Consideration of the Word’s Neighborhood in GATs for Information Extraction in Semi-structured Documents

Autor: Yolande Belaïd, Djedjiga Belhadj, Abdel Belaïd

Publikováno v: Document Analysis and Recognition – ICDAR 2021 ISBN: 9783030863302
ICDAR (2)
Lecture Notes in Computer Science
16th International Conference on Document Analysis and Recognition ICDAR 2021
16th International Conference on Document Analysis and Recognition ICDAR 2021, Sep 2021, Lausanne, Switzerland. pp.854-869, ⟨10.1007/978-3-030-86331-9_55⟩

Most administrative documents take a semi-structured form (invoices, payslips, etc.). Extracting information from this type of document is still challenging because of the variability of its structure brought about by the change of layout style of th

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::68f18d70cf247709f7ae14ac9ce206f4
https://doi.org/10.1007/978-3-030-86331-9_55

Arabic/Latin and Machine-printed/Handwritten Word Discrimination using HOG-based Shape Descriptor

Autor: Asma Saidani, Afef Kacem, Abdel Belaïd

Publikováno v: Recercat: Dipósit de la Recerca de Catalunya
Varias* (Consorci de Biblioteques Universitáries de Catalunya, Centre de Serveis Científics i Acadèmics de Catalunya)
Electronic Letters on Computer Vision and Image Analysis
Electronic Letters on Computer Vision and Image Analysis, Computer Vision Center Press, 2015, 2 (14), pp.24. ⟨10.5565/rev/elcvia.762⟩
Dipòsit Digital de Documents de la UAB
Universitat Autònoma de Barcelona
ELCVIA: electronic letters on computer vision and image analysis; Vol. 14, Núm. 2 (2015); p. 1-23
Electronic Letters on Computer Vision and Image Analysis, 2015, 2 (14), pp.24. ⟨10.5565/rev/elcvia.762⟩
ELCVIA Electronic Letters on Computer Vision and Image Analysis, Vol 14, Iss 2 (2015)
Recercat. Dipósit de la Recerca de Catalunya
instname

International audience; In this paper, we present an approach for Arabic and Latin script and its type identification based onHistogram of Oriented Gradients (HOG) descriptors. HOGs are first applied at word level based on writingorientation analysis

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::ab2b5d3d86756da1e5e96c60aecae2bc
http://hdl.handle.net/2072/407509

Arabic Handwritten Documents Segmentation into Text-lines and Words using Deep Learning

Autor: Chemseddine Neche, Afef Kacem-Echi, Abdel Belaïd

Publikováno v: ASAR
ASAR, Sep 2019, Sydney, Australia
ASAR@ICDAR

International audience; One of the most important steps in a handwriting recognition system is text-line and word segmentation. But, this step is made difficult by the differences in handwriting styles, problems of skewness, overlapping and touching

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::7a4150032d5881361d4f873a3696a56e
https://hal.inria.fr/hal-02460880

Semi-Synthetic Data Augmentation of Scanned Historical Documents

Autor: Romain Karpinski, Abdel Belaïd

Publikováno v: ICDAR
ICDAR, Sep 2019, Sydney, Australia

International audience; This paper proposes a fully automatic new method for generating semi-synthetic images of historical documents to increase the number of training samples in small datasets. This method extracts and mixes background only images

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::4adc2ecf590112efc25611573ebfc6aa
https://hal.inria.fr/hal-02460891

Combination of Two Fully Convolutional Neural Networks for Robust Binarization

Autor: Romain Karpinski, Abdel Belaïd

Publikováno v: ACCV'18 Asian Conference on Computer Vision
ACCV'18 Asian Conference on Computer Vision, Dec 2018, PERTH, Australia. pp.509-524, ⟨10.1007/978-3-030-20893-6_32⟩
Computer Vision – ACCV 2018 ISBN: 9783030208929
ACCV (3)

To be able to process historical documents, it is often required to first binarize the image (background and foreground separation) before applying the processing itself. Historical documents are challenging to binarize because of the numerous degrad

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b037b1dddf500a275684b095bba84128
https://inria.hal.science/hal-01981602