Feature Extraction Techniques for Handwritten Text in Various Scripts: a Survey
Vandita Singh1, Bhupendra Kumar2, Tushar Patnaik3
1Vandita Singh, C-DAC-Noida, Noida, India.
2Bhupendra Kumar, C-DAC-Noida, Noida, India.
3Tushar Patnaik, C-DAC-Noida, Noida, India.
Manuscript received on February 04, 2013. | Revised Manuscript received on February 27, 2013. | Manuscript published on March 05, 2013. | PP: 238-241 | Volume-3 Issue-1, March 2013. | Retrieval Number: A1361033113/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Published By: Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Optical Character Recognition (OCR) Systems aim to recognize text and bring it to editable form from the given document image, where the input text can be in machine printed, hand written or hand printed form. Many recognition systems have been developed for languages based on various scripts and digits all over the world, taking input in either of the online and offline modes, with considerable efficiencies. These systems have proved to be highly applicable in the fields of Banking, Education, IT systems and Postal Sector for digitization of processes and automated information retrieval. In this paper, we present a survey of techniques for recognition of handwritten and hand printed documents in off-line mode, with an emphasis on the Feature Extraction phase and the corresponding classification technique has also been mentioned with the recognition rates achieved.
Keywords: Optical Character Recognition, Feature Extraction, Classification.