An object-oriented descriptive language to facilitate advanced handwritten form processing

https://doi.org/10.1016/S0262-8856(98)00053-5Get rights and content

Abstract

This paper presents an object-oriented form description language which has been developed to enhance the performance and flexibility of advanced form reader systems. The paper also outlines the structure of a form reader which would allow the language's capabilities to be fully utilised. The form description language presented here is novel in two ways: firstly, it allows the comprehensive utilisation of contextual information; and secondly, it enables the storage of form models which economically describe the logical and layout variations in the form design which are allowed for a particular class of form.

Section snippets

Preliminary definitions

Throughout this paper, the term `object' refers to either a field or a hand-printed character on a form. `Logical information' refers to an object's logical properties (i.e. whether an object's occurrence is mandatory or optional). The term `layout information' refers to the geometric placement and size of objects.

`Form design' is the term used to describe the appearance of a form prior to any human data entry on that form. When a number of different forms are designed in order to collect

A new form description language

The language which is described here has been developed specifically to promote the generation of more flexible and robust form reader systems by supporting the comprehensive storage of contextual constraints and allowing economic storage of high-level information which describes the logical and layout variations in form design which are allowed for a particular class of form.

Application of the new language

In the situations which current form readers can handle, each meaningful word (e.g. name, county, date) of hand-printed text is expected to occur within a known, fixed, geometric area on the form. However, in the following two situations the location of words cannot be so precisely determined a priori, and a more flexible approach is required to form reading and to the form description which facilitates that reading.

  • 1.

    Where we wish to read form designs which do not enforce a totally constrained

Conclusions

An object-oriented form description language which facilitates advance handwritten form processing has been presented in this paper. The language allows logical, contextual and layout information to be recorded in a flexible and powerful manner. It has been proposed that the utilisation of such information will allow future form reader systems to demonstrate a substantial increase in flexibility and recognition performance.

Tools have been developed which are capable of reading and writing valid

References (23)

  • G. Nagy

    At the frontiers of OCR

    Proceedings of the IEEE

    (1992)
  • C.Y. Suen et al.

    Automatic recognition of handprinted characters—the state of the art

    Proceedings of the IEEE

    (1980)
  • L. Du, A.C. Downton, S.M. Lucas, B. Al-Badr, Hypothesis verification for hand-printed text by constrained lazy...
  • F. Cesarini, M. Gori, S. Marinai, G. Soda, A system for data extraction from form of known class, in: Proceedings of...
  • M.A. Eshera et al.

    An image understanding system using attributed symbolic representation and inexact graph-matching

    IEEE Transactions on Pattern Analysis and Machine Intelligence

    (1986)
  • Y. Ishitani, Model matching based on association graph for form image understanding, in: Proceedings of the IEEE...
  • S.L. Taylor, R. Fritzson, J.A. Pastor, Extraction of Data from Preprinted Forms, Document Image Analysis. IEEE Computer...
  • R.G. Casey et al.

    Intelligent forms processing

    IBM Systems Journal of Research and Development

    (1990)
  • S.W. Lam, L. Javanbakht, S.N. Srihari, Anatomy of a form reader, in: Proceedings of the IEEE International Conference...
  • Y.Y. Tang, C.Y. Suen, C. De Yan, M. Cheriet, Financial document processing based on staff line and description...
  • J. Liu, X. Ding, Y. Wu, Description and recognition of form and automated data entry, in: Proceedings of the IEEE...
  • View full text