Academic Journals Database
Disseminating quality controlled scientific knowledge

Text line and word segmentation of Indian Script Handwritten Document

Author(s): Varsha Hole | Leena Ragha | Pravin Hole

Journal: International Journal of Computer Applications
ISSN 0975-8887

Volume: icwet;
Issue: 3;
Date: 2012;
Original page

Keywords: Optical character recognition | Pre-processing | Global skew detection and correction | Line segmentation | Word segmentation

Based on the analysis of Indian script character shapes and literature survey, it presents a new sequence of line and word segmentation method to handle some of the deformations usually present in the handwritten document like touching components, overlapping components, skewed lines, words with individual skews etc. and build a proper text image with all these deformations removed. Line segmentation procedure is applied using Hough transform. The word segmentation is done with the computation of the distances of adjacent components in the text line image and classification of the previously computed distances as either inter-word gaps or inter-character gaps in a Gaussian mixture modeling framework. The proposed method of line segmentation is a sufficiently accurate to extract the text lines from unconstrained handwritten text documents. Word segmentation procedure also works well on different language scripts. Average result of word segmentation for complex Document on different language script is 76% and average result of word segmentation for good Document of different language script is 90%.
Save time & money - Smart Internet Solutions     

Tango Jona
Tangokurs Rapperswil-Jona