作者：
Mehri, MarouaSellami, AkremTabbone, SalvatoreUniv Sousse
LATIS Lab Adv Technol & Intelligent Syst Ecole Natl Ingenieurs Sousse Sousse 4023 Tunisia Univ Lorraine
IDMC Inst Sci Digital Management & Cognit Pole Herbert Simon 13 Rue Michel Ney F-54000 Nancy France Univ Lille
CNRS UMR 9189 CRIStAL Campus SciBatiment ESPRITAve Henri Poincare F-59655 Villeneuve Dascq France Univ Lorraine
LORIA CNRS UMR 7503 Campus Sci615 Rue Jardin Bot F-54506 Vandoeuvre Les Nancy France
due to the idiosyncrasies of historical document images (HdI), growing attention over the last decades is being paid for proposing robust HdI analysis solutions. Many research studies have shown that Gabor filters are...
详细信息
ISBN:
(纸本)9783031416842;9783031416859
due to the idiosyncrasies of historical document images (HdI), growing attention over the last decades is being paid for proposing robust HdI analysis solutions. Many research studies have shown that Gabor filters are among the low-level descriptors that best characterize texture information in HdI. On the other side, deep neural networks (dNN) have been successfully used for HdI segmentation. As a consequence, we propose in this paper a HdI segmentation method that is based on combining Gabor features anddNN. The segmentation method focuses on classifying each document image pixel to either graphic, text or background. The novelty of the proposed method lies mainly in feeding a dNN with a Gabor filtered image (obtained by applying specific multichannel Gabor filters) instead of an original image as input. The proposed method is decomposed into three steps: a) filtered image generation using Gabor filters, b) feature learning with stacked autoencoder, and c) image segmentation with 2du-net. In order to evaluate its performance, experiments are conductedusing two different datasets. The results are reported and compared with those of a recent state-of-the-art method.
暂无评论