Comprehensive Study on Recognition of offline Handwritten Gujarati Numerals

Abstract

Last few decades, Handwritten Character Recognition (HCR) is one of the most intricate and tricky area in the field of pattern recognition. Research on Optical Character Recognition (OCR) of Gujarati script is very difficult and challenging task due to its complex structural attributes. The most important difficulty with handwritten text is the variability of writing styles, between different writers and between separate examples from the same writer overtime. There is no standard dataset available for handwritten Gujarati characters. Researcher has to develop their own character dataset. Handwritten character or Numeral recognition has received large attention in academics and production fields. This paper will act as guide and update for the readers, working in the field of OCR of Gujarati language. An overview of various statistical and structural features used for recognition based on the various research papers is presented and reviewed.

Keywords: Gujarati Numeral Recognition, Off-line Handwriting Recognition, Feature Extraction, Classification

View PDF