Text/graphic separation using a sparse representation with multi-learned dictionaries - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2012

Text/graphic separation using a sparse representation with multi-learned dictionaries

Abstract

In this paper, we propose a new approach to extract text regions from graphical documents. In our method, we first empirically construct two sequences of learned dictionaries for the text and graphical parts respectively. Then, we compute the sparse representations of all different sizes and non-overlapped document patches in these learned dictionaries. Based on these representations, each patch can be classified into the text or graphic category by comparing its reconstruction errors. Same-sized patches in one category are then merged together to define the corresponding text or graphic layers which are combined to createfinal text/graphic layer. Finally, in a post-processing step, text regions are further filtered out by using some learned thresholds.
Fichier principal
Vignette du fichier
DO_extractText_SparseRepresentation.pdf (357.98 Ko) Télécharger le fichier
Origin : Files produced by the author(s)
Loading...

Dates and versions

hal-00759554 , version 1 (04-12-2012)

Identifiers

  • HAL Id : hal-00759554 , version 1

Cite

Thanh Ha Do, Salvatore Tabbone, Oriol Ramos Terrades. Text/graphic separation using a sparse representation with multi-learned dictionaries. 21st International Conference on Pattern Recognition - ICPR 2012, Nov 2012, Tsukuba, Japan. ⟨hal-00759554⟩
203 View
278 Download

Share

Gmail Facebook X LinkedIn More