The First Robust Mongolian Text Reading Dataset CSIMU-MTR

  • Yunxue ShaoEmail author
  • Guanglai Gao
  • Linbo Zhang
  • Zhong Zhang
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 322)


Text extraction from various text containers like document, born digital images, real scenes and videos has been a continuous interest in this field for more than a decade. Although a lot of work has been done on printed Mongolian document image analysis, there has little work on Mongolian text extraction from complex images. For the design and evaluation of Mongolian text extraction algorithms and systems, the availability of large-scale dataset is important. This paper first introduces a dataset named CSIMU-MTR which is built by the College of Computer Science of Inner Mongolia University. And then presents benchmark results using two state-of-the-art methods in text detection on this new dataset. The reported results serve as a baseline for evaluating the further works.


Mongolian text extraction dataset Scene text detection Maximally stable extremal regions 



This work was supported by program of higher-level talents of Inner Mongolia University.


  1. 1.
    Antonacopoulos A, Karatzas D, Ortiz Lopez J (2001) Accessing textual information embedded in internet images. In: Proceedings of SPIE, Internet Imaging II, vol 4311, pp 198–205Google Scholar
  2. 2.
    Merino-Gracia C, Lenc K, Mirmehdi M (2012) A head-mounted device for recognizing text in natural scenes. In: CBDAR’11, LNCS, vol 7139, pp 29–41Google Scholar
  3. 3.
    Gao J, Yang J (2001) An adaptive algorithm for text detection from natural scenes. In: CVPR’01, vol 2, pp 84–89Google Scholar
  4. 4.
    Epshtein B, Ofek E, Wexler Y (2010) Detecting text in natural scenes with stroke width transform. In: 2010 I.E. conference on computer vision and pattern recognition, pp 2963–2970Google Scholar
  5. 5.
    Weinman JJ, Butler Z, Knoll D, Feild J (2014) Toward integrated scene text reading. IEEE Trans Pattern Anal Mach Intell 36(2):375–387CrossRefGoogle Scholar
  6. 6.
    Yi C, Tian Y (2011) Text string detection from natural scenes by structure-based partition and grouping. IEEE Trans Image Process 20(9):2594–2605MathSciNetCrossRefGoogle Scholar
  7. 7.
    Koo H, Kim D (2013) Scene text detection via connected component clustering and nontext filtering. IEEE Trans Image Process 22(6):2296–2305MathSciNetCrossRefGoogle Scholar
  8. 8.
    Shahav A, Shafait F, Dengel A (2011) ICDAR 2011 robust reading competition challenge 1: reading text in scene images. In: 2011 international conference on document analysis and recognition, pp 1491–1496Google Scholar
  9. 9.
    Karatzas D, Mestre SR, Mas J, Nourbakhsh F, Roy PP (2011) ICDAR 2011 robust reading competition-challenge 1: reading text in born-digital images (web and email). In: 2011 international conference on document analysis and recognition (ICDAR), pp 1485–1490Google Scholar
  10. 10.
    Karatzas D, Antonacopoulos (2007) A colour text segmentation in web images based on human perception. Image Vision Comput 25(5):564–577Google Scholar
  11. 11.
    Lopresti D, Zhou J (2000) Locating and recognizing text in WWW images. Inf Retr 2:177–206CrossRefGoogle Scholar
  12. 12.
    Perantonis SJ, Gatos B, Maragos V (2003) A novel web image processing algorithm for text area identification that helps commercial OCR engines to improve their web image recognition efficiency. In: Second international workshop on web document analysis (WDA2003), pp 61–64Google Scholar
  13. 13.
    Gao G, Li W, Hou H, Li Z (2003) Multi-agent based recognition system of printed Mongolian characters. In: Proceedings of the international conference on active media technology, pp 376–381Google Scholar
  14. 14.
    Gao G, Su X, Wei H, Gong Y (2011) Classical Mongolian words recognition in historical document. In: Proceedings of the 11th international conference on document analysis and recognition (ICDAR), pp 692–697Google Scholar
  15. 15.
    Wei H, Gao G (2014) A keyword retrieval system for historical Mongolian document images. Int J Doc Anal Recognit 17(2):33–45CrossRefGoogle Scholar
  16. 16.
    Peng L, Liu C, Ding X et al (2010) Multi-font printed Mongolian document recognition system. Int J Doc Anal Recognit 13(2):93–106CrossRefGoogle Scholar
  17. 17.
    Karatzas D et al (2013) ICDAR 2013 robust reading competition. In: 2013 international conference on document analysis and recognition, pp 1484–1493Google Scholar
  18. 18.
    Dalal N, Triggs B (2005) Histograms of oriented gradients for human detection. In: Proceedings of IEEE conference on computer vision and pattern recognition, pp 886–893Google Scholar
  19. 19.
    Vedaldi A, Fulkerson B (2008) An open and portable library of computer vision algorithms.
  20. 20.
    Wolf C, Jolion J.-M (2006) Object count/area graphs for the evaluation of object detection and segmentation algorithms. Int J Doc Anal Recognit 8(4):280–296CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2015

Authors and Affiliations

  • Yunxue Shao
    • 1
    Email author
  • Guanglai Gao
    • 1
  • Linbo Zhang
    • 2
  • Zhong Zhang
    • 3
  1. 1.College of Computer ScienceInner Mongolia UniversityInner MongoliaPR China
  2. 2.China Academy of Transportation Sciences (CATS)BeijingChina
  3. 3.College of Electronic and Communication EngineeringTianjin Normal UniversityTianjinChina

Personalised recommendations