A New Hierarchical Approach for Recognition of Unconstrained Handwritten Numerals

  title={A New Hierarchical Approach for Recognition of Unconstrained Handwritten Numerals},
  author={Jhing-Fa Wang and Gwo-En Wang},
  journal={IEEE International Conference on Consumer Electronics},
  • Jhing-Fa Wang, G. Wang
  • Published 1 August 1994
  • Computer Science
  • IEEE International Conference on Consumer Electronics
A new hierarchical approach for the recognition of unconstrained handwritten numerals is proposed. In order to obtain a reliable skeleton of the observed character, some preprocessing operations including smoothing, noise removal, normalization, and a thinning process are first applied to each character. Then, some interesting feature points are extracted from this reliable skeleton of the character. In the first stage of preclassification, a set of structural features named four-zone codes is… 

Tables from this paper

A multi-layer classifier for recognition of unconstrained handwritten numerals
  • G. Wang, Jhing-Fa Wang
  • Computer Science
    Proceedings of 3rd International Conference on Document Analysis and Recognition
  • 1995
A three layer Bayesian neural network with 20 dimensional global feature vectors is designed for fine classification of the confusing classes of handwritten numerals to promote the recognition rate to be a practical OCR system.
Disconnected handwritten numeral image recognition
  • L. Ling, N. R. Gomes
  • Computer Science
    Proceedings of the Fourth International Conference on Document Analysis and Recognition
  • 1997
Experimental results obtained from testing laboratory-prepared data and handwritten numerals extracted from real Brazilian bank checks show that recognition rates of 85% and 92.4% were achieved, respectively.
Handwritten Numeral Recognition via Fuzzy Logic and Local Discriminating Features
A system to recognize disconnected handwritten numerals based on the concept of fuzzy logic and discriminating local features extracted from numeral images and using three fuzzy sets to classify branch segments into straight line segments, parts of circles and circles is described.
Multiple classifier decision combination strategies for character recognition: A review
This paper explicitly reviews the field of multiple classifier decision combination strategies for character recognition, from some of its early roots to the present day and illustrates explicitly how the principles underlying the application of multi-classifier approaches to character recognition can easily generalise to a wide variety of different task domains.
Virtual Handwritten Recognition for Advanced Smart Learning Process
The proposed system can improve the standard of smart learning system by improving the way of teaching system by recognizing numeral or character input virtually using Support Vector Machines.
A Prototype for Brazilian Bankcheck Recognition
The proposed automatic bankcheck recognition prototype was intensively tested by real bankcheck data as well as simulated data providing the following performance results: for skilled forgery, 4.7% equal error rate; for random forgeries, zero Type I error and 7.3% Type II error; for bankcheck numerals, 92.7%" correct recognition rate.
Distinctive feature analysis for Thai handwritten character recognition based on modified stroke changing sequence
From character head center, stroke changing sequence is drawn both horizontally and vertically to detect stroke change to improve recognition rate and speed.


A new approach for recognition of unconstrained handwritten numerals
  • G. Wang, Jhing-Fa Wang
  • Computer Science
    Proceedings of TENCON '93. IEEE Region 10 International Conference on Computers, Communications and Automation
  • 1993
A new approach for recognizing unconstrained handwritten numeral is proposed, using smoothing and coloring processes to smooth the shape and to obtain the number of components of the observed numeral.
Computer Recognition of Handwritten Numerals by Polygonal Approximations
The outlines of handwritten numerals are approximated by polygons using a method previously developed by Pavlidis and Horowitz [10]. This enables a simple evaluation of many intuitively descriptive
Computer Recognition of Totally unconstrained Handwritten ZIP Codes
  • P. Ahmed, C. Suen
  • Computer Science
    Int. J. Pattern Recognit. Artif. Intell.
  • 1987
The techniques, image processing together with recognition, described in this thesis present a complete model of an optical reader for ZIP code recognition.
A high-accuracy syntactic recognition algorithm for handwritten numerals
A new set of topological features (primitives) for use with a syntactic classifier for high-accuracy recognition of handwritten numerals is proposed and made possible with minimal preprocessing of the test pattern.