Text and Graphics Analysis in Engineering Drawings

  • Thomas C. Henderson


The meaning of an engineering drawing is expressed through text and graphics and the relations between them. Chapter 1 provided a detailed summary of the major approaches to their segmentation, and here we describe our own contributions on some specific applications. The goal is the fully automatic segmentation of text and graphics in an engineering drawing image, as well as its interpretation; that is, characters represented as pixels must be interpreted as to which specific character they represent. Of course, this is made difficult in that most engineering drawings use a variety of fonts, sizes, and orientations for characters—indeed, some are even hand-written. In addition, character segmentation is generally only part of a larger process: for example, dimension set analysis. Since the names and numbers extracted by the system are quite significant for manufacturing purposes, say in a reverse engineering application, then more likely than not, hypotheses put forward by the image analysis system will need to be corroborated by a human.


Radial Basis Function Branch Point Foreground Pixel Part Number Character Segmentation 
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. 10.
    H. Arai and K. Okada. Form Processing based on background Region Analysis. In Proceedings International Conference on Document Analysis and Recognition, pages 164–169, Ulm, Germany, 1997.Google Scholar
  2. 20.
    A.K. Chhabra. Neural Network based Text Recognition for Engineering Drawing Conversion. In IEEE International Conference on Neural Networks, pages 3774–3779. IEEE, 1994.Google Scholar
  3. 27.
    S. Diana, E. Trupin, Y. Lecourtier, and J. Labiche. Document Modeling for Form Class Identification. In N.A. Murshed and F. Bortolozzi, editors, Advances in Document Image Analysis, pages 176–187, Berlin, 1997. Springer Verlag.Google Scholar
  4. 41.
    T. Ha and H. Bunke. Model-based Analysis and Understanding of Check Forms. In H. Bunke, P.S.P. Wang, and H.S. Baird, editors, Document Image Analysis, pages 57–84, Singapore, 1994. World Scientific Pub Co.Google Scholar
  5. 43.
    T.C. Henderson. Discrete Relaxation. Oxford University Press, Oxford, 1990.Google Scholar
  6. 48.
    T.C. Henderson and L. Swaminathan. Form Analysis with the Nondeterministic Agent System (NDAS). In Proceedings of 2003 Symposium on Document Image Understanding Technology, pages 253–258, April 2003.Google Scholar
  7. 71.
    J. Liang, J. Ha, R.M. Haralick, and I.T. Phillips. Document Layout Structure Extraction using Bounding Boxes of Different Entities. In Proceedings 3rd IEEE Workshop on Applications of Computer Vision (WACV ’96), pages 278–283, December 1996.Google Scholar
  8. 81.
    D. Marr. Vision. W.H. Freeman and Company, New York, NY, 1982.Google Scholar
  9. 82.
    S. Marsland. Machine Learning, An Algorithmic Approach. Chapman & Hall, CRC Press, Boca Raton, FL, 2009.Google Scholar
  10. 87.
    R. Mohr and T.C. Henderson. Arc and Path Consistency Revisited. Artificial Intelligence, 28(2):225–233, March 1986.Google Scholar
  11. 93.
    M. Ondrejcek, J. Kastner, R. Kooper, and P. Bajcsy. Information Extraction from Scanned Engineering Drawings. Image Spatial Analysis Group NCSA-ISDA09-001, National Center for Supercomputing Applications, December 2009.Google Scholar
  12. 116.
    Y. Tang and J. Lin. Information Acquisition and Storage of Forms in Document Processing. In Proceedings International Conference on Document Analysis and Recognition, pages 170–174, Ulm, Germany, 1997.Google Scholar
  13. 125.
    D. Wang and S.N. Srihari. Analysis of Form Images. International Journal of Pattern Recognition and Artificial Intelligence, 8(5):1031–1051, 1994.CrossRefGoogle Scholar
  14. 126.
    Q. Wang, J. Shi, and D.D. Feng. A Uniform Framework of Representation and Structure Reconstruction for Generic Form Image. In Proceedings 7th International Conference on Signal Processing (ICSP), pages 1052–1055, August 2004.Google Scholar

Copyright information

© Springer Science+Business Media New York 2014

Authors and Affiliations

  • Thomas C. Henderson
    • 1
  1. 1.University of UtahSalt Lake CityUSA

Personalised recommendations