Extended zone based handwritten Malayalam character recognition using structural features
P. V. Raveena, Ajay James, C. Saravanan · 2017 Second International Conference on Electrical, Computer and Communication Technologies (ICECCT) · 2017
Optical Character Recognition can be defined as the process of detecting and identifying text from a scanned image. There are a number of techniques by which recognition is carried out in several languages. The main steps of optical character recognition are Line segmentation, Word segmentation, Character segmentation and Character recognition. Character recognition has two phases: Feature extraction and classification. Difference in human handwriting style make the character recognition is difficult for a machine. Feature extraction and character recognition are different for different languages and become most complex task among the phases of OCR. Feature extraction for each language may vary depending on various characteristics of that language. The handwritten recognition for Malayalam language that propose here uses extended zone based method using structural features. The zonal features along with language dependent features are used in this work. This paper proposes an efficient method on feature extraction for Malayalam handwritten character recognition using extended zone based structural feature extraction.