/tmbdev/ocropy
  35. Homemade Manuscript OCR (1): OCRopy, Sacré Grlhttps://graal.hypotheses.org/786
  36. Breuel Thomas M., Ul-Hasan Adnan, Al-Azawi Mayce Ali, Shafait Faisal, High-Performance OCR for Printed English and Fraktur Using LSTM Networks, 10.1109/icdar.2013.140
  37. Saund Eric, Lin Jing, Sarkar Prateek, PixLabeler: User Interface for Pixel-Level Labeling of Elements in Document Images, 10.1109/icdar.2009.250
  38. Stamatopoulos Nikolaos, Gatos Basilis, Louloudis Georgios, Pal Umapada, Alaei Alireza, ICDAR 2013 Handwriting Segmentation Contest, 10.1109/icdar.2013.283
  39. PRImAhttp://www.primaresearch.org/tools/Aletheia
  40. Clausner C., Pletschacher S., Antonacopoulos A., Aletheia - An Advanced Document Layout and Text Ground-Truthing System for Production Environments, 10.1109/icdar.2011.19
Bibliographic reference Kesiman, Made Windu Antara ; Valy, Dona ; Burie, Jean-Christophe ; Paulus, Erick ; Suryani, Mira ; et. al. Benchmarking of Document Image Analysis Tasks for Palm Leaf Manuscripts from Southeast Asia. In: Journal of Imaging, Vol. 4, no.43, p. 27 (2018)
Permanent URL http://hdl.handle.net/2078.1/211073