Ines Ben Messaoud

New Binarization Approach Based on Text Block Extraction

Document analysis and recognition systems include, usually, several levels, annotation, preprocessing, segmentation, feature extraction, classification and post-processing. Each level may be dependent on or independent from the other levels. The presence of noise in images can affect the performance of the entire system. This noise can be introduced by the digitization step or from the document itself. In this paper, we present a new binarization approach based on a combination between a preprocessing step and a localization step. The aim of the present approach is the application of binarization algorithms on selected objects-of-interest. The evaluation of the developed approach is performed using two benchmarking datasets from the last two document binarization contests (DIBCO 2009 and H-DIBCO 2010). It shows very promising results.

A design of a preprocessing framework for large database of historical documents

The objective of document preprocessing is to ease the text recognition or the document indexing processes. The analysis of historical documents seems to be a big challenge because the majority of those documents are noisy and present many degradations. In this paper we propose a preprocessing framework for a large dataset of historical documents. The proposed framework is decomposed of two phases, the selection and the evaluation. During the first phase one or multiple methods are corresponded for each book of the used database. The validation of the selection results is performed during the evaluation. The experiments are applied on printed and handwritten documents extracted respectively from Google-Books and Bayerische Staatsbibliothek databases. The results returned during the evaluation are very promising.

New method for the selection of binarization parameters based on noise features of historical documents

Historical documents contain generally different kind of degradations. Due to this degradations the application of methods of noise removal during a preprocessing stage seems to be necessary. Since the noise which, exists in the original document can not be eliminated using a simple noise removal algorithm and it influences the preprocessing result e.g. the binarization, a function of noise detection seems to be necessary. We present in this paper a method for the selection of the input parameters of binarization methods according to the noise type detected in the image. The tests are achieved on benchmarking datasets used at DIBCO 2009 and H-DIBCO 2010. The results returned by the binarization methods using the noise features are promising.

A Multilevel Text-Line Segmentation Framework for Handwritten Historical Documents

Text-line segmentation is considered as a crucial step of document analysis and recognition systems because its output is considered as the input of recognition systems. Due to the reason that the same handwritten image page has different characteristics, we propose in this paper a multilevel segmentation framework for handwritten historical documents. In this framework, one or many segmentation methods are selected according to the input document features. This framework is tested on the IAM historical database (60 images) and on images from the segmentation competition for handwritten document segmentation held at ICFHR 2010. The evaluation of the segmentation framework is based on several evaluation metrics. The tests show that the proposed framework gives promoting results.

Automatic Annotation for Handwritten Historical Documents Using Markov Models

This paper presents a system for automatic annotation of handwritten historical documents based on Markov models. The proposed system first extracts XML schema which describes a specific domain and than a Mapping algorithm is used for the generation of the new XML schemes. Mapping algorithm has as inputs two schemes reference schema and a specific schema. XML schemes are generated using Markov models, this model is used to calculate the Mapping efficiency. In the first model the Mapping increased according to the common number of nodes between the entries XML schemes. Mapping is pertinent when the common nodes number is over

Document Preprocessing System - Automatic Selection of Binarization

0.5\%

Region Based Local Binarization Approach for Handwritten Ancient Documents

of Markov model states. In the second model the Mapping changes randomly according to the in common number of nodes between

A model for the gray-intensity distribution of historical handwritten documents and its application for binarization

0.05

Binarization effects on results of text-line segmentation methods applied on historical documents

and

New evaluation framework for metadata mapping approaches based on Markov models

0.4