An Ancient Graphic Documents Indexing Method Based on Spatial Similarity

Karray, Ali; Ogier, Jean-Marc; Kanoun, Slim; Alimi, Adel M.

doi:10.1007/978-3-540-88188-9_13

“…The binarization of historical documents is described also in [12]. Other papers dealing with early printed books addressed the retrieval of ornamental initials [13,14].…”

Section: Early Printed Booksmentioning

confidence: 99%

Text retrieval from early printed books

Marinai

¹

2009

Proceedings of the Third Workshop on Analytics for Noisy Unstructured Text Data

View full text Add to dashboard Cite

Retrieving text from early printed books is particularly difficult because in these documents, the words are very close one to the other and, similarly to medieval manuscripts, there is a large use of ligatures and abbreviations.To address these problems, we propose a word indexing and retrieval technique that does not require word segmentation and is tolerant to errors in character segmentation. Two main principles characterize the approach. First, characters are identified in the pages and clustered with self-organizing map (SOM). During the retrieval, the similarity of characters is estimated considering the proximity of cluster centroids in the SOM space, rather than directly comparing the character images. Second, query words are matched with the indexed sequence of characters by means of a dynamic time warping (DTW)-based approach. The proposed technique integrates the SOM similarity and the information about the width of characters in the string matching process. The best path in the DTW array is identified considering the widths of matching words with respect to the query so as to deal with broken or touching symbols. The proposed method is tested on four copies of the Gutenberg Bibles.

show abstract

“…Graph-based representations are used in many applications, for instance, to represent circuit diagrams [4], for shape recognition [8], image matching [14,2], or old document analysis [12]. Other works on graph-based representation [1,3,18,17], use different methods to incorporate features of the document image.…”

Section: Introductionmentioning

confidence: 99%

“…Bunke [4] illustrates an example of converting a circuit diagram to a graph by representing the lines in the circuit diagram; each graph node represents a line endpoint, corner or intersection point, and node attributes record the image coordinates (x,y) of this feature. In [12] the authors manipulate initial letters from old documents. They proceed by segmenting the initial letter into different information layers to obtain "Information layers of homogeneous zones".…”

Section: Introductionmentioning

confidence: 99%

Comparing Graph Similarity Measures for Graphical Recognition

Jouili

¹

,

Tabbone

²

,

Valveny

³

2010

Graphics Recognition. Achievements, Challenges, and Evolution

View full text Add to dashboard Cite

Abstract. In this paper we evaluate four graph distance measures. The analysis is performed for document retrieval tasks. For this aim, different kind of documents are used including line drawings (symbols), ancient documents (ornamental letters), shapes and trademark-logos. The experimental results show that the performance of each graph distance measure depends on the kind of data and the graph representation technique.

show abstract

Text retrieval from early printed books

Marinai

¹

2011

IJDAR

View full text Add to dashboard Cite

Retrieving text from early printed books is particularly difficult because in these documents, the words are very close one to the other and, similarly to medieval manuscripts, there is a large use of ligatures and abbreviations.To address these problems, we propose a word indexing and retrieval technique that does not require word segmentation and is tolerant to errors in character segmentation. Two main principles characterize the approach. First, characters are identified in the pages and clustered with self-organizing map (SOM). During the retrieval, the similarity of characters is estimated considering the proximity of cluster centroids in the SOM space, rather than directly comparing the character images. Second, query words are matched with the indexed sequence of characters by means of a dynamic time warping (DTW)-based approach. The proposed technique integrates the SOM similarity and the information about the width of characters in the string matching process. The best path in the DTW array is identified considering the widths of matching words with respect to the query so as to deal with broken or touching symbols. The proposed method is tested on four copies of the Gutenberg Bibles.

show abstract

An Ancient Graphic Documents Indexing Method Based on Spatial Similarity

Cited by 5 publications

References 10 publications

Text retrieval from early printed books

Text retrieval from early printed books

Comparing Graph Similarity Measures for Graphical Recognition

Text retrieval from early printed books

Contact Info

Product

Resources

About