
<?xml version="1.0" encoding="UTF-8"?>
<record>
  <title>An Analysis of Optical Character Recognition (OCR) Methods</title>
  <journal>International Journal of Computational Linguistics Research</journal>
  <author>Nabeel Ashraf, Syed Yasser Arafat, Muhammad Javed Iqbal</author>
  <volume>10</volume>
  <issue>3</issue>
  <year>2019</year>
  <doi>https://doi.org/10.6025/jcl/2019/10/3/81-91</doi>
  <url>http://www.dline.info/jcl/fulltext/v10n3/jclv10n3_3.pdf</url>
  <abstract>This survey paper presents a comprehensive study of Urdu Optical Character Recognition (OCR) methodologies.
The main focus of the study is detail investigation of the techniques used to recognize the Nastaliq, Naskh and other similar
scripts fonts. These script fonts are used to write Urdu, Arabic, Pashto and Sindhi etc. languages. Several methods of text
recognition and classification of Urdu like cursive scripts are discussed. The survey contains the comparison and description
of each method in a brief way which identifies handwritten, printed and online text recognition as well. For each optical
character recognition (OCR) the phases of pre-processing, segmentation, feature extraction, classification and finally
recognition are discussed. After the comprehensive analysis of all methodologies critics and future work in Urdu cursive
scripts, i.e. Naskh and Nastaliq scripts are also proposed.</abstract>
</record>
