Hidden Loop Recovery for Handwriting Recognition
.
Hidden loop recovery for handwriting recognition.
In FHR02, 375-380, 2002
Online Version
A pdf version is available for download.
Abstract
One significant challenge in the recognition of offline handwriting is in the interpretation of loop structures. Although this information is readily available in online representation, close proximity of strokes often merges their centers making them difficult to identify. In this paper a novel approach to the recovery of hidden loops in offline scanned document images is presented. The proposed algorithm seeks blobs that resemble truncated ellipses. We use a sophisticated form analysis method based on mutual distance measurements between the two sides of a symmetric shape. The experimental results are compared with the ground truth of the online representations of each offline word image. More than 86% percent of the meaningful loops are handled correctly.
Co-authors
Bibtex Entry
@inproceedings{DoermannIRS02i,
title = {Hidden loop recovery for handwriting recognition},
author = {David Doermann and Nathan Intrator and Ehud Rivlin and Tal Steinherz},
year = {2002},
booktitle = {FHR02},
pages = {375-380},
abstract = {One significant challenge in the recognition of offline handwriting is in the interpretation of loop structures. Although this information is readily available in online representation, close proximity of strokes often merges their centers making them difficult to identify. In this paper a novel approach to the recovery of hidden loops in offline scanned document images is presented. The proposed algorithm seeks blobs that resemble truncated ellipses. We use a sophisticated form analysis method based on mutual distance measurements between the two sides of a symmetric shape. The experimental results are compared with the ground truth of the online representations of each offline word image. More than 86% percent of the meaningful loops are handled correctly.}
}