A Deep Transfer Learning Approach to Document Image Quality Assessment

Tan Lu; Ann Dooms

doi:10.1109/ICDAR.2019.00221

A Deep Transfer Learning Approach to Document Image Quality Assessment

Research output: Chapter in Book/Report/Conference proceeding › Conference paper

Abstract

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.

Original language	English
Title of host publication	2019 International Conference on Document Analysis and Recognition (ICDAR)
Place of Publication	Sydney, Australia
Publisher	IEEE
Pages	1372-1377
Number of pages	6
Volume	September-2019
Edition	September-2019
ISBN (Electronic)	978-1-7281-3014-9
ISBN (Print)	978-1-7281-3015-6
DOIs	https://doi.org/10.1109/ICDAR.2019.00221
Publication status	Published - 3 Feb 2020
Event	15th International Conference on Document Analysis and Recognition - International Convention Centre Sydney, Sydney, Australia Duration: 20 Sept 2019 → 25 Sept 2019 https://icdar2019.org/

Publication series

Name	Proceedings of 2019 International Conference on Document Analysis and Recognition (ICDAR)
Publisher	IEEE
Number	September-2019
Volume	September-2019
ISSN (Print)	1520-5363
ISSN (Electronic)	2379-2140

Conference

Conference	15th International Conference on Document Analysis and Recognition
Abbreviated title	ICDAR2019
Country/Territory	Australia
City	Sydney
Period	20/09/19 → 25/09/19
Internet address	https://icdar2019.org/

Keywords

document image quality assessment
OCR accuracy
deep convolutional neural network
deep transfer learning

Access to Document

10.1109/ICDAR.2019.00221

Handle.net

Persistent link

Cite this

Lu, T., & Dooms, A. (2020). A Deep Transfer Learning Approach to Document Image Quality Assessment. In 2019 International Conference on Document Analysis and Recognition (ICDAR) (September-2019 ed., Vol. September-2019, pp. 1372-1377). Article 19318631 (Proceedings of 2019 International Conference on Document Analysis and Recognition (ICDAR); Vol. September-2019, No. September-2019). IEEE. https://doi.org/10.1109/ICDAR.2019.00221

@inproceedings{cfa76a9172b644429f6c2d3d61eb2ba9,

title = "A Deep Transfer Learning Approach to Document Image Quality Assessment",

abstract = "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.",

keywords = "document image quality assessment, OCR accuracy, deep convolutional neural network, deep transfer learning",

author = "Tan Lu and Ann Dooms",

year = "2020",

month = feb,

day = "3",

doi = "10.1109/ICDAR.2019.00221",

language = "English",

isbn = "978-1-7281-3015-6",

volume = "September-2019",

series = "Proceedings of 2019 International Conference on Document Analysis and Recognition (ICDAR)",

publisher = "IEEE",

number = "September-2019",

pages = "1372--1377",

booktitle = "2019 International Conference on Document Analysis and Recognition (ICDAR)",

edition = "September-2019",

note = "15th International Conference on Document Analysis and Recognition ; Conference date: 20-09-2019 Through 25-09-2019",

url = "https://icdar2019.org/",

}

Lu, T & Dooms, A 2020, A Deep Transfer Learning Approach to Document Image Quality Assessment. in 2019 International Conference on Document Analysis and Recognition (ICDAR). September-2019 edn, vol. September-2019, 19318631, Proceedings of 2019 International Conference on Document Analysis and Recognition (ICDAR), no. September-2019, vol. September-2019, IEEE, Sydney, Australia, pp. 1372-1377, 15th International Conference on Document Analysis and Recognition, Sydney, Australia, 20/09/19. https://doi.org/10.1109/ICDAR.2019.00221

A Deep Transfer Learning Approach to Document Image Quality Assessment. / Lu, Tan ; Dooms, Ann.
2019 International Conference on Document Analysis and Recognition (ICDAR). Vol. September-2019 September-2019. ed. Sydney, Australia: IEEE, 2020. p. 1372-1377 19318631 (Proceedings of 2019 International Conference on Document Analysis and Recognition (ICDAR); Vol. September-2019, No. September-2019).

Research output: Chapter in Book/Report/Conference proceeding › Conference paper

TY - GEN

T1 - A Deep Transfer Learning Approach to Document Image Quality Assessment

AU - Lu, Tan

AU - Dooms, Ann

PY - 2020/2/3

Y1 - 2020/2/3

N2 - 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.

AB - 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.

KW - document image quality assessment

KW - OCR accuracy

KW - deep convolutional neural network

KW - deep transfer learning

UR - https://ieeexplore.ieee.org/document/8978068

U2 - 10.1109/ICDAR.2019.00221

DO - 10.1109/ICDAR.2019.00221

M3 - Conference paper

SN - 978-1-7281-3015-6

VL - September-2019

T3 - Proceedings of 2019 International Conference on Document Analysis and Recognition (ICDAR)

SP - 1372

EP - 1377

BT - 2019 International Conference on Document Analysis and Recognition (ICDAR)

PB - IEEE

CY - Sydney, Australia

T2 - 15th International Conference on Document Analysis and Recognition

Y2 - 20 September 2019 through 25 September 2019

ER -

Lu T , Dooms A. A Deep Transfer Learning Approach to Document Image Quality Assessment. In 2019 International Conference on Document Analysis and Recognition (ICDAR). September-2019 ed. Vol. September-2019. Sydney, Australia: IEEE. 2020. p. 1372-1377. 19318631. (Proceedings of 2019 International Conference on Document Analysis and Recognition (ICDAR); September-2019). doi: 10.1109/ICDAR.2019.00221

A Deep Transfer Learning Approach to Document Image Quality Assessment

Abstract

Publication series

Conference

Keywords

Access to Document

Handle.net

Other files and links

Fingerprint

Cite this