Efficient system for Devnagari script segmentation

Character segmentation plays an important role in Optical Character Recognition (OCR) for Indian Script. The image to be segmented consists of various characters and is broken down into different segmented images of individual characters. These segmented characters are then fed to the Recognition phase of OCR. This paper covers only the segmentation of Devnagari script, as it consists of vowels, consonants and various modifiers, recognition of this script becomes quite a difficult task if segmentation is not done properly. Overall accuracy for character and word are 87.73% and 86.4% is achieved respectively.