Structured document labeling and rule extraction using a new recurrent fuzzy-neural system

Sainz-Palmero, G.I.; Dimitriadis, Yannis

doi:10.1109/icdar.1999.791754

Cited by 21 publications

(14 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Then, our recognition rate correspond to the number of black pixels that has been properly labeled. We can notice that this level of precision is necessary because, contrary to other method like, 4 we have to evaluate classification but also segmentation into blocks.…”

Section: Results On a Large Basementioning

confidence: 99%

See 1 more Smart Citation

<title>A generic method for structure recognition of handwritten mail documents</title>

2008

View full text Add to dashboard Cite

This paper presents a system to extract the logical structure of handwritten mail documents. It consists in two joined tasks: the segmentation of documents into blocks and the labeling of such blocks. The main considered label classes are: addressee details, sender details, date, subject, text body, signature. This work has to face with difficulties of unconstrained handwritten documents: variable structure and writing.We propose a method based on a geometric analysis of the arrangement of elements in the document. We give a description of the document using a two-dimension grammatical formalism, which makes it possible to easily introduce knowledge on mail into a generic parser. Our grammatical parser is LL(k), which means several combinations are tried before extracting the good one. The main interest of this approach is that we can deal with low structured documents. Moreover, as the segmentation into blocks often depends on the associated classes, our method is able to retry a different segmentation until labeling succeeds.We validated this method in the context of the French national project RIMES, which proposed a contest on a large base of documents. We obtain a recognition rate of 91.7% on 1150 images.

show abstract

Section: Results On a Large Basementioning

confidence: 99%

“…Thus, Sainz Palmero et al 4 propose a complete analysis of mail documents based on a recurrent neuro-fuzzy architecture. To classify blocks, they use structural and geometric data, such as x-y positions, number of lines, and contextual data like the relative positions of blocks.…”

mentioning

confidence: 99%

<title>A generic method for structure recognition of handwritten mail documents</title>

2008

View full text Add to dashboard Cite

show abstract

“…The use of learning modules leads to more adaptable systems as those presented by Sainz and Dimitriadis [Palmero and Dimitriadis, 1999] and Li and Ng [Li and Ng, 1999]. These latter two systems ignore the important role of textual content.…”

Section: Related Workmentioning

confidence: 99%

Textual Article Clustering in Newspaper Pages

Aiello

Pegoretti

2006

Applied Artificial Intelligence

View full text Add to dashboard Cite

In the analysis of a newspaper page an important step is the clustering of various text blocks into logical units, i.e., into articles. We propose three algorithms based on text processing techniques to cluster articles in newspaper pages. Based on the complexity of the three algorithms and experiment on actual pages from the Italian newspaper L'Adige, we select one of the algorithms as the preferred choice to solve the textual clustering problem.

show abstract

“…Such systems are presented by Sainz and Dimitriadis [21] and Li and Ng [15]. Again, their systems ignore the important role of the textual content.…”

Section: Introductionmentioning

confidence: 99%

“…• systems processing specific document classes, able to adapt for processing similar document classes [28,21,15];…”

Section: Introductionmentioning

confidence: 99%

Document understanding for a broad class of documents

Aiello¹,

Monz²,

Todoran³

et al. 2002

International Journal on Document Analysis and Recognition

View full text Add to dashboard Cite

We present a document analysis system able to assign logical labels and extract reading order in a broad set of documents. All information sources, from geometric features and spatial relations to the textual features and content are employed in the analysis. To deal effectively with these information sources, we define a document representation general and flexible enough to represent complex documents. To handle such a broad document class, it uses generic document knowledge only. The generic document knowledge used is identified explicitly. Our system integrates components based on computer vision, artificial intelligence, and natural language processing techniques. Experimental results for each component and for the entire system are presented. The performance of the system is good, especially when considering the variety of documents in the collection. Document Understanding for a Broad Class of Documents

show abstract

Structured document labeling and rule extraction using a new recurrent fuzzy-neural system

Cited by 21 publications

References 6 publications

<title>A generic method for structure recognition of handwritten mail documents</title>

<title>A generic method for structure recognition of handwritten mail documents</title>

Textual Article Clustering in Newspaper Pages

Document understanding for a broad class of documents

Contact Info

Product

Resources

About