NAVIDOMASS: Structural-based Approaches Towards Handling Historical Documents

Jouili, Salim; Coustaty, Mickaël; Tabbone, Salvatore; Ogier, Jean-Marc

doi:10.1109/icpr.2010.237

Cited by 13 publications

(16 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are a handful of other image matching techniques optimized for initial letters that we do not compare to here. For example, [15] proposes a method that works by segmenting the initial letters, using a Zipf Law distribution to label the segmented regions, measuring four features of these regions, constructing a graph of these regions and finally using (an approximation of) the graph edit distance to measure the similarity between two images in their graph representation. We do not compare to this work because there are more than eight parameters that must be set, and there is little guidance on setting them.…”

Section: Discussionmentioning

confidence: 99%

“…However, as we will show in Section 4, we can exploit an internal binarized representation to produce lower bounds to make our search algorithms significantly faster. To binarize a page P we transform the grayscale page into black or white using a simple binary threshold filter [15]. To identify the provenance of an unlabeled page, experts compare any existing initial letters, as shown in Figure 1, in the page to the reference initial letters.…”

Section: Definitions and Notationmentioning

confidence: 99%

See 1 more Smart Citation

Image Mining of Historical Manuscripts to Establish Provenance

Rakthanmanon

Campana

et al. 2012

Proceedings of the 2012 SIAM International Conference on Data Mining

View full text Add to dashboard Cite

he recent digitization of more than twenty million books has been led by initiatives from countries wishing to preserve their cultural heritage and by commercial endeavors, such as the Google Print Library Project. Within a few years a significant fraction of the world's books will be online. For millions of intact books and tens of millions of loose pages, the provenance of the manuscripts may be in doubt or completely unknown, thus denying historians an understanding of the context of the content. In some cases it may be possible for human experts to regain the provenance by examining linguistic, cultural and/or stylistic clues. However, such experts are rare and this investigation is clearly a time-consuming process. One technique used by experts to establish provenance is the examination of the ornate initial letters appearing in the questioned manuscript. By comparing the initial letters in the manuscript to annotated initial letters whose origin is known, the provenance can be determined. In this work we show for the first time that we can reproduce this ability with a computer algorithm. We leverage off a recently introduced technique to measure texture similarity and show that it can recognize initial letters with an accuracy that rivals or exceeds human performance. A brute force implementation of this measure would require several years to process a single large book; however, we introduce a novel lower bound that allows us to process the books in minutes.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Definitions and Notationmentioning

confidence: 99%

Image Mining of Historical Manuscripts to Establish Provenance

Rakthanmanon

Campana

et al. 2012

Proceedings of the 2012 SIAM International Conference on Data Mining

View full text Add to dashboard Cite

show abstract

“…In order to enrich drop caps semantically, by adding meta-data or semantic annotations, many works proposed to describe, to classify and to compare them using some statistical or structural signatures. This context therefore encompasses the ongoing development of content-based image retrieval (CBIR) systems for historical drop caps [1]. The development of such image retrieval systems is complex, since these historical drop caps present a large variety and wide range of models and styles, and because the images contain a lot of information (e.g.…”

Section: Related Workmentioning

confidence: 99%

Local Enlacement Histograms for Historical Drop Caps Style Recognition

Clément

Coustaty

Kurtz

et al. 2017

2017 14th IAPR International Conference on Document Analysis and Recognition (ICDAR)

Self Cite

View full text Add to dashboard Cite

This article focuses on the specific issue of drop caps image recognition in the context of cultural heritage preservation. Due to their heterogeneity and their weakly structured properties, these historical images represent challenging data. An important aspect in the recognition process of drop caps is their background styles, which can be considered as discriminative features to identify both the printer and the period. Most existing methods for style recognition are based on low-level features such as color or texture properties. In this article, we present a novel framework for the recognition of drop caps style based on features of higher levels. We propose to capture the spatial structure carried by these images using relative position descriptors modeling the enlacement between local cells of pixel layers obtained from a document segmentation step. Such descriptors are then exploited in an efficient bag-of-features learning procedure. Experimental results obtained on a dataset of historical drop caps images highlight the interest of this approach, and in particular the benefit of considering spatial information.

show abstract

“…It is generated from a database of lettrines images manually populated by historians using a specific thesaurus ; • In the second ontology (ontology of image processing), each lettrine image is described by a set of regions. These regions correspond to the segmentation of principal shapes in the image, using automatic image processing treatment that we have developed [4], [5]. These regions are described by statistical informations and by spatial relations between them.…”

Section: Introduction and Contextmentioning

confidence: 99%

Using Ontologies to Reduce the Semantic Gap between Historians and Image Processing Algorithms

Coustaty

Bouju

Bertet

et al. 2011

2011 International Conference on Document Analysis and Recognition

View full text Add to dashboard Cite

Abstract-To reduce the gap between pixel data and thesaurus semantics, this paper presents a novel approach using mapping between two ontologies on images of drop-capitals (also named dropcaps or lettrines): In the first ontology, each dropcap image is endowed with semantic information describing its content. It is generated from a database of lettrines images -namely Ornamental Letter Images DataBase -manually populated by historians with dropcap images annotations. For the second ontology we have developed image processing algorithms to extract image regions on the basis of a number of features. These features, as well as spatial relations, among regions form the basis of the ontology. The ontologies are then enriched by inference rules to annotate some regions to automatically deduce their semantics. In this article, the method is presented together with preliminary experimental results and an illustrative example.

show abstract

NAVIDOMASS: Structural-based Approaches Towards Handling Historical Documents

Cited by 13 publications

References 14 publications

Image Mining of Historical Manuscripts to Establish Provenance

Image Mining of Historical Manuscripts to Establish Provenance

Local Enlacement Histograms for Historical Drop Caps Style Recognition

Using Ontologies to Reduce the Semantic Gap between Historians and Image Processing Algorithms

Contact Info

Product

Resources

About