Header Information

NPRP 7 - 442 - 1 - 082
NPRP 07
Qatar University
Award Closed
01 Mar 2015
Dr. Somaya Al-Maadeed
3 Year(s)
24 Jun 2018
Resubmission of [NPRP 6 - 267 - 1 - 057]
Intelligent System to Digitally Support Paleographic Analysis of Ancient Manuscripts in Qatar

Project Summary
Domain of handwriting in the Arabic script presents unique technical challenges and has been addressed more recently than other domains. There are hundreds of thousands of rare Arabic documents and old manuscripts in libraries, museums, and private collections in the state of Qatar. A rich archival heritage of Arab world and Qatar are preserved in the Qatar National Library (QNL) and the Museum of Islamic Art (MIA) in Doha. It is estimated that about seven million rare manuscripts documenting the contributions of the Arab Islamic civilization in Astronomy, Medicine, Mathematics, Science, Engineering, Art, Religion, Philosophy, etc. still survive in libraries, museums, and private collections around the world. The goal of this research project is the development of techniques and tools for the classification, investigation, and analysis of such sources to support paleographic studies by automatically deriving the writer, date, and geographic location of the relevant scripts. Furthermore, true physical image degradation modeling, which includes stochastic processes, will be developed to model the degradation phenomena and other characteristics of these documents (e.g. creation, age). Therefore, the general objectives is to plan, analyze, design, build, and test a novel classification algorithms and tools to support Palaeographic Analysis of historical Arabic manuscripts. Considering a true fiber network for paper we will use the Boltzmann lattice method for obtaining the ink distribution. We will develop a data-driven pattern analysis method that adaptively learns enhancement and restoration models for the collection under study, with the autonomous authority to validate the adapted model with less reliance on reference data and expert feedback
This research program will lead to novel paradigms to process and understand ancient, culturally-important manuscripts to Qatar using coherent, data-driven frameworks with tractable solutions, enabling paleographs as well students to experience practical and multidisciplinary problems. Ancient manuscripts are one of heritage carriers, and are being intensively digitized all over the globe to better preserve them and to ultimately access their contents. This framework provides a novel and efficient roadmap to preserve and access ancient manuscripts, transforming the way we access and mine these documents with a data-driven and autonomous approach. The proposed tool will provide to historians and other researchers new models and approaches to address the problem of writer identification, document age and location recognition of ancient manuscripts. At the end of this project, our system will allow users to classify and authenticate ancient manuscripts of the Qatar National Library (QNL) and the Museum of Islamic Art (MIA) in Qatar. The system will use the digital representation of a manuscript as a tool to support the paleographical analysis.
Image processing; Handwriten document recognition; Documentary heritage; Document analysis; Pattern recognition
Applied research
1. Natural Sciences
1.2 Computer and Information Sciences
Computer Sciences
Yes
No
6. Humanities
6.1 History and Archaeology
History
No
Yes

Institution
Qatar University
Qatar
Submitting Institution
Northumbria University
United Kingdom
Collaborative Institution
École de Technologie Supérieure
Canada
Collaborative Institution

Personnel
Lead PI
Dr. Somaya Al-Maadeed
Qatar University
Co-Lead PI
Dr. Somaya Al-Maadeed
Qatar University
PI
Prof. Mohamed Cheriet
École de Technologie Supérieure
PI
Prof. Ahmed Bouridane
Northumbria University
PI
Dr. Sherine Al-Menshawy
Qatar University

Outputs/Outcomes
Conference Paper
Unsupervised feature selection method for improved human gait recognition
I. Rida; S. A. Maadeed; A. Bouridane
DOI:10.1109/EUSIPCO.2015.7362559
Conference Paper
Direct Unsupervised Text Line Extraction from Colored Historical Manuscript Images Using DCT
Baig, Asim, Al-Maadeed, Somaya, Bouridane, Ahmed and Cheriet, Mohamed
DOI:10.1007/978-3-319-41501-7_84
Journal Paper
Robust off-line text independent writer identification using bagged discrete cosine transform features
Faraz Ahmad Khan a , Muhammad Atif Tahir, Fouad Khelifi , Ahmed Bouridane, Resheed Almotaeryi
ISSN:09574174
Conference Paper
Offline text independent writer identification using ensemble of multi-scale local ternary pattern histograms
Khan, Faraz Ahmad, Muhammad Atif Tahir, Fouad Khelifi, and Ahmed Bouridane
DOI:10.1109/EUVIP.2016.7764587
Conference Paper
BigCrypt for big data encryption
Abdullah Al Mamun , Khaled Salah , Somaya Al-maadeed , and Tarek R. Sheltami
DOI:10.1109/SDS.2017.7939147
Online Paper
AN EVALUATION OF MA'IL QUR'AN MANUSCRIPT
Sherine El Menshawy , Sumaya Al-Maadeed and Kalthoum Adam
DOI:10.5281/zenodo.1048239
Journal Paper
Automatic segmentation and reconstruction of historical manuscripts in gradient domain
Asim Baig ; Somaya A.S. Al-Ma'adeed ; Ahmed Bouridane 2; Mohamed Cheriet
ISSN:17519667
Online Paper
Writer identification approach based on bag of words with OBI features
Amal Durou, Ibrahim Aref, Somaya Al-Maadeed, Ahmed Bouridanea, Elhadj Ben khelifa
DOI:10.1016/j.ipm.2017.09.005
Conference Paper
Measuring and optimising performance of an offline text writer identification system in terms of dimensionality reduction techniques
Amal Durou, Ibrahim Aref, Mosa Elbendak, Sumaya Al-Maadeed and Ahmed Bouridane
DOI:10.1109/EST.2017.8090393
Journal Paper
Subjective and objective quality assessment of degraded document images
Atena Shahkolaei, Hossein Ziaei Nafchi, Somaya Al-Maadeed, Mohamed Cheriet,
ISSN:12962074
Conference Paper
Letter-based classification of Arabic scripts style in ancient Arabic manuscripts: Preliminary results
Kalthoum Adam, Somaya Al-Máadeed, Ahmed Bouridane
DOI:10.1109/ASAR.2017.8067767