Elsevier

Pattern Recognition

Volume 27, Issue 7, July 1994, Pages 895-902
Pattern Recognition

Nonlinear shape normalization methods for the recognition of large-set handwritten characters

https://doi.org/10.1016/0031-3203(94)90155-4Get rights and content

Abstract

Recently, several nonlinear shape normalization methods have been proposed in order to compensate for shape distortions in large-set handwritten characters. In this paper, these methods are reviewed from the two points of view: feature projection and feature density equalization. The former makes feature projection histogram by projecting a certain feature at each point onto horizontal- or vertical-axis and the latter equalizes the feature densities of input image by re-sampling the feature projection histogram. Then, the results of quantitative evaluation for these methods are presented. These methods have been implemented on a PC in C language and tested with a large variety of handwritten Hangul syllables. A systematic comparison of them has been made based on the following criteria: recognition rate, processing speed, computational complexity and degree of variation.

References (10)

There are more references available in the full text version of this article.

Cited by (83)

  • A prototype classification method and its use in a hybrid solution for multiclass pattern recognition

    2006, Pattern Recognition
    Citation Excerpt :

    The second group consists of two large-scale classification tasks, in which we use full sets of ETL8B (comprised of 956 character types) and ETL9B (comprised of 3036 character types). For all data sets, we use a feature extraction method that consists of three basic techniques [9,19,20]: non-linear normalization [21,22], directional feature extraction [9,20], and feature blurring [23]. According to Umeda [24], these techniques represent major breakthroughs in handwritten Chinese/Hiragana character recognition.

  • Pseudo two-dimensional shape normalization methods for handwritten Chinese character recognition

    2005, Pattern Recognition
    Citation Excerpt :

    The moment normalization method of Casey [1] and the geometric projection method of Nagy and Tuong [2] were proposed for normalizing alphanumeric characters. For normalizing Chinese characters that contain multiple strokes, the nonlinear normalization (NLN) method based on line density equalization is proven very efficient [3]. The line density can be computed in various ways yet the methods of Tsukumo and Tanaka [4] and Yamada et al. [5] have been widely adopted.

  • An improved handwritten Chinese character recognition system using support vector machine

    2005, Pattern Recognition Letters
    Citation Excerpt :

    Lee and Park (1994) compared the performance and computational complexity of existing nonlinear normalization methods and experimental results indicated that Yamada et al.’s method based on line density (Yamada et al., 1990) performed better while having a higher computational cost and side effects of zigzags.

  • Benchmark dataset for offline handwritten character recognition

    2017, Proceedings - 2017 13th International Conference on Emerging Technologies, ICET2017
View all citing articles on Scopus

A preliminary version of this paper has been presented at the 2nd International Conference on Document Analysis and Recognition, Tsukuba Science City, Japan, October 1993.

View full text