Enchanced Automatic Offline Character Image Pre-processing and Recogintion Using Single Layer Network

R. Radha, R.R. Aparna


In this paper the pre-processing, feature extraction phases, classification and recognition of an offline handwritten numeric character are discussed.. This paper discusses in detail some of the algorithms used in the pre-processing stages of an offline handwritten numeric character in the form of an image file and digit recognition using artificial neural network. This paper serves as the beginning part of a future research work that aims at recognizing handwritten isolated and cursive alphabets.. The pre-processing phase starts from reading in the input file, the process of binarization, filtering, edge enhancement, segmentation and feature extraction of the character image for further use in the next stage of labeling and recognition process by neural network. The handwritten digits considered here are non-slant characters with less noise but feature varies from writer to writer


Keywords: Pre-processing; feature extraction; chain-code;binarization; handwritten character;unsharp masking, single- layer neural network .

Full Text:


DOI: https://doi.org/10.26483/ijarcs.v3i4.1277


  • There are currently no refbacks.

Copyright (c) 2016 International Journal of Advanced Research in Computer Science