A Novel Two Stage Evaluation Methodology for Word Segmentation Techniques

Abstract

Word segmentation is a critical stage towards word and character recognition as well as word spotting and mainly concerns two basic aspects, distance computation and gap classification. In this paper, we propose a robust evaluation methodology that treats the distance computation and the gap classification stages independently. The detection rate calculated for every distance metric corresponds to the maximum detection rate that we could have achieved if we had a perfect classifier for the gap classification stage. The proposed evaluation framework has been applied to several state-of-the-art techniques using a handwritten as well as a historical typewritten document set. The best combination of distance metric computation and gap classification state-of-the-art techniques is proposed.

A Survey on Methods for Basic Unit Segmentation in Off-line Handwritten Text Recognition

“…[24] In 1994, a comprehensive discussion on word separation have been carried out by Seni and Cohen etc [7]. Eight different distance metrics have been analyzed, and these metrics are used to correct word segmentation results by measuring the number of text lines.…”

“…In 2009, a robust evaluation method that was independent in the distance computation and the gap classification stages proposed by Louloudis and Stamatopoulos [24]. In 2009 ，two effective techniques for segmenting handwritten documents into text lines and words were presented by Papavassiliou et al [26].…”

Morphological Operations and Projection Profiles based Segmentation of Handwritten Kannada Document

“…The authors have used the concept of variable sized window, that is, the window whose size can be adjusted according to needs. In [18], the authors have proposed a novel two stage evaluation methodology for word segmentation Techniques. They have proposed a robust evaluation methodology that treats the distance computation and the gap classification stages independently.…”

Fast document image comparison in multilingual corpus without OCR

“…The second category is recognition-based segmentation [16] which is used to search the image for components. The third is holistic methods [17] which seek to recognize words as a whole and avoid the need of segmentation.…”