Efficient Word Segmentation and Baseline Localization in Handwritten Documents Using Isothetic Covers

Efficient Word Segmentation and Baseline Localization in Handwritten Documents Using Isothetic Covers

Mousumi Dutt (Bengal Engineering and Science University, India), Aisharjya Sarkar (Bengal Engineering and Science University, India), Arindam Biswas (Bengal Engineering and Science University, India), Partha Bhowmick (Indian Institute of Technology, Kharagpur, India) and Bhargab B. Bhattacharya (Indian Statistical Institute, India)
Copyright: © 2011 |Pages: 13
DOI: 10.4018/jdls.2011070101
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

Analysis of handwritten documents is a challenging task in the modern era of document digitization. It requires efficient preprocessing which includes word segmentation and baseline detection. This paper proposes a novel approach toward word segmentation and baseline detection in a handwritten document. It is based on certain structural properties of isothetic covers tightly enclosing the words in a handwritten document. For an appropriate grid size, the isothetic covers successfully segregate the words so that each cover corresponds to a particular word. The grid size is selected by an adaptive technique that classifies the inter-cover distances into two classes in an unsupervised manner. Finally, by using a geometric heuristic with the horizontal chords of these covers, the corresponding baselines are extracted. Owing to its traversal strategy along the word boundaries in a combinatorial manner and usage of limited operations strictly in the integer domain, the method is found to be quite fast, efficient, and robust, as demonstrated by experimental results with datasets of both Bengali and English handwritings.
Article Preview

Introduction

A handwriting portrays the characteristics of an individual, and hence has been studied in numerous disciplines including experimental psychology, neuroscience, engineering, anthropology, forensic science, etc. (Plamondon, 1993; Plamondon & Leedham, 1990; Simner et al., 1994, 1996; Galen & Morasso, 1998; Galen & Stelmach, 1993; Wan et al., 1991). The analysis of handwritings has been quite important in recent times with the advancements of document digitization (Hole & Ragha, 2011; Saba, 2011; Terrades, 2010; Zhu et al., 2009), biometric authentication (Henniger & Franke, 2004; Hoque et al., 2008; Low et al., 2009; Makrushin, 2011; Schimke et al., 2005; Vielhauer, 2006; Vielhauer & Scheidat, 2005), forensic science (Franke & Köppen, 2001; Máadeed et al., 2008; Mahmoudi et al., 2009; Pervouchine et al., 2008), etc. The result of analysis strives to interpret, verify, and recognize a particular handwritten document. The most difficult problem in the area of handwriting recognition is segmentation of cursive handwriting. The infinitude of different types of human handwritings amidst the similarities in the shapes of different characters renders the problem even more difficult. Hence, over the last few years, various works have been presented for specific domains, e.g., Bengali character recognition (Majumdar & Chaudhuri, 2007; Parui et al., 2008), text line identification (Chaudhuri & Bera, 2009), numeral recognition (Bhattacharya & Chaudhuri, 2009), check sorting (Gorski et al., 1999), address reading (Srihari & Keubert, 1997), tax reading (Srihari et al., 1996), office automation (Gopisetty, 1996), automated postal system (Vajda et al., 2009), etc.

Complete Article List

Search this Journal:
Reset
Volume 5: 2 Issues (2015)
Volume 4: 2 Issues (2014)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing