Thai handwritten character corpus

Abstract

The paper describes a Thai handwritten character corpus, developed in Thailand by the National Electronics and Computer Technology Center (NECTEC). The corpus consists of online and offline handwritten characters, including isolated, touching, and cursive handwritten characters. Total numbers of images are more than 44,000. The corpus has two purposes. One is to reduce the budget and time consumption of research work. The other is to provide a standard corpus for comparing the performance between handwritten character recognition engines.

Showing 1-4 of 4 references

Sang-Tae Park Em-lung Kim Sang-Hwn Paek and Sung-Yang Bang Handwritten Korean Character Image Database PE9Z

  • Dae-Hwan Kim Young-Sup Hwang
  • 1996

Development of Very Large Corpora in Thailand Collection. and Analysis of Handwriting Sample Image Database

  • Rachad Thongprasirr, Thatsanee Charenpom, Wain Sinthupinya, Virach Sortlertlamvanich
  • 1994

CEDAR On-line Handwriting Dabbase'', HlUllUl Language Technology

Daw Sets for OCR and Document Image Understanding Research Handbook of Character Recapition and Document Image Analysis

  • Isabelle Guyon, Robert M Haralick, Jonathan I Hull, Ihsi Tsaipn Pbihps