Analysis of Document Snippets as a Basis for Reconstruction

Simple item page
dc.contributor.authorDiem, Markusen_US
dc.contributor.authorKleber, Florianen_US
dc.contributor.authorSablatnig, Roberten_US
dc.contributor.editorKurt Debattista and Cinzia Perlingieri and Denis Pitzalis and Sandro Spinaen_US
dc.date.accessioned2014-01-31T15:27:44Z
dc.date.available2014-01-31T15:27:44Z
dc.date.issued2009en_US
dc.description.abstractIn Archaeography, Philology, Forensics, and related research areas fragments of documents are very common. These fragments are the basis for the subsequent reconstruction process, where the goal is to make the original information spread over several fragments visible again. The fragments can originate from paper shredders, hand torn pages or in the case of ancient manuscripts this is due to bad storage conditions, or other destroying facts. So we can distinguish between an "on-purpose" destruction because the information contained on the pages should not be readable anymore or a "time-induced" destruction for ancient documents which is unintentional. Nevertheless the reconstruction of document fragments is an interesting research question. This paper shows a preliminary step for the page reconstruction namely the automatic orientation of snippets in order to eliminate the rotation in the later reconstruction (puzzling) process. Furthermore features like paper color and the color of the inks used are analyzed as a pre-classification step to find matching snippets. In the case of "on-purpose" destruction there is no a-priori information on which fragment belongs to which page which makes a reconstruction based on thousands of fragments from unknown sources difficult since the combinatorial effort explodes (NP-hardness). Preliminary results on orientation and color segmentation are presented and show that these pre-processing steps can be performed reliably and can be used for reconstruction and snippet classification.en_US
dc.description.seriesinformationVAST: International Symposium on Virtual Reality, Archaeology and Intelligent Cultural Heritageen_US
dc.identifier.isbn978-3-905674-18-7en_US
dc.identifier.issn1811-864Xen_US
dc.identifier.urihttps://doi.org/10.2312/VAST/VAST09/101-108en_US
dc.publisherThe Eurographics Associationen_US
dc.subjectCategories and Subject Descriptors (according to ACM CCS): I.4.0 [Image Processing and Computer Vision]: General - I.5.4 [Pattern Recognition]: Applications - Text Processingen_US
dc.titleAnalysis of Document Snippets as a Basis for Reconstructionen_US
Files
Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
101-108.pdf
Size:
1.7 MB
Format:
Adobe Portable Document Format
Download
Collections
VAST09: The 10th International Symposium on Virtual Reality, Archaeology and Intelligent Cultural Heritage