Sourour Njah - Academia.edu (original) (raw)

Papers by Sourour Njah

arXiv (Cornell University), Apr 16, 2018

Actually, the ability to identify the documents authors provides more chances for using these doc... more Actually, the ability to identify the documents authors provides more chances for using these documents for various purposes. In this paper, we present a new effective biometric writer identification system from online handwriting. The system consists of the preprocessing and the segmentation of online handwriting into a sequence of Beta strokes in a first step. Then, from each stroke, we extract a set of static and dynamic features from new proposed model that we called Extended Beta-Elliptic model and from the Fuzzy Elementary Perceptual Codes. Next, all the segments which are composed of N consecutive strokes are categorized into groups and subgroups according to their position and their geometric characteristics. Finally, Deep Neural Network is used as classifier. Experimental results reveal that the proposed system achieves interesting results as compared to those of the existing writer identification systems on Latin and Arabic scripts.

arXiv (Cornell University), Apr 16, 2018

Actually, the ability to identify the documents' authors provides more chances for using these do... more Actually, the ability to identify the documents' authors provides more chances for using these documents for various purposes. In this paper, we present a new effective biometric writer identification system from online handwriting. The system consists of the preprocessing and the segmentation of online handwriting into a sequence of Beta strokes in a first step. Then, from each stroke, we extract a set of static and dynamic features from new proposed model that we called Extended Beta-Elliptic model and from the Fuzzy Elementary Perceptual Codes. Next, all the segments which are composed of N consecutive strokes are categorized into groups and subgroups according to their position and their geometric characteristics. Finally, Deep Neural Network is used as classifier. Experimental results reveal that the proposed system achieves interesting results as compared to those of the existing writer identification systems on Latin and Arabic scripts.

The presented code consists of the preprocessing and the segmentation of online handwriting into ... more The presented code consists of the preprocessing and the segmentation of online handwriting into a sequence of Beta strokes in a first step. Then, from each Beta stroke, we extract a set of static and dynamic features using four features extraction techniques based on the Beta-Elliptic model and the Fuzzy Elementary Perceptual Codes. Next, all the segments which are composed of N consecutive Beta strokes are categorized into groups and subgroups according to their position and their geometric characteristics. - The first features extraction technique called Beta-Elliptic model. - The second features extraction technique called Extended Beta-Elliptic Model. - The third features extraction technique using combination between Beta-Elliptic model and Fuzzy Elementary Perceptual Codes. - The fourth features extraction technique using combination between Extended Beta-Elliptic Model and Fuzzy Elementary Perceptual Codes.

This code is for the segmentation of online handwriting into a sequence of Beta strokes in a firs... more This code is for the segmentation of online handwriting into a sequence of Beta strokes in a first step. Then, from each Beta stroke, we extract a theta, angle of deviation of each stroke. Using this features, we classify each elliptic stoke into Elementary Perceptual Codes "EPCs" using Fuzzy Logic. After classifying into EPCs, we use LSTM for character recognition.

The presented code consists of the preprocessing and the segmentation of online handwriting into ... more The presented code consists of the preprocessing and the segmentation of online handwriting into a sequence of Beta strokes in a first step. Then, from each Beta stroke, we extract a set of static and dynamic features using four features extraction techniques based on the Beta-Elliptic model and the Fuzzy Elementary Perceptual Codes. Next, all the segments which are composed of N consecutive Beta strokes are categorized into groups and subgroups according to their position and their geometric characteristics. - The first features extraction technique using Beta-Elliptic model called the Advanced Overlapped Beta Strategy (AOBS). - The second features extraction technique using Beta-Elliptic model called the Simplified Beta Strategy (SBS). - The third features extraction technique using combination between Advanced Overlapped Beta Strategy and Fuzzy Elementary Perceptual Codes (AOBSFEPC). - The fourth features extraction technique using combination between Simplified Beta Strategy and...

Multimedia Tools and Applications, 2022

In this paper, we present PerTOHS theory for on-line handwriting s egmentation, based on the fact... more In this paper, we present PerTOHS theory for on-line handwriting s egmentation, based on the fact that in order to identify patterns, our human perceptual system is based on basic features called perceptual codes. Analysing handwriting, we notice the existence of elementary and global ones. Perceptual organization of elementary perceptual codes in various constraints generates global ones, and to obtain different forms of handwriting, we proceed by combining them. We develop a new approach to improve handwriting segmentation via perceptual codes. The proposed architecture uses the Beta-elliptic model for the generation of on-line handwriting scripts, the fuzzy set theory to detect the elementary perceptual codes and the genetic algorithms for the global perceptual ones. This theory has been tested on the developed MAYASTROUN database, and IRONOFF database. The achieved results show successful representations of handwritten script via perceptual codes and good segmentation rate is ob...

ArXiv, 2019

Due to the omnipresence of mobile devices, online handwritten scripts have become the most import... more Due to the omnipresence of mobile devices, online handwritten scripts have become the most important feeding input to smartphones and tablet devices. To increase online handwriting recognition performance, deeper neural networks have extensively been used. In this context, our paper handles the problem of online handwritten script recognition based on extraction features system and deep approach system for sequences classification. Many solutions have appeared in order to facilitate the recognition of handwriting. Accordingly, we used an existent method and combined with new classifiers in order to get a flexible system. Good results are achieved compared to online characters and words recognition system on Latin and Arabic scripts. The performance of our two proposed systems is assessed by using five databases. Indeed, the recognition rate exceeds 98%.

2019 International Conference on Document Analysis and Recognition Workshops (ICDARW), 2019

Handwriting recognition is challenging research filed in spite of the progress of techniques used... more Handwriting recognition is challenging research filed in spite of the progress of techniques used on its recognition. Deeper neural networks have achieved good results in this field. Current neural networks especially deep convolutional networks, neglect spatial and temporal information of script and deal only with it as an image. Features can be a crucial fact for separating between handwriting scripts. In this paper, we propose a CNN based on Beta-elliptic parameters and Fuzzy Elementary Perceptual Codes for Online Arabic Characters Recognition. Experimental results on two databases, LMCA and MAYASTROUN, indicate that our novel system based on CNN is possible on an online script and gives good accuracy of 98.90% compared to recent works in the state of the art.

Document Analysis and Recognition – ICDAR 2021 Workshops, 2021

Document Analysis and Recognition – ICDAR 2021 Workshops, 2021

Multimedia Tools and Applications, 2021

2017 1st International Workshop on Arabic Script Analysis and Recognition (ASAR), 2017

Arabic character recognition is one of the most challenging tasks and exciting areas of research.... more Arabic character recognition is one of the most challenging tasks and exciting areas of research. In this one, we will present a mobile application for handwriting recognition. Our Mobile application proposes an approach for Arabic handwriting recognition based on the fact that handwriting is defined as a sequence of elementary and perceptual codes. After applying PerTOHS theory which is a perceptual theory of on-line handwriting segmentation, each character is entirely transformed into a sequence of perceptual codes. By using decision trees and perceptual codes, we recognize Arabic characters and digits. Experimental results show that the high recognition rate depends on the way how the character is written.

Computers & Security, 2020

This is a PDF file of an article that has undergone enhancements after acceptance, such as the ad... more This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

In this paper, we present PerTOHS theory for on-line handwriting segmentation, based on the fact ... more In this paper, we present PerTOHS theory for on-line handwriting segmentation, based on the fact that in order to identify patterns, our human perceptual system is based on basic features called perceptual codes. Analysing handwriting, we notice the existence of elementary and global ones. Perceptual organization of elementary perceptual codes in various constraints generates global ones, and to obtain different forms of handwriting, we proceed by combining them. We develop a new approach to improve handwriting segmentation via perceptual codes. The proposed architecture uses the Beta-elliptic model for the generation of on-line handwriting scripts, the fuzzy set theory to detect the elementary perceptual codes and the genetic algorithms for the global perceptual ones. This theory has been tested on the developed MAYASTROUN database, and IRONOFF database. The achieved results show successful representations of handwritten script via perceptual codes and good segmentation rate is obtained.

2010 12th International Conference on Frontiers in Handwriting Recognition, 2010

... The obtained results show successful representations of Arabic handwritten script via percept... more ... The obtained results show successful representations of Arabic handwritten script via perceptual codes. ... 6], we note that there are common basic features necessary to the identification of handwriting ... system starts with the segmentation of the on-line Arabic script into elliptic ...

International Conference on Education and e-Learning Innovations, 2012

Today many students produce a wrong and illegible handwriting. The traditional approach for handw... more Today many students produce a wrong and illegible handwriting. The traditional approach for handwriting teaching needs a long hour of handwriting practice, and teacher needs a lot of time to check the handwriting errors. Unfortunately, this is not feasible in many cases. In this paper we introduce an automated educational tool for Arabic Handwriting detection errors, such as the stroke production errors, stroke sequence errors, stroke relationship errors and stroke interline errors, to help students to generate clear and readable handwriting. Firstly, we used an attributed relational graph to locate the handwriting errors. Secondly, an immediate feedback is provided to the students to correct them.

In this paper we present a new method to extract the elementary perceptual codes from on-line han... more In this paper we present a new method to extract the elementary perceptual codes from on-line handwriting scripts. This approach uses the advantages of fuzzy sets theory to classify the on-line handwriting strokes into elementary perceptual codes and the Betaelliptic model for the generation of complex handwriting movements. Human perceptual system is based on some basic features in order to read, write and recognize handwriting. These features are the elementary perceptual codes (EPC) which are on the base to identify patterns. The writing process implies the use of many constraints including the psychological and the physiological ones of the writer. This new method has been tested on the developed data base containing characters, digits and Arabic texts. The achieved results show successful representations of handwritten script with EPCs. Good recognition rates are also reached where the elementary perceptual codes are applied as the characteristics vector of an on-line handwriting Arabic recognition system.

SPIE Proceedings, 2017

The paper handles the problem of segmentation of handwriting on mobile devices. Many applications... more The paper handles the problem of segmentation of handwriting on mobile devices. Many applications have been developed in order to facilitate the recognition of handwriting and to skip the limited numbers of keys in keyboards and try to introduce a space of drawing for writing instead of using keyboards. In this one, we will present a mobile theory for the segmentation of for handwriting uses PerTOHS theory, Perceptual Theory of On line Handwriting Segmentation, where handwriting is defined as a sequence of elementary and perceptual codes. In fact, the theory analyzes the written script and tries to learn the handwriting visual codes features in order to generate new ones via the generated perceptual sequences. To get this classification we try to apply the Beta-elliptic model, fuzzy detector and also genetic algorithms in order to get the EPCs (Elementary Perceptual Codes) and GPCs (Global Perceptual Codes) that composed the script. So, we will present our Android application M-PerTOHS for segmentation of handwriting.

arXiv (Cornell University), Apr 16, 2018

Actually, the ability to identify the documents authors provides more chances for using these doc... more Actually, the ability to identify the documents authors provides more chances for using these documents for various purposes. In this paper, we present a new effective biometric writer identification system from online handwriting. The system consists of the preprocessing and the segmentation of online handwriting into a sequence of Beta strokes in a first step. Then, from each stroke, we extract a set of static and dynamic features from new proposed model that we called Extended Beta-Elliptic model and from the Fuzzy Elementary Perceptual Codes. Next, all the segments which are composed of N consecutive strokes are categorized into groups and subgroups according to their position and their geometric characteristics. Finally, Deep Neural Network is used as classifier. Experimental results reveal that the proposed system achieves interesting results as compared to those of the existing writer identification systems on Latin and Arabic scripts.

arXiv (Cornell University), Apr 16, 2018

Actually, the ability to identify the documents' authors provides more chances for using these do... more Actually, the ability to identify the documents' authors provides more chances for using these documents for various purposes. In this paper, we present a new effective biometric writer identification system from online handwriting. The system consists of the preprocessing and the segmentation of online handwriting into a sequence of Beta strokes in a first step. Then, from each stroke, we extract a set of static and dynamic features from new proposed model that we called Extended Beta-Elliptic model and from the Fuzzy Elementary Perceptual Codes. Next, all the segments which are composed of N consecutive strokes are categorized into groups and subgroups according to their position and their geometric characteristics. Finally, Deep Neural Network is used as classifier. Experimental results reveal that the proposed system achieves interesting results as compared to those of the existing writer identification systems on Latin and Arabic scripts.

The presented code consists of the preprocessing and the segmentation of online handwriting into ... more The presented code consists of the preprocessing and the segmentation of online handwriting into a sequence of Beta strokes in a first step. Then, from each Beta stroke, we extract a set of static and dynamic features using four features extraction techniques based on the Beta-Elliptic model and the Fuzzy Elementary Perceptual Codes. Next, all the segments which are composed of N consecutive Beta strokes are categorized into groups and subgroups according to their position and their geometric characteristics. - The first features extraction technique called Beta-Elliptic model. - The second features extraction technique called Extended Beta-Elliptic Model. - The third features extraction technique using combination between Beta-Elliptic model and Fuzzy Elementary Perceptual Codes. - The fourth features extraction technique using combination between Extended Beta-Elliptic Model and Fuzzy Elementary Perceptual Codes.

This code is for the segmentation of online handwriting into a sequence of Beta strokes in a firs... more This code is for the segmentation of online handwriting into a sequence of Beta strokes in a first step. Then, from each Beta stroke, we extract a theta, angle of deviation of each stroke. Using this features, we classify each elliptic stoke into Elementary Perceptual Codes "EPCs" using Fuzzy Logic. After classifying into EPCs, we use LSTM for character recognition.

The presented code consists of the preprocessing and the segmentation of online handwriting into ... more The presented code consists of the preprocessing and the segmentation of online handwriting into a sequence of Beta strokes in a first step. Then, from each Beta stroke, we extract a set of static and dynamic features using four features extraction techniques based on the Beta-Elliptic model and the Fuzzy Elementary Perceptual Codes. Next, all the segments which are composed of N consecutive Beta strokes are categorized into groups and subgroups according to their position and their geometric characteristics. - The first features extraction technique using Beta-Elliptic model called the Advanced Overlapped Beta Strategy (AOBS). - The second features extraction technique using Beta-Elliptic model called the Simplified Beta Strategy (SBS). - The third features extraction technique using combination between Advanced Overlapped Beta Strategy and Fuzzy Elementary Perceptual Codes (AOBSFEPC). - The fourth features extraction technique using combination between Simplified Beta Strategy and...

Multimedia Tools and Applications, 2022

In this paper, we present PerTOHS theory for on-line handwriting s egmentation, based on the fact... more In this paper, we present PerTOHS theory for on-line handwriting s egmentation, based on the fact that in order to identify patterns, our human perceptual system is based on basic features called perceptual codes. Analysing handwriting, we notice the existence of elementary and global ones. Perceptual organization of elementary perceptual codes in various constraints generates global ones, and to obtain different forms of handwriting, we proceed by combining them. We develop a new approach to improve handwriting segmentation via perceptual codes. The proposed architecture uses the Beta-elliptic model for the generation of on-line handwriting scripts, the fuzzy set theory to detect the elementary perceptual codes and the genetic algorithms for the global perceptual ones. This theory has been tested on the developed MAYASTROUN database, and IRONOFF database. The achieved results show successful representations of handwritten script via perceptual codes and good segmentation rate is ob...

ArXiv, 2019

Due to the omnipresence of mobile devices, online handwritten scripts have become the most import... more Due to the omnipresence of mobile devices, online handwritten scripts have become the most important feeding input to smartphones and tablet devices. To increase online handwriting recognition performance, deeper neural networks have extensively been used. In this context, our paper handles the problem of online handwritten script recognition based on extraction features system and deep approach system for sequences classification. Many solutions have appeared in order to facilitate the recognition of handwriting. Accordingly, we used an existent method and combined with new classifiers in order to get a flexible system. Good results are achieved compared to online characters and words recognition system on Latin and Arabic scripts. The performance of our two proposed systems is assessed by using five databases. Indeed, the recognition rate exceeds 98%.

2019 International Conference on Document Analysis and Recognition Workshops (ICDARW), 2019

Handwriting recognition is challenging research filed in spite of the progress of techniques used... more Handwriting recognition is challenging research filed in spite of the progress of techniques used on its recognition. Deeper neural networks have achieved good results in this field. Current neural networks especially deep convolutional networks, neglect spatial and temporal information of script and deal only with it as an image. Features can be a crucial fact for separating between handwriting scripts. In this paper, we propose a CNN based on Beta-elliptic parameters and Fuzzy Elementary Perceptual Codes for Online Arabic Characters Recognition. Experimental results on two databases, LMCA and MAYASTROUN, indicate that our novel system based on CNN is possible on an online script and gives good accuracy of 98.90% compared to recent works in the state of the art.

Document Analysis and Recognition – ICDAR 2021 Workshops, 2021

Document Analysis and Recognition – ICDAR 2021 Workshops, 2021

Multimedia Tools and Applications, 2021

2017 1st International Workshop on Arabic Script Analysis and Recognition (ASAR), 2017

Arabic character recognition is one of the most challenging tasks and exciting areas of research.... more Arabic character recognition is one of the most challenging tasks and exciting areas of research. In this one, we will present a mobile application for handwriting recognition. Our Mobile application proposes an approach for Arabic handwriting recognition based on the fact that handwriting is defined as a sequence of elementary and perceptual codes. After applying PerTOHS theory which is a perceptual theory of on-line handwriting segmentation, each character is entirely transformed into a sequence of perceptual codes. By using decision trees and perceptual codes, we recognize Arabic characters and digits. Experimental results show that the high recognition rate depends on the way how the character is written.

Computers & Security, 2020

This is a PDF file of an article that has undergone enhancements after acceptance, such as the ad... more This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

In this paper, we present PerTOHS theory for on-line handwriting segmentation, based on the fact ... more In this paper, we present PerTOHS theory for on-line handwriting segmentation, based on the fact that in order to identify patterns, our human perceptual system is based on basic features called perceptual codes. Analysing handwriting, we notice the existence of elementary and global ones. Perceptual organization of elementary perceptual codes in various constraints generates global ones, and to obtain different forms of handwriting, we proceed by combining them. We develop a new approach to improve handwriting segmentation via perceptual codes. The proposed architecture uses the Beta-elliptic model for the generation of on-line handwriting scripts, the fuzzy set theory to detect the elementary perceptual codes and the genetic algorithms for the global perceptual ones. This theory has been tested on the developed MAYASTROUN database, and IRONOFF database. The achieved results show successful representations of handwritten script via perceptual codes and good segmentation rate is obtained.

2010 12th International Conference on Frontiers in Handwriting Recognition, 2010

... The obtained results show successful representations of Arabic handwritten script via percept... more ... The obtained results show successful representations of Arabic handwritten script via perceptual codes. ... 6], we note that there are common basic features necessary to the identification of handwriting ... system starts with the segmentation of the on-line Arabic script into elliptic ...

International Conference on Education and e-Learning Innovations, 2012

Today many students produce a wrong and illegible handwriting. The traditional approach for handw... more Today many students produce a wrong and illegible handwriting. The traditional approach for handwriting teaching needs a long hour of handwriting practice, and teacher needs a lot of time to check the handwriting errors. Unfortunately, this is not feasible in many cases. In this paper we introduce an automated educational tool for Arabic Handwriting detection errors, such as the stroke production errors, stroke sequence errors, stroke relationship errors and stroke interline errors, to help students to generate clear and readable handwriting. Firstly, we used an attributed relational graph to locate the handwriting errors. Secondly, an immediate feedback is provided to the students to correct them.

In this paper we present a new method to extract the elementary perceptual codes from on-line han... more In this paper we present a new method to extract the elementary perceptual codes from on-line handwriting scripts. This approach uses the advantages of fuzzy sets theory to classify the on-line handwriting strokes into elementary perceptual codes and the Betaelliptic model for the generation of complex handwriting movements. Human perceptual system is based on some basic features in order to read, write and recognize handwriting. These features are the elementary perceptual codes (EPC) which are on the base to identify patterns. The writing process implies the use of many constraints including the psychological and the physiological ones of the writer. This new method has been tested on the developed data base containing characters, digits and Arabic texts. The achieved results show successful representations of handwritten script with EPCs. Good recognition rates are also reached where the elementary perceptual codes are applied as the characteristics vector of an on-line handwriting Arabic recognition system.

SPIE Proceedings, 2017

The paper handles the problem of segmentation of handwriting on mobile devices. Many applications... more The paper handles the problem of segmentation of handwriting on mobile devices. Many applications have been developed in order to facilitate the recognition of handwriting and to skip the limited numbers of keys in keyboards and try to introduce a space of drawing for writing instead of using keyboards. In this one, we will present a mobile theory for the segmentation of for handwriting uses PerTOHS theory, Perceptual Theory of On line Handwriting Segmentation, where handwriting is defined as a sequence of elementary and perceptual codes. In fact, the theory analyzes the written script and tries to learn the handwriting visual codes features in order to generate new ones via the generated perceptual sequences. To get this classification we try to apply the Beta-elliptic model, fuzzy detector and also genetic algorithms in order to get the EPCs (Elementary Perceptual Codes) and GPCs (Global Perceptual Codes) that composed the script. So, we will present our Android application M-PerTOHS for segmentation of handwriting.