A Deep Transfer Learning Approach to Document Image Quality Assessment

Onderzoeksoutput: Conference paperResearch

Samenvatting

Document image quality assessment (DIQA) is an important process for various applications such as optical character recognition (OCR) and document restoration. In this paper we propose a no-reference DIQA model based on a deep convolutional neural network (DCNN), where the rich knowledge of natural scene image characterization of a previously-trained DCNN is exploited towards OCR accuracy oriented document image quality assessment. Following a two-stage deep transfer learning procedure, we fine-tune the knowledge base of the DCNN in the first phase and bring in a task-specific segment consisting of three fully connected (FC) layers in the second phase. Based on the fine-tuned knowledge base, the task-specific segment is trained from scratch to facilitate the application of the transferred knowledge on the new task of document quality assessment. Testing results on a benchmark dataset demonstrate that the proposed model achieves state-of-the-art performance.
Originele taal-2English
Titel2019 International Conference on Document Analysis and Recognition (ICDAR)
UitgeverijIEEE
Pagina's1372-1377
Aantal pagina's6
ISBN van geprinte versie978-1-7281-3014-9, 978-1-7281-3015-6
DOI's
StatusPublished - 3 feb 2020
Evenement15th International Conference on Document Analysis and Recognition - International Convention Centre Sydney, Sydney, Australia
Duur: 20 sep 201925 sep 2019
https://icdar2019.org/

Publicatie series

NaamProceedings of the International Conference on Document Analysis and Recognition, ICDAR
ISSN van geprinte versie1520-5363

Conference

Conference15th International Conference on Document Analysis and Recognition
Verkorte titelICDAR2019
Land/RegioAustralia
StadSydney
Periode20/09/1925/09/19
Internet adres

Bibliografische nota

Publisher Copyright:
© 2019 IEEE.

Vingerafdruk

Duik in de onderzoeksthema's van 'A Deep Transfer Learning Approach to Document Image Quality Assessment'. Samen vormen ze een unieke vingerafdruk.

Citeer dit