6,462 Hits in 6.4 sec

A Complete Pyramidal Geometrical Scheme for Text Based Image Description and Retrieval [chapter]

Guillaume Joutel, Véronique Eglin, Hubert Emptoz
2008 Lecture Notes in Computer Science  
This paper presents a general architecture for ancient handwriting documents content description and retrieval.  ...  The complete implementation scheme is validated with a content based images retrieval (CBIR) application on the medieval database from the IRHT 1 and on the European 18 th century correspondences corpus  ...  The two other on-line applications that are also based on the Curvelet geometric decomposition (spotting and writing styles clustering) are currently under development.  ... 
doi:10.1007/978-3-540-69905-7_54 fatcat:mgpnplgwxzh3bnmientwekcopm

Curvelets Based Queries for CBIR Application in Handwriting Collections

G. Joutel, V. Eglin, S. Bres, H. Emptoz
2007 Proceedings of the International Conference on Document Analysis and Recognition  
Current results of ancient manuscripts retrieval are very promising with very satisfying precisions and recalls.  ...  In this way, Curvelets coefficients are used as a representation tool for handwriting when searching in large manuscripts databases by finding similar handwritten samples.  ...  Acknowledgments: We want to thank the palaeographer Denis Muzerelle from the IRHT of Paris for his advices for medieval handwritings classification and also Catherine Volpilhac-Auger from the CERPI of  ... 
doi:10.1109/icdar.2007.4376995 dblp:conf/icdar/JoutelEBE07 fatcat:bzvuox3qefeern5624vgnka2gi

Word segmentation of off-line handwritten documents

Chen Huang, Sargur N. Srihari, Berrin A. Yanikoglu, Kathrin Berkner
2008 Document Recognition and Retrieval XV  
This paper describes an approach to separate a line of unconstrained (written in a natural manner) handwritten text into words.  ...  The system is evaluated using an unconstrained handwriting database, which contains 50 pages (1026 line, 7562 words images) handwritten documents.  ...  It is based on modeling the lines as bi-variate gaussian densities that provide for accurate association of components to the respective lines.  ... 
doi:10.1117/12.767055 dblp:conf/drr/HuangS08 fatcat:4tp4te3b2rc7re6k2kkke6kzcy

Curvelets based feature extraction of handwritten shapes for ancient manuscripts classification

Guillaume Joutel, Véronique Eglin, Stéphane Bres, Hubert Emptoz, Xiaofan Lin, Berrin A. Yanikoglu
2007 Document Recognition and Retrieval XIV  
The main experimental results are very promising and show 78% of good retrieval (as precision) on the Middle-Ages database and 89% on the humanistic database.  ...  We propose a global approach of writers' classification based on Curvelets based features in relation with two discriminative shapes properties, the curvature and the orientation.  ...  Acknowledgments We want to thank the palaeographer Denis Muzerelle from the IRHT of Paris for his advices and recommendations for medieval handwritings classification and also Catherine Volpilhac-Auger  ... 
doi:10.1117/12.704450 dblp:conf/drr/JoutelEBE07 fatcat:ifftqvsqffch7oe4piriha3zcm

Algorithm based on normal coordinate vectors with 16 segments for the data fusion from hand-written Arabic text implemented with MATLAB

Said S. Saloum, Iván García-Magariño
2021 PeerJ Computer Science  
This work presents a novel handwriting recognition approach based on the application of coordinate vectors to find similarities in different kinds of deformations.  ...  In order to perform data fusion from different data sources, handwriting automatic recognition involves barriers such as different ways of writing letters and deformation due to many reasons.  ...  This ranking is based on the fact that information retrieval can be improved by applying data fusion techniques in handwriting documents.  ... 
doi:10.7717/peerj-cs.705 pmid:34604527 pmcid:PMC8444069 fatcat:tnest5tj2jbb3lwlp2l2txwr3e

HAH manuscripts: A holistic paradigm for classifying and retrieving historical Arabic handwritten documents

Zaher Al Aghbari, Salama Brook
2009 Expert systems with applications  
Second, several features are extracted from these connected parts and then combined to represent a word with one consolidated feature vector.  ...  Finally, a generalized feedforward neural network is used to learn and classify the different styles/fonts into word classes, which are used to retrieve Arabic handwritten text documents.  ...  Then, these feature vectors are combined to represent the corresponding word with one consolidated feature vector.  ... 
doi:10.1016/j.eswa.2009.02.024 fatcat:wyyleohskjbpdndurdxavibyja

Retrieval of online handwriting by synthesis and matching

C.V. Jawahar, A. Balasubramanian, Million Meshesha, Anoop M. Namboodiri
2009 Pattern Recognition  
One of the major advantages of this framework is that information retrieval techniques such as ranking relevance, detecting stopwords and controlling word forms are extended to work with search and retrieval  ...  The proposed approach provides a keyboard-based search interface that enables to search handwritten data from any platform, in addition to pen-based and example-based queries.  ...  The following are the characteristic differences of the current work, when compared to other approaches to retrieval of online handwriting data. (1) Use of Synthesis: Our approach employs a handwriting  ... 
doi:10.1016/j.patcog.2008.08.017 fatcat:krfrckza5zg4pbvws3pvy3abuq

Generic scale-space process for handwriting documents analysis

Guillaume Joutel, Veronique Eglin, Hubert Emptoz
2008 Pattern Recognition (ICPR), Proceedings of the International Conference on  
In that context, no lexical information nor any other statistical language models are required.  ...  It covers all analysis steps from the content description of the document (layout analysis, handwriting shape characterization) to three dedicated Digital Libraries applications (CBIR 1 in great ancient  ...  A mixed approach that combines global and local shapes description seemed to be an interesting way for handwriting images analysis.  ... 
doi:10.1109/icpr.2008.4761827 dblp:conf/icpr/JoutelEE08 fatcat:hmilm7hxx5bv5fm5rzs46chaha

Machine learning in a multimedia document retrieval framework

M. P. Perrone, G. F. Russell, A. Ziq
2002 IBM Systems Journal  
Croft, “A Language Model Approach to In- formation Retrieval,” Proceedings of the 21st Annual Inter- national ACM SIGIR Conference on Research and Develop- ment in Information Retrieval (August 1998),  ...  The data set size in this experiment is much smaller than typical databases used in text retrieval research, though it is typical of available on-line handwriting recognition databases.  ... 
doi:10.1147/sj.413.0494 fatcat:egp4a3gh4bdsxpw7pip7ulwqua

Using content based image retrieval techniques for the indexing and retrieval of Thai handwritten documents

Seksan Sangsawad, Chun Che Fung
2010 2010 The 2nd International Conference on Computer and Automation Engineering (ICCAE)  
This paper also examined off-line signature recognition techniques in order to adapt to Thai handwriting system for matching data.  ...  The objective of the proposal is to develop a semiautomated method to index and retrieve Thai handwritten documents based on sampled keywords by combining CBIR and signature recognition techniques.  ...  In terms of document management, the key of information retrieval is the ability to extract and match some desired data. Several tools have emphasized on line and word segmentation.  ... 
doi:10.1109/iccae.2010.5451992 fatcat:fl4pvxasorhgpkufiibu2zubje

Boosted decision trees for word recognition in handwritten document retrieval

Nicholas R. Howe, Toni M. Rath, R. Manmatha
2005 Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval - SIGIR '05  
We propose a novel approach to recognizing and retrieving handwritten manuscripts, based upon word image classification as a key step.  ...  Retrieval performance also shows substantially improved results over previously published results on this database.  ...  (The on-line handwriting problem is easier because of access to information about pen velocity and position.)  ... 
doi:10.1145/1076034.1076099 dblp:conf/sigir/HoweRM05 fatcat:yswgefl7dfco5nbmshlyd3zfka

Text-Independent Writer Identification and Verification Using Textural and Allographic Features

Marius Bulacu, Lambert Schomaker
2007 IEEE Transactions on Pattern Analysis and Machine Intelligence  
At the texture level, we use contour-based joint directional PDFs that encode orientation and curvature information to give an intimate characterization of individual handwriting style.  ...  The proposed methods are applicable to free-style handwriting (both cursive and isolated) and have practical feasibility, under the assumption that a few text lines of handwritten material are available  ...  Writer identification is performed in an information retrieval framework, while writer verification is based on the mutual information between the grapheme distributions in the two handwritings that are  ... 
doi:10.1109/tpami.2007.1009 pmid:17299226 fatcat:fkgu47xnqbcjjaffyl2ddi26gq

Generic Scale-Space Architecture for Handwriting Documents Analysis [chapter]

Joutel Guillaume, Eglin Veronique, Emptoz Hubert
2010 Pattern Recognition Recent Advances  
The content based information retrieval system is based on a global information about the handwriting in the document and the word spotting system is based on local information about a keyword.  ...  retrieve information from manuscripts.  ...  /pattern-recognition-recentadvances/generic-scale-space-architecture-for-handwriting-documents-analysis  ... 
doi:10.5772/9358 fatcat:nky4w4k4orbn7os23qjoybxyem

Handwritten Text Image Authentication using Back Propagation [article]

A. S. N. Chakravarthy, Penmetsa V. Krishna Raja, P. S. Avadhani
2011 arXiv   pre-print
is a trusted one.  ...  The authentication of information can pose special problems (especially man-in-the-middle attacks), and is often wrapped up with authenticating identity.  ...  ACKNOWLEDGEMENTS An assemblage of this nature could never have been attempted without reference to and inspiration from the works of others whose details are mentioned in reference section.  ... 
arXiv:1110.1488v1 fatcat:yogtbzyt3refjbj4slmkbhwoti

Biological inspired tools for patrimonial handwriting denoising and categorization

V. Eglin, S. Bres, C. Rivero, H. Emptoz
2005 Eighth International Conference on Document Analysis and Recognition (ICDAR'05)  
In this paper, we propose a global segmentation free methodology for patrimonial documents denoising, handwriting characterization and categorization that are based on biological inspired approaches.  ...  The current results of handwriting categorization and indexing are very promising and show that it is possible to analyze handwritten drawings without any a priori graphemes segmentation.  ...  The originality of our approach is to consider a multiscale approach from a macroscopic (linked to the page dimension and the global text lines orientations) to a microscopic (linked to the grapheme or  ... 
doi:10.1109/icdar.2005.58 dblp:conf/icdar/EglinBRE05 fatcat:7iywzptqr5fx3kjltgzoh4gzdu
« Previous Showing results 1 — 15 out of 6,462 results